Pdfjoiner

Industrial Instrumentation
Prof. Alok Barua

Department of Electrical Engineering
Indian Institute of Technology - Kharagpur
Lecture - 1
Introduction
Good morning! This is a basic course on industrial instrumentation. As you know, the
instrumentation is very important subject in all process industry. As you know, in the
industry like steel, petrochemicals, then fertilizers, all this type of, I mean, industry, the
instrumentations is a, I mean subject which is a concern for everybody. The reason is
there are various sensors in the instrumentation, excuse me and you will find that some of
the sensors are ……, especially flow, temperature and all those things you have to
measure, then you have to convert it in some convenient domain which can be, I mean
signal processed later on and transmit those signal to the control room.
Now, ultimately you know that all these measurements and instrumentation is basically to
make your product or quality of the product better. Suppose in the fertilizer plants the
quality of the product should be good, urea that means the purity of the urea or in the case
of steel industry, percentage of the carbon in the steel all those things are important and in
every, I mean step you will find that we have to measure some parameters. What are
those parameters? The parameters are pressure, temperature, flow, humidity, viscosity,
pH, all those things. The reason that we have given the name of the course as industrial
instrumentation instead of instrumentation is that you will see we will always focus on to
some process industries. That means what type of measurements they are doing and how
that measurement is converted to electrical domain, so that I can get a current of 4 to 20
milliampere and all those things.
Now, today is the first lecture, so obviously I will introduce the different components and
I will cover to some extent the static characteristics of the, I will cover full the static
characteristics and the dynamic characteristics I will start in the next lesson. The lesson 1,
let us look at the contents.
(Refer Slide Time: 3:14)
The lesson 1, it is basically as I told, Introduction. The contents of this lesson is getting
started that means basic introduction to the system. Then, I will cover the static
characteristics. Now, the static characteristics, it means that the accuracy, linearity,
sensitivity, etc., are the parameters of an instrument or sensor and these attributes are
collectively known as the static characteristics of an instrument and are given in the data
sheet for a particular instrument.

Now, at the end of this lesson, the viewer will know the static characteristics of the
industrial instrumentation, the importance of instrumentation in process industry.
Now, classification of instruments, so they, I mean instrument can be classified in 4

different categories. I mean, it can have active and passive instruments. Then, in the case
of active instruments, you know that you have displacement sensors or potentiometer. So,
in the case of displacement sensor or I mean you can have in the case of active
instruments that means displacement sensor, suppose I have a displacement sensor, it
looks like this that I have, sorry, displacement, suppose I have a potentiometer like this
and I am giving, energize it by battery and I am taking the output voltage from this one.
So, this is the wiper or the jockey of your potentiometer displacement sensors and I am
taking the output from these two regions. I am sorry, it should look like this.
It should be, I mean, I have a displacement sensor, so I am giving a, this is our jockey and
I am taking ….. So, suppose this is the voltage e naught, if it supply the ex, excitation
voltage and this voltage e naught, obviously you can see that e naught if you displace it in
this direction, suppose x, so, that e naught will be a function of x. If the x changes our e
naught also will change. So, in this case, you can find that this is a displacement sensor
and I need an excitation voltage here. If you do not give the excitation voltage, my
potentiometer obviously will not work and I will not get any output voltage there. So, this
type of instrument I should call it or we should call it an active instrument or active
sensor.
Now, there are some sensors which are passive. Passive in the sense, suppose I have a
Bourdon gauge which is used and we will discuss that later on, that Bourdon gauge you
know there is a C type of tube, it looks like this.
I have a C tube and the area of cross section of the tube is elliptical. Now, if you increase
the pressure here, so the tube will try to make in circular form. That means it will have a
circular form like this one. So, accordingly this tip of our tube will move in this direction.
So, this movement of the tip, this movement of the tip can be utilized to move some
pointer on the scale, so that we get some arrangements of a rack and pinion and all those
things, you can find that it will move on a scale like this one. So, in this case, that it will
work as an, it is a passive instrument, it is not an active instrument.
Moreover, please note that active instrument does not mean it is only electrical. Active
instruments can also be pneumatic, because I have told about the active instrument which
is only passive, but active instrument is also, it can be pneumatic instruments. It can have
also hydraulic instrument, where there is supply, always supply. Now, force balance
systems or you can flapper nozzle systems most of them are familiar, we thought we will
cover this. I am sorry, you are not familiar with that right now. You will be familiar with
the systems after sometime or after few lectures that you will find that this, the flapper
nozzle system you need to always have a power supply. I mean it is a pneumatic supply,
so without supply, the flapper nozzle systems cannot work. Similarly hydraulic systems
also; I need some hydraulic, some liquids like oil or water or kerosene that type of things,
so that it will be activated. So, that type of instruments are active instruments and the
choice between the active and passive instruments involves carefully balancing the
measurement, the resolution. These are the requirements against cost.
The reason you see that in the case of, I mean just let us go back to again and through this
potentiometer you see that when we discussed the potentiometer we have seen that the,
we have the potentiometer here, we have a power supply here and we are taking, this is
our wiper which moves in this direction, x and I am getting an output voltage e naught.
Now, you see that resolution of this potentiometer that is the output voltage can be made
better and better, if I, suppose if I increase this supply voltage value, whereas and this is
very simple. I mean you just increase the supply voltage ex, excitation voltage and the
measurement resolution will increase, right; whereas in the case of the passive
instrument, it is not very easy. Suppose as I told you, as I show you both the tube which
is used for measurement of, I mean measurement of pressure or suppose one of the, one
of the good example is mercury in glass manometers all of you are familiar with. How
does it look?
It looks like all of you have seen that we have a tube like this one and there is a mercury
column bulb here. So, mercury will heat up, it will go up like this one. So, in this case,
the resolution if I want to change, the resolution it is not very easy ….. I mean you have
to change the entire constructions of the system. That means the capillary should be more
thinner and all those things, so that you will get for a smaller and smaller measurements
of temperature. But, that is not necessary in the case of, in the case of electrical
instruments or suppose I have, I have some excitations or pneumatic instruments or
hydraulic instrument; that we already discussed.
Now, there are null and deflection type of instrument. This is very important, null and
deflection type of instruments. You see that if I say here how does it, what is this null
deflection type? All of you are familiar with the Wheatstone bridge. How does it look? It
looks like this. This is our Wheatstone bridge. I have a power supply here and I am
measuring the output voltage here, e naught. Suppose this is R 1, R 2, R 3, R 4; simple, I
can write that R 1 is equal to R 4 by R 3 into R 2. Isn’t it? At balance I can write like this
one, when there is no, I mean no current flowing through the detector, this, suppose this
is the detector or voltmeter.
Now, null type means, you see that I can utilize, suppose this R 1 varies; suppose R 1 is a
thermistor which is measuring some voltage, some temperatures and this obviously what
will happen if the temperature rises? The value of the R 1 will fall. Now, I can utilize
either R 4 or R 2 to balance the bridge. See, if I use some resistance R 4 and R 2, suppose
R 2 is also varying, so in that case, it is a null type of instrument. Whereas, if I utilize
this, suppose whenever there is a, this R 1 initially suppose all the bridge, all are equal, so
bridge is balanced.
Now, R 1 since it is a thermistor, as the temperature rises, this resistance will fall. I will
get an unbalanced voltage at the detector side. So, this voltage actually will, you can say
it is, now the instrument, now our system is a deflection type of system; it is no more a
null. I mean I can calibrate this output voltage. That means I calibrate this unbalanced
voltage in terms of the temperature or in terms of the resistance R 1, so that is possible.
So, this is the difference between null and deflection type of instrument. Sometimes we
prefer to have a null type of instrument, sometimes we, I mean prefer to have a deflection
type of instrument. Especially in the case of electrical circuits, when we need always
some output voltage we should go for the deflection type of instrument. But, in some
precisions we need a null, because always whenever if you make the bridge null,
obviously it is the most accurate form of measurement.
Now, monitoring and transmitting instrument, what does it mean? It means that that
mercury in glass manometer, that is a good example of the monitoring instrument.
Mercury in glass manometer, as you know, you have seen or suppose a flow meter,
suppose a rotometer, it is the common example of the, I mean how does it look is what I
will discuss in details later on. So, it has a shape like this one, a vessel and it is, there is a
float or bob. So, it will go like this one. So, liquid is flowing through this and going out.
So, this liquid will move up and down, right. The liquid will move up and down and these
positions of the bob, position of the bob can be calibrated in terms of the flow, right. So,
this type of instrument is only monitoring instrument. You cannot say it is a transmitting
instrument, right or mercury in glass manometer that is also a transmitting instrument,
because you are reading only the temperature. But suppose, if I use a thermocouple there
instead of, I mean instead of mercury in glass manometer if I use a thermocouple, in the
case of thermocouple what will happen? You know that you can, you will get a voltage
there. Whenever there is a change, difference of temperature between the two junctions of
the thermocouple of dissimilar metals, you will get a voltage there. So, that voltage can
be calibrated in terms of temperature.
Similarly, in the case of RTD or resistance temperature detectors, which is very accurate
form of measurements of temperature you will find also there that if the temperature
varies the resistance will vary. So, that change of resistance can be taken, as I have shown
in last slide that can be utilized as an unbalanced voltage. So, that unbalanced voltage can
be calibrated in terms of the temperature. So, this is also because that signal can be
transmitted. Now, one thing you should know that in the industry that usually or you
cannot transmit voltage, you have to transmit current.
Even though it is not very relevant whether it is monitoring or transmitting instruments,

but I am discussing its importance. Since it is industrial instrumentation, you must know
what is the practical aspects of the sensor signal conditioning circuitry. Any signal which
is electrical in nature, you are not supposed to, I mean, transmit voltage. Voltage, if you
transmit it will be totally corrupted by the noise, you have to transmit current. So, typical
standard for industry is 4 to 20 milliampere of current that is for the total range of
temperature, pressure, flow, etc. So, this is the difference between monitoring and
transmitting sort of instruments.
So, all the instruments, there are some instruments which are both monitoring and
transmitting. That means it has monitoring system also. It is monitoring the temperatures.
Onboard you will find the, on the panel itself you will find the temperature itself. Also, it
has a transmitting capability. If it is has no transmitting capability there is a transmitter
which is to be utilized to convert that signal, I mean to the current domain. Transmitter is
a very common form. In industry you will find that many a times we are talking about
this transmitter. The function of this transmitter is to convert the signal whether it is
pneumatic signal, electrical signal, hydraulic signal, to convert to 4 to 20 milliampere of
electrical current. This will be converted, I mean transmitted to the control room for
packing the control actions which will ultimately activate all the control valve like this
one, to control flow or heater, heating to the some heating to the some heater or heating
power to the heater and all those things or we can have an analog or digital instrument.
Now, in the case of analog instrument, output varies continuously. It is a, there can be
infinite …. positions of the pointer. Suppose I have a simple, if I say that if I have simple
voltmeter, suppose I have a voltmeter, if I take a, well I have voltmeter, so voltmeter,
what will happen? The pointer will move, isn’t it? I have a calibrate scale here in volts, so
pointer will move. So, in case of analog instrument, when you talk of analog instrument
you will find there is an infinite, infinite position of the pointer. You can have many
positions, whereas in the case of digital instruments this is not the case. In the digital
instrument, output varies in the discrete forms; it is not continuous. So, in that case, we
will consider the digital instrument.
Sometimes when people talk about, I mean talk about suppose a revolution counter
whether that is digital, I will not consider that as a digital. Digital instrument is, so far is,
I mean today is concerned we talk of the digital instruments where you have some digital
circuitry and the output is also digital. Suppose if you have a virtual instrumentation. This
is actually I mean digital instrumentation, but you are showing that like a conventional
analog meter. It is not necessary it is an analog instrument. So, in that sense I say, if the
signal conditioning is there itself inside the system, instrumentation system digital in
nature, we call it digital instrument, whereas in the case of analog instruments, it is
analog in nature. But please note that in the analog instruments there can be analog meter,
there can be many positions of the pointer on the scale, whereas in the case of digital
instrument there are finite number of positions of the output. This is all about the digital
and analog instrument.
Now, let us look at a process diagram which will show you glimpse, I mean give you
some idea of the instrumentation system and why we have given the name of the
industrial instrumentation, instead of simple instrumentation, instrumentation devices and
all those things. Now, this is a typical process, you look at. This is a typical process.
Here, you see that a liquid is flowing through this pipe and it is going out, right and there
is an orifice plate. Now, what will happen, you see. There are two tappings here. One is
upstream tapping and another is downstream tapping. So, this tapping means that it will,
the pressure, the flow, whatever the liquid is flowing through the pipe, our ultimate goal I
have a set point that I have a set point, some fixed value of suppose L litre per minute that
liquid I have to, I should always or that liquid should always, this rate of the liquid should
be maintained through this pipe. That is our ultimate goal.
In many a situation you will find this type of, whether steel, chemical plants, you will
find this type of situations. That might be water, might be naphtha and so many things are
there, where you have to measure the flow. Now, you see, what will happen is that liquid
is flowing. So, there is an upstream tapping and downstream tapping. What is the
function? We will see later on. You see that in the case of flow meter, the flow is
converted in terms of pressure, because as you know that when the liquid is flowing
through this pipe here the flow is, whenever it is flowing through this region the flow is
high.
If the flow is high, then what will happen? The pressure will drop. So, there is a high
pressure zone and it is a low pressure zone. So, this I will take through a tapping. This is
called upstream tapping and downstream tapping and I am giving to a DP transmitter. DP
transmitter means it is a differential pressure transmitter. Sometimes back people used to
call it a DPDT, differential pressure differential transmitter. But nowadays, more
reasonably people call it just DP transmitter or differential pressure transmitter. Why it is
differential pressure, because you see, I have to measure, I have to look at, I have to take
the tapping both upstream and downstream. So, because this is a restriction, so this we
call upstream, this downstream and this differential pressure is going to DP transmitter.
Now, this DP transmitter is, I mean according to flow it is giving a signal, pneumatic
signal of 3 to 15 PSI. PSI means pounds per square inch. Even though this is, nowadays
as you know it is an SI system, but still in some cases people use, more convenient with
the PSI for some reason. Even though while we solve the problem we will find that we
are using SI system, but steel industry they use this type of units. It is not a problem.
Now, this 3 to 15 PSI of pressure that means varies. That means for almost no flow I will
get 3 PSI of pressure. For full, for full flow of the liquid, I will get 15 PSI of pressure.
So, this flow, I mean this will go to the pressure to current convertor. Now, pressure to
current convertor, what it will do? It will convert that signal, pneumatic signal to the
current domain. As I told you earlier, you see that you have to transmit the pneumatic
signal, electrical signal, hydraulic signal to the electrical domain, so 4 to 20 milliampere
of current. Now, what is this 4 to 20 milliampere? As you say, it is from for 3 PSI I have
a 4 milliampere of current, for 15 PSI I have a 20 milliampere of current. This range you
have to maintain. Whatever the range of the flow, it does not matter; accordingly it will
…., because this will remain constant. It might be in a huge flow, might be it can be 10 to
the power of 5 litre per hour in that type.
Suppose in a power station or water pumping station and all those things, a huge amount
of liquid is flowing or in some narrow pipe that might be, it is few, I mean it does not
matter. The range will be always, which is converted will be always 3 to 15 PSI for the
pneumatic signal and for the, for the current signal it is 4 to 20 milliampere.
But nowadays, as you know, these, the entire things are incorporated in one DP
transmitter, where you will get directly this 4 to 20 milliampere of current. This you do
not, you do not need P to I convertor and all those things; everything is incorporated
inside. We will see that you will utilize this differential pressure to make a, we will utilize
it for a change of positions of a diaphragm plate in the case of capacitive measurement, so
that I will get differential of capacitance, we will discuss later on and accordingly I will
get a current.
Now, this current is coming to a controller. The controller, the function of the controller
is to give a control signal to the pneumatic control valve. I have a set point, because set
point, what is set point? Set point will set the value of the flow of the liquid, L litre.
Suppose the 1000 litre per minute or that type of very high, 1000 litre per minute or I
should say 1000 litre per hour that type of set points we have done. Now, if for some
reason or the other this falls, then what will happen? I have to open the valve, so that
more liquid will flow or for some reason or the other suppose this, this is, this has
increased that means flow has increased, then what you will get? I will get a, I will get a
flow. I have to close the valve, right. So, that type of signal you have to, you have to put
to the control valve. So, that control, where I will get that signal? I will get the signal, I
will subtract this signal, I have a set point and I have a signal. So, these two signals will
be subtracted. When you will, in future we will study the control system; we will find all
those things in details. That means I have a system here and I have a, this is our set point
and this is our error. So, this will be kept and this will come to the controller, okay. This,
I have a process, it will take actions or utilize this one, anyway.
So, I have a, that means I have a controller process. I have a measurement, it is coming
like … So, this is our set point, this is our output from the DP transmitter, right. So, this is
coming …... So, our goal is to make this R equal to C. That means these two should
remain the same. If there is a difference, I will get an error voltage. So, our controller will
take some actions. So, it will send some signal to the control valve and all those things. It
will take some action, it will make, so that it will move in such a direction, so that ES will
be again zero, right.
Here also it is same if you see that. What will happen if it increases? If the flow increases,
my valve is to be closed, so the flow, if the flow increase this current will also increase.
So, there is a, there is a set point. There is an error, so this is an error signal which is
coming from there. This error signal also should be 4 to 20 milliampere, because I cannot
transmit anything other than 4 to 20 milliampere. So, this will come to the, because this
valve is pneumatic. Most of the industrial valve, as you know it is a pneumatic valve,
because in many industry like the, especially hydrocarbon industries and all those things,
petrochemical industries and all those things, you cannot use our solenoid valve,
electrically operated valve, because of the safety reasons, because the operating voltage
should be below 50 volt. So, there is a restriction.
So, for a large pipe and all those things, I cannot use there all those small control valves.
So, I have to use pneumatic control. But, pneumatic control valve means that signal
should be pneumatic to close or open. So, to close or open, I have to give pneumatic
signal. So, ultimately what is that? That is very simply, precisely I should say that if I, the
flow is increased, so I will get a signal through this transmitter. This will convert to the
current domain moving to the controller. I have a set point, so obviously flow has
increased. There is a, there is an error voltage, it will come, so the stem of the valve will
go down, the flow of the valve will be controlled. This action will take until and unless
this error signal will become zero, right or this volt, this current and this set point will
remain the same. It is in other way also.
If the flow suppose decreases, then what will happen? The flow decreases, I will again
also get some error voltage, but of opposite sign, obviously. So, what will happen? So, I
have to take some action, so that the control valve should stem, should go out, so that I
will get a flow which is actual to the set points. It will, the action will be taken until
unless this error current will become zero, right.
Now, with this, I will go to the instrumentation laboratory of IIT Kharagpur, where you
can see some of the instruments or process instruments. You will get the glimpses of how
it looks like, how the different flow meter works and what actually the process looks like.
We have some small, I mean process, I mean by which you can get the idea of what is the
instrumentation system we talk about. We will come back after some time.
This is a flow and level control system, which is actually the scaled down version of the
actual process and what actually it is, the system is doing that it has different sensors.
What I want to mean that you will see the instrumentation is everywhere. You will find
that there are several sensors. There are float level sensors, there are flow sensors, there
are several control valves. So, the goal of this process is that basically to control the level
as well as flow in a particular process. Now, the several sensors will take, I mean will
play actually very key role in a system, because, you see everywhere you will find that
the sensors output is most important for actual control of the, whatever the control
systems you are using that is it is important as well as whatever the sensor output which
is coming from the sensors is also important. So, the sensor’s calibrations, its accuracy,
its resolutions also will control the basic process.
Any process has some output. It has some, suppose if you go to the petrochemical
process, it has some desired output or petrol output or kerosene output for any other type
of material or if you go to fertilizer plant, you will find that there are different types of
fertilizers are coming out, but the quality of the fertilizer or quality of diesel, quality of
petrol will typically be decided by, not only by the control elements which you are using
in the plant as well by the sensor. If the sensors are not accurate, the instrumentation
system is not accurate, then you will find that the entire quality control is out of order. So,
this tells that how instrumentation is important in industry.

Now, I will describe at least two sensors. One of this is, usually this is a rotameter which
is a flow meter. It will give the estimate of the flow of the liquid in the entire process.
Now, you see, a rotameter is a tapered tube with a very small angle and the liquid is
coming in through this pipe and liquid is coming out through that and then, the float,
some technical persons they call it bob also. As the liquid flow increases, this bob will
move up. Now, the details of these different forces which is acting on the float, we will
discuss later on. Now, you can just remember here what will happen that when the float
goes up that means the liquid flow also increases here.
Here, you can see here there is one float sensor, it is a pollution free float sensors. Due to
the different level of the, you can see that this liquid can move up.
You can see that this float will go up like this one as the liquid goes up and potentiometer
will move and it will give the electrical output. Now, any process or any instrumentation
system, one thing we must remember that, which we will discuss later on also, that you
cannot transmit voltage, you have to transmit current.
Now, for that you see here there is a, you have a float level transmitter which will
transmit 4 to 20 milliampere of current to the process. This system, this, first it will go
through this chord, it will go to the process interface.
From the process interface, it will come to the digital computer. Now, then the digital
computer, there are control algorithms by which you can set the different algorithms. You
can tune the controller, it will feedback the system to the process controller, process
controller will take the action. Then, the signal will go back through this chord again to
this control valve; control valve will make the decision, the flow. Accordingly, you can
achieve the particular flow or particular level in this process. This is the miniature
versions of the actual process actual process might be large, but whatever the components
you can see here, this you will find also in the process.
In this assembly, you can find several flow meters. There are rotameters, there are optical
flow meters, there are hot wire anemometers and there is also a magnetic flow meter. On
top of it, you will find there is a weir, it is called V notch weir. As you know, the flow
meters are different types. These are, some of them are for the open channel and some of
them are for the closed channel. This V notch weir is basically an open channel meter,
which will give you the, extensively used for the irrigation purposes, so that how much
water is needed in the different fields, to get the estimate of that.
Now, you can see that if you increase the flow here by this, you can have a pump by
which you can increase the flow, you see the flow is, I have reduced the flow, so the bob
of the flow meter is coming down and this tapered glass is calibrated in terms of
centimeter cube per minute. So, it will give you the flow and this height is also calibrated
in terms of flow and you can see here, interestingly this is one electro optics flow meter.
If you go, if I go to the backside, you can see there are four spots and there are lights
which are falling on the flow meter and for each complete rotations, I am getting four
pulses. So, accordingly, suppose you have 10 rotations, so accordingly I can calculate
how much the flow is moving. So, it is a, it has a direct electrical output, wherein you can
see here this flow meter has no electrical output. Similarly, hot wire anemometer also has
electrical output and this is the hot wire anemometer. You cannot see from there, because
it is inside and in the magnetic flow meter also it is electrical output. So, all these flow
meters except this rotameter have electrical output.
Now, electrical output has the advantage. If you have in the instrumentation systems
electrical output, you can directly convert that signal to the current domain, 4 to 20
milliampere of current domain. Then, you can digitize this if it is necessary, whereas if
you have some non-electrical output as a pneumatic output or some mechanical
movements that type of things, you have to convert this to the electrical domain by using
either pneumo electric convertors sort of things or any other form of devices. So, this will
give you the glimpses of the different flow sensors which is used in the industry, in the
process. In the flow meters, you will find in any process industry these are thousands in
numbers. Unfortunately, here I cannot show you that type of thing, but this will give you
the idea of the how the flow is measured in the process.
If you see here the backside of this flow meter system, even though you cannot see here
clearly the hot wire anemometer which is also a resistance based flow meter and
electromagnetic flow meter, but you can see here the electro optic flow meter.
You see, what will happen in this instrumentation system that there is a light source
which will be launched and there are four spots in the backside or four white spots in the
backside, four white or silver. So, lighting gets reflected from that and there is a receiver
on the other side. So, what will happen that when it rotates, for each complete rotation of
the system we will get four clock pulses. So, if I can count the number of clock pulses,
which obviously you have to, I mean you have to read the single process, we have to
make little single processing in the sense that you have to pulse the signal shaper like
switch here, we will get output and that can be counted for 60 minutes, so that you can
get the flow because if you know the flow by diameters and all those things you can
immediately tell that how much is the flow rate in a minute. So, this will give you the
direct electrical output, obviously that can be processed, that can be easily interfaced to
the computer. So, this will give the glimpses of the different flow sensors in the system
and that can be calibrated obviously in this rotameter, which is in the front side, which I
have shown at the beginning of this lecture.
Welcome back to the class room. So, we will now study the instrument characteristics.
The static characteristics of an instrument are concerned only with the steady state
readings. It is not concerned with any transient readings. So, we are concerned only with
the steady state readings. So, we will call one by one now, we will read one by one the
different steady characteristics. First of all the span; now, if in a measuring instrument the
highest point of calibration is X 2 units and the lowest point of calibration is X 1 units, I
can tell that then the instrumentation range is X 2 units and the instrument span is X 2
minus X 1 units. It is very simple. That means suppose I have a, if I have a, if I have a
total, I mean range, suppose I have an instrument temperature sense which can measure 5
degree centigrade to 200 degree centigrade in that case, range I will call it 200 degree
centigrade, span I will call it 195 degree centigrade.
Now, mean and standard deviation of measurement, see we will need this in future. You
will find that is the reason we are explaining. Suppose we are making a set of n
measurement, X 1, X 2, X n and the mean value will be given by X mean; X 1 plus X 2
plus 1 X n upon n, this is X mean. This is mean of the measurement and the spread of any
measured value X i can be expressed as a deviation which can be expressed as, given as d
i equal to X i minus X mean.

The extent to which the measured values are spread about the mean value are known as,
known and now we express as standard deviation sigma, where sigma equal to d 1 square
plus d 2 square so on d n square upon n minus 1 whole to the power half, under the
square root. So, this is the standard deviations. We need it in future. So, we will find that
is the reason we are explaining it here.
Now, accuracy actually means, accuracy and precisions, accuracy is usually expressed as
accurate to within x percent; even though we do not use exactly, use the same, I mean
word, so accuracy and accurate, it is slightly different. It is usually, we do not, I mean
like that if somebody express the same word, I mean some definition with the, with the
implanted in the same word in the definition itself, but these two are different. So,
accuracy, we will explain this more. Accuracy is usually expressed as accurate to within
x percent. It means accurate to within plus minus x percent of the instrument span at all
calibration points of the scale unless otherwise stated.
When a temperature transducer with an error of plus minus 1% indicates a 100 degree
centigrade, the true temperature is somewhere between a 99 degree centigrade and 101
degree centigrade. Thus, the measurement accuracy of plus minus 1% defines how close
the measurement is to the actual measured quantity. So, this is the main test. Thus the
measurement accuracy of plus minus 1% defines how close the measurement is to the
actual measured quantities. How much is the deviation? That is actually we talk about the
accuracy.
Linearity, if the calibration is very important, because in instrumentation you know

always we prefer a linear sensor. There is a lot of problem if you have a, do not have a
linear sensor; we will find that. What is mean? Means the input output relation should be
linear. Forget about the accuracy, we can have a, I can have a characteristics like this one
also. I can say it is a, it is non-linear, but it can be very accurate. But there is a, if the
linear ease of calibration will be there that means that if i can calibrate an instrument, if it
is linear instrument its calibration is quite simple. Its sensitivity will be same at all
positions.
Why? You see here, I have this type of relation. So, it does not matter where I take. The
sensitivity, sorry the ……. will remain the same; whereas if you, if you take some other,
suppose here in this curve, sensitivity something here and somewhere here, it will be
different. So, while you are making the signal conditioning circuitry, it will be very
difficult. The task will be very, very hard. That means, I mean your amplifier should also
be adaptive. It is a very difficult task. So, always you prefer linear instrument. But, some
instruments, some sensor like thermocouple, thermistor, it is non-linear sensor. We
cannot do anything, because of its low cost, it is ……, there are so many other positive
points we have sacrificed that, sacrificed that it is a non-linear sensor. I mean we want to
accommodate it even though it is a non-linear sensor.
Now, it is normally desirable that the output reading of an instrument is linearly

proportional to the quantity being measured. Figure 1.2 shows a calibration curve that is a
plot of the typical output reading, cross marked, of an instrument when a sequence of
input quantities is applied to it; we will show in the next slide. A good fit straight line will
give the calibration curve of the instrument. How does it look?
Non-linearity is thus defined as the maximum deviations of any of the output readings
marked with a cross from the straight line and it is usually expressed as a percentage of
the full scale reading. You see, that is we talked about the Figure 1.2 measured quantity.
You see here, that this is our measured quantity. So, you make a best fit curve. There are
various softwares available, so by which, least square method is there by which you can
make a best fit curve, so that it will be a best fit for all the points which you have given.
So, this is our deviations, okay. It is not that this is our deviations, this is our deviations,
okay, this is our deviations, this is our deviations. So, this non-linearity is thus defined as
the maximum deviations of any of the output readings marked with a cross from this
straight line and it is usually expressed as a percentage of full scale reading. Always it
should be expressed in percentage of full scale reading.
Now, tolerance, it is a term that is closely related to the accuracy and defines the
maximum error which is to be expected on some value. It is a not exactly a static
characteristics; obviously, you know that. However, some instrument manufacturers
usually quote that, I mean quote this tolerance figure. So, you must know what is the
tolerance? One of the good examples, those who are not familiar with the instrument
tolerance, is the maximum deviation of a manufactured component from some specified
value; you are familiar with that.
Why? You see for an example, a resistor having a nominal value of 100 ohm and a
tolerance of 5% might be having an actual value anywhere between 95 ohm and 105
ohm. There are resistors available of varying, various tolerance; 1%, 10%, 2% depending
on … So, if it is, tolerance band is important, because not always you will find the 0.1%
tolerance. If you, with the 5% tolerance, I mean measurement is okay, there is no
problem.
Why it is saying I am saying, you see that if my measurement error is 5% that means
tolerance band is coming 5%. Ultimately what we will do with this, such a precise
measurements? Ultimately the quality of the product, if the quality of the product does
not deteriorate, I can amount, I mean I can allow certain amount of tolerances, is not it? If
it is not, if the product, ultimately product falls, supposing some, I mean chemical plants
you can, you have to be precise. If there is 5 degree centigrade temperature variations,
suppose there is a, you are controlling some temperatures of suppose 600 degree
centigrade. It hardly matters if the temperature varies to 598 to 602 degree centigrade,
whereas in some process suppose in the case of bio process we will find that where the
cell grows, in that type of situations with even the difference of temperature of 0.5 degree
might be detrimental to the cell, cell will die. So, in that type of situations you cannot
allow the temperature to, so this tolerance are there everywhere.
Now, static error it is the difference between the true value of a time invariant measurable
quantity and the value indicated by the instrument. The static error is expressed as plus y
units or minus y units and for static error in units, true value plus static error is the
instrument reading or true value equal to instrument reading plus static correction, quite
obviously because it is opposite of that it is the static error.
Now, repeatability is very important. Repeatability is, in instrumentation you see that the,
I mean if you, there are thousands of instruments in a big plant, thousands of sensors, I
mean sub temperature pressure flow, you cannot expect that you will calibrate the
instruments every day or every once in 7 days. Usually the typical norm is that there is a
routine, maintenance routine, I mean checking of all these calibrations. But if there is a, if
the fall of quality of some of the, then you will, for some of the product, we will talk
about the instrument calibrations or sensor calibrations, otherwise not. So repeatability,
repeatability is defined as the, of an instrument is the degree of closeness with which a
measurable quantity may be repeatedly measured.
What does it mean? Suppose I have a voltmeter. So, I am giving a battery voltage. So,
battery voltage, suppose a dry cell battery voltage is, voltage is fixed 1.5 volts. If I even
measure 1000 times it should always give 1.5 volt by that instrument or voltmeter. I
would say that its repeatability is good, because battery voltage is not changing, supposed
to. Suppose if it is a dry cell battery, if you are not drawing any current from that, because
usually voltmeter doesn’t draw much current, so it is, if you take 1000 reading, it should
give you the exactly same value 1.5 volt. If it does not, that means that repeatability is
poor.
Now, mathematically how will you define this repeatability? Mathematically, it is defined
as the measure of the variation in the measured data known as the standard deviation
sigma. It is expressed in terms of maximum repeatability error as a percentage of full
scale output range, right. So, it is repeatability 2 sigma full scale output range into 100.
This is the percentage repeatability.
Now, static sensitivity is very important. The slope of a static calibration curve evaluated
at the input value is the static sensitivity, right. It means that as shown in Figure 1.3, we
will show, it will come, the static sensitivity at any particular input value say X 1 is
expressed as this. This is Figure 1. You can see here, so this is a calibration curve, this is
my input, sorry, yes, you see here, this is our, this is our calibration curve; this is our
calibration curve, right. It is going like this. It is non-linear, it does not matter. So, you
see here that this is my input and this is the output reading. The input, say some
parameter and output we are getting some reading, excuse me.
The static sensitivity is defined as at X 1 value suppose if I say this is the X 1, at the
value X 1 the slope of this, I mean calibration curve, so it is dy by dx at X equal X 1.
Interestingly you see this static sensitivity varies. You see, here is some, there is some
value, here it is some different value. So, the static sensitivity for the non-linear curve
will change. It depends under what point you are measuring, right. You will find this
typically true that in the case of thermocouple. Thermocouple you will find that, typically
we say something suppose the platinum, platinum rhodium thermocouple, we say that the
static, I mean its sensitivity is around 10 to 12 microvolt per degree centigrade. It is non-
linear. Suppose in some range you will find 10 microvolt per degree centigrade, some
other range you may find 12 microvolt per degree centigrade, some other even less, might
be 6 to 7 microvolt per degree centigrade.
Now calibration, actually it is very important. It is a procedure that involves a comparison

of the particular instrument with either a primary standard number 1, a secondary
standard with higher accuracy than the instrument which is to be calibrated or a known
output source, right. What does it mean? It is a primary standard. That means I can
calibrate my instruments with a very accurate instrument by going to some National
Laboratory or some other places. I can calibrate that instrument with a secondary
standard of higher accuracy. I can use some other instruments which I know they are
accurate. So, I can calibrate. Because you see, it is a hardly, in industrial instrumentation
you will find that the instruments are of absolute type. That means you do not, it does not
need any calibrations. You have to calibrate each and every instrument, so most of the
instruments you will find, in that case what will happen that I must calibrate either with
some primary standard.
Suppose I have, I mean I have given you a RTD. So, you must calibrate that with some
small range, suppose I can calibrate with some standard mercury in glass manometers.
So, I can calibrate like that or I can calibrate with some suppose boiling water which is
supposed to …. normal pressure and temperature. At normal pressure it should be 100
degree centigrade for the pure water. So, that way we can calibrate, right or I can
calibrate with a known output source. What is that, I will show you. For an example, a
flow meter can be calibrated with a standard flow meter available in the national
laboratory. This is number 1.
Number 2, it can be calibrated with another flow meter of known accuracy. It can be
directly calibrated with a primary measurement such as a, such as weighing a certain
amount of water in a tank or a recording time required for this quantity to flow through
the meter. How does it look? It looks like this. It looks like, suppose I have a flow meter,
I have a vessel. So, I know how much is the height, I know how much is the water
flowing out through this one. So, I can tell what is the flow rate, because if I have a stop
watch I can tell. So, this is some other form of a calibration. So, I know the quantity of
water, I know the time, so I know the volume. So, I can find the volume rate. That means
V by time. If I, so I can, litre per minute I can tell that how much is the volume. So, I can
calibrate in that way also, right.
Now, dead zone or dead space is another important factor. Now, in many instrument you
will find it has a dead zone. Dead zone is the largest value of a measured variable for
which the instrument does not respond, right. What does it mean? Suppose I have a
pressure gauge which is measuring a pressure of 10 to the power 5 psi. So, in that type of
pressure gauge if you give a 50 psi you will find that there is no movement of the pointer.
So, that is the, 50 psi is the dead zone of that type of instrument, right. So, dead zone
usually occurs with a friction in a mechanical measurement system.
Hysteresis is another important … Hysteresis error refers to the difference between the
upscale sequence of calibration and downscale sequence of calibration and the hysteresis
error of an instrument is given by y upscale minus y downscale at X equal to X 1. What
does it mean? It looks like this.

The hysteresis error is depicted in figure and it is usually expressed in terms of the
maximum hysteresis error that is the percentage of the full scale output range. Hysteresis
error equal to hysteresis, maximum hysteresis error upon full scale output range
multiplied by 100. It looks like this.
In this figure you can see here. You see, so this is our hysteresis error at X equal to X 1.
So, this if you take the full scale output range, so you will get the, I mean total hysteresis
is there or I should that it is non-matching of the upscale calibration and downscale
calibrations. It may happen in some instrument that it does not match. In that case we will
call it hysteresis.
The input impedance at the input of each component in a measuring system there exist a
variable X i1 and at the same point there is associated with a X i2 another variable, such
that the product has the dimension of power. When these two signals are identified, we
can define the input impedance as shown by Z i equal to X i1 upon X i2. This is
important because some instrument and the, I mean …

So, here the power drain is P d equal to X i by Z i, X i1 square by Z i and that a large
input impedance is needed to keep the power drain small. Because in some instruments,
some sensor you will find like a, like a pH meter, all these things need very small, should
draw very small current. So, this is very important, because if it draws large current it
cannot measure the voltage actually, because it has internal large source impedance, right.
Now, resolution is another important … The measurement precision of an instrument

defines the smallest change in measured quantity that can be observed, right. Suppose I
have a thermometer, 0.2 degree centigrade. It is the minimum, so it is 10.2 centigrade.
For example, it looks like this. It is a temperature transducer. If the 0.2 centigrade is the
smallest temperature change observed, then the measurement resolution is 0.2 degree
centigrade.
Now, bias is another important … It means a constant error exists over the full range of
measurement of an instrument. The error can be easily removed by calibration. For an
example, a voltmeter perhaps shows a reading of 1 volt with no input voltage to its
terminal. If now a known voltage of 30 volt is applied to the voltmeter, the reading
should be 31 volt. So, obviously I can, this constant bias of 1 volt can be removed by
calibrations or by simply mechanical electrical means. Suppose by adjusting the needle or
in case of …. voltmeter, I mean we should take care, we should charge some capacitors
when there is no, I mean input voltage, so that that biased voltage will be charged.
Accordingly, it will be subtracted or added to the final value, when you are measuring
some voltage.
Drift is the calibration of an instrument. It is usually performed under some control

conditions of temperature and pressure. Because, not always we have control conditions,
so there will be drift in the measurement. As variations occurs in the ambient
temperature, some static characteristics change, namely zero drift and the sensitivity;
sensitivity may change, zero of the instrument may change. So, zero drift, it is, it
describes the effect where the zero reading of an instrument is modified for the change in
ambient conditions.
For an example, the zero drift of a voltmeter, zero of voltmeter for the ambient
temperature change is expressed in volts per degree centigrade. This is sometimes called
the zero drift coefficient related to the temperature changes. If there are several
environmental parameters, then it will have several zero drift coefficients. Effect of zero
drift is to impose bias on the instrument. Sensitivity drift is another, because sensitivity
also will change. The difference, the amount by which an instrument sensitivity, it
defines the amount by which an instrument sensitivity of the measurement varies as
ambient condition changes, right.
So, this is all about basic introductions of the industrial instrumentations. I will keep the
static characteristics in details and you have understood by this time, how the role or what
is the role of the instrumentation and how important it is and the responsibility of the
instrumentation engineer is to maintain all those sensors, signal conditioning circuitry and
all types of things. This ends the lesson 1.
Prof. Alok Barua
Indian Institute of Technology – Kharagpur
Lecture - 2
Dynamic Characteristics
Lesson 2 of industrial instrumentation, we will discuss basically the dynamic

characteristics of instrument or sensor. Now, one thing I told you that a sensor or
instruments, we are I mean rather using very loosely here the term, because sometimes
we call it sensors, sometimes we call it instruments as such. When I will call the second
order instrument or first order instrument, it does not necessarily mean only instrument as
a whole, it may mean also the sensor.
Now, the contents of this lesson are: dynamic characteristics, zero order instrument, first
order instrument and its responses to step input to ramp input to sinusoidal input. You
see, in both the cases, in the case of first order and second order instruments, we will
consider these three different inputs and we will compare putting these side by side.
1
At the end of this lesson, the viewer will know the characteristic equation of an
instrument, zero order instrument and example of some zero order instrument, first order
instrument and its responses to step, ramp and sinusoidal input. In the instrumentation
system, you will find that most of the time you will face these three different inputs. Even
though ramp input is not very usual, we will find, instead of ramp the input will be
sinusoidal input. However, we will consider the ramp input also and sinusoidal input is
very common like ambient temperature variations or in some biomedical applications, we
will find these types of inputs will be there.
Step input is very common in the electrical system, because suppose you are using a
voltmeter, you are giving the input to that voltmeter, it is considered as a step input and
also you consider the error in measurement. How much the error, because until unless
you know the error you cannot reduce it. What are the factors on which this error will
depend, you must know that thing and how the errors will behave as the time goes that
also you should know. For that reasons, we should also consider error in the
measurements in all the three different, for the three different inputs and physical
parameters that influence the error that I told you just now.
2
Now, if I look at the dynamic characteristics, the dynamic response of an instrument to a

signal input may be described by the nth order differential equation such as following. It
looks like a n, nth order derivative of y with the respect to t plus a n minus 1, n minus
oneth derivative of y with respect to t, so on until it is a 1 multiplied by dy by dt plus a
naught y equal to b naught x, where y is the measured quantity or value indicated by the
instrument, right, x is the input quantity.
For any measurement systems or sensor, we will actually, we will want that whatever the
x, y should show that without any time delay, without any phase lag. But, this will not
happen. You will see that there will be some lag in the system. There will be some, some
delay in the system. t is the time and where a naught, a 1, a 2, so on up to a n and b 0 are
the constants which are the combination of the system physical parameters. What is the
meaning of the system physical parameter? Suppose I have an electrical circuit. That R
and C will be the, I mean suppose in the case of, I mean electrical circuit R and C will be
the factors which will depend, suppose I have a circuit like this, I have a simple electrical
circuit.
3
So, I have a resistance capacitance. So, if I draw, so the value, R and C value, so the
physical constant will depend on the value of R and C.
Now, here I will show you a block diagram of a measurement system. I have an input
quantity X t, I have an output quantity Y t and I have some initial conditions. These are
necessary; these initial conditions are necessary to solve the differential equation. In the
4
case of first order instrument, it will be first order differential equation. Now, zero order
instrument is the simplest instrument we have, but there are some instruments also, some
sensors in that which are zero, zero order in nature.
The simplest model of a measurement system is a zero order differential equation. It

looks like this: a naught y equal to b naught x or y equal to b naught by a naught x,
multiplied by x, equal to K x. You see, if you go back to our initial differential equation,
it will look like, you see, this is our differential equation.
5
So, this differential equation, nth order differential equation if you put all zero that means
from a n up to a 1 that means a n, a n minus 1 up to a 1, if you put all equal to zero, I will
get the zero order instrument which looks like this. That actually I have shown just now.
So, only this part if you take, it is a zero order instrument, right. So, this is our zero order
instrument.
6
You can see here y output equal to b naught by a naught multiplied by x or equal to K x,
where K equal to b naught by a naught. I am sorry, this will be and this will be b naught
by a naught, b naught by a naught and K is called the static sensitivity of the system. K is
called the static sensitivity of the system. Static sensitivity has a lot of influence. You will
find the calibration curve of the entire system will depend on the static sensitivity.
Remarks - you will see that it is obvious that x may vary with time. However, the
instrument output follows it perfectly with no distortion or time lag. Whatever the input
in the system, it does not matter. It will immediately appear at the output without any
phase lag in the system. So, that is a zero order instrument. It does not matter what is the
input. It might be step input, it might be ramp input, it might be your sinusoidal input. So,
immediately it will appear at the output without any lag, without any phase change. So,
that means it is a zero order instrument.
Now, in zero order behaviour, the system output is considered to respond to the input
signal instantly. The static sensitivity is found from the static calibration curve of the
measurement system. It is the slope of the calibration curve. How does it mean?
7
It means that if I look at it will be, suppose I have this characteristics calibration curve,
this is x, this is y and I have a linear relation like this one. So, slope of this calibration
curve, okay that is my dy by dx will give you the static sensitivity. So, this will give you
the static sensitivity and since it is linear, you see anywhere the value of dy by, del y by
del x, so if it is del x, it is del y, it is same.
8
Now, I will take an example of a potentiometer. Potentiometer is a very important sensor
you will find in, in all electronic circuits. Also, in instrumentation you will find it is used
as position sensors; rotational positions, linear positions for all this different reasons it is
used and in the electronic circuits, potentiometers are extensively used. Whether you call
it potentiometer or you call preset or you call the rheostat, all are same. These are
basically potentiometer.
Potentiometer is a three terminal device. You see here, this is our potentiometer. You can
look here. So, I have a, supply excitation voltage I have given and total angular rotation
of the potentiometer, because if you look at the physical shape, it looks like this. I have,
the potentiometer looks like this, isn’t it? Here, so I have a wiper or jokey. In the case of,
in the case of potentiometer or preset we call it wiper, in the case of rheostat we call it
jockey, but the principle is same. Both are zero order instruments.
Potentiometer is used in electrical, electronic circuits and rheostat is used actually in the
case of electrical circuits and presets are used for, it is not for multiple use, but for limited
number of ways if you use 1 or 2 or 3 times you can use preset which is basically nothing
but a low cost potentiometer. So, I can say, the rheostat, potentiometer and the preset are
basically same. So, you see here, the total rotation is theta T. So, I can, may be I have any
position theta, so I am giving the voltage, battery voltage here, connected here. So, this is
my e, right and I can measure this voltage, voltage between this point and this point. That
will be my output voltage e naught. So, this can be e naught, right; this can be my e
naught, clear. So, this will give you the, so total is theta T. So, this will give the, I mean
symmetric view of a potentiometer.
9
A potentiometer is the example of a zero order instrument and its equation looks like e
naught equal to theta by theta T multiplied by a supply voltage or excitation voltage,
which is equal to K theta, where K equal to E by theta T which is volts per radian.
Because E is a constant, it does not depend on the position of the potentiometer. Theta T
is the total angular … These are constant. For a particular potentiometer, you know this is
constant. That is K equal to E by theta T which is volts on radian and this is the static
sensitivity of a potentiometer.
10
You can see here that the potentiometer, that it always should, should whatever the input
it will get, immediately it will be shown at the output without any time lag or without any
phase change. That actually happens in the case of potentiometer. That means whatever
the inputs you are giving, it immediately appears at the output. I can use it as position
sensors also, because as you can see the output voltage is directly proportional to the
position. What is that position?
11
If I go back, I can see here, you see here that output voltage is directly proportional to
theta. So, e is a function of theta, so which can be utilized to make a position sensor and it
is a zero order instrument. So, obviously there is no phase lag or no time lag, phase
change or time lag in the input output, okay.
Now, next we will consider the first order instrument.
The transducer that contains the storage element cannot respond instantaneously to
change in input. The mercury in glass thermometer is an example of a first order
instrument. The bulb takes energy from the environment until the two are at the same
temperature or steady state condition has been reached, right? The temperature of the
bulb will change with the time until the equilibrium is reached. The rate at which the
temperature change with time can be modeled with the first order derivative and the
thermometer behaviour is modeled as the first order differential equation.
12
Therefore, the dynamic characteristics of a first order instrument is given by, looks like
this: a 1 dy by dt plus a naught y equal to b naught x. This is similar. You see that if you
remember, our nth order differential equation that a n, then d n y by dt n, all those things,
if you put all the coefficients from, from a n to a 2 from a n, a n minus 1 so on, a 2 equal
to all zero, I will get a first order characteristics equation or a first order instrument.
So, exactly we got that thing, you can see here.
13
So, a naught, so you can see here, so that is we have multiplied entire equations by a 0. I
got the equation a 1 by a 0 y dot plus y equal to b naught by a naught into x. So, dot we
just replaced by dy by dt. So, tau, now I replace a 1 by a 0 equal to tau. Tau y dot plus y
equal to b naught by a naught, upon a naught x. Finally, we have written tau y dot plus y
equal to K x, right. K we remember.
For the zero order instruments also K is a static sensitivity of the system and tau is the
time constant of the system and this is the equation number 1, where tau equal to a naught
upon a zero and K equal to b 0 by a 0. Obviously all this, I mean time constant of the
system like, as you told, because if you take an example of a thermometer this a 1 upon a
0 and static sensitivity b 0 upon a 0, depends on some of the physical parameters of the
system. What are those physical parameters? That means what type of, what is the size of
the bulb, of the thermometer, materials which you are using as a thermometer liquid, all
these factor will actually tell you the value of the tau.
If the, if I use a for an example, just from intuition you can tell if you take a large bulb, so
that I, if I have a large, I mean length of the mercury, so in that case I can use, the time
constants of the system will also increase. Similarly, the K also will be actually controlled
14
by the physical parameters of the system and if you remember, our first, the nth order
differential equations, we have written there clearly this value. That means a 0, a 1, b 0,
all are the basically, depends on the physical parameters of the system. Tau is called the
time constant of the system and it always has a dimension of time, quite obviously.
Now, now you slowly consider one by one, the inputs. Now, first I consider the step
input.
Now, step input is very common to the system. Especially in the electrical systems, step
input is very common. Step input function is defined as X equal to 0 for t less than 0 and
X equal to x s for t greater than equal to 0. This is equation number 2. These set of
equations, we have given the equation number 2. These are necessary to solve the
differential equations, right. How does it look? It looks very simple. It looks like that my
input, …. time axis if I plot, so it looks like this. Then, suddenly it goes. This is my input.
If I take different colours, well little more, so it will, my input looks like this and go like
that.
15
So, this is our zero. That means at t less than 0, it is X equal to 0 and this is our X and for
t greater than 0, X is a constant. x s, x s actually I want to signify the steady state value,
because this will be our steady state value of the input. I want, my output will be always
exactly equal to x s, because if it is x s, …. the steady state value, right.
Now, substituting equation 2 in equation 1, for t greater than equal to 0, we get tau y dot
plus y, tau y dot plus y equal to Kx s. This, I have given the equation number 3. The
solution of this differential equation, equation 3 gives for t greater than equal to zero
equal to, y equal to Kx s 1 minus e to the power minus t by tau, right; very carefully, it is
e to the power minus t by tau. It has lot of significance, the meaning of this and KX s. As
you can see that as the time goes, what will happen to this that as the time goes, this is a
transient part, this will die out. So, ultimately we will find that the output will follow the
input, because at that time this will be y equal to KX s into 1. This becomes zero, so this
equal to 1.
This means that whatever the static sensitivity multiplied by the steady state value of x
will give, will be the output, right. But, you will see here that this tau carries a lot of, I
mean lot of influence, at what time you will get that y will be equal to KX s. If the tau is
16
large it will be, it will take long time to die out and if the tau is small it will take a very
quick time that it will reach the final steady state value. Output will reach the final steady
state value, because it is minus term is there. So, let us play it so that, this we have done
in flash, so it will be better if you look like this.
x-axis is my input, y is, my y-axis is my output y by KX s. Now, interestingly, you see

one thing. I have normalized it, because this is very much necessary, okay. Let it finish,
then we will start again. Hopefully it has finished. Now, see here that I have taken this is
my y by K axis. I have taken normalized value. So, y will be the unit less quantity. So,
since it is normalized, so it is unit less, right and this value, maximum value will be
always 1. Since it is now 1.0, 1.0, so this will be always 1 and you have plotted t in the x-
axis.
As you can see that, as I increase the value of the tau, it is taking longer and longer time
to reach the final steady state. This is our final steady state value. So, as I increase the tau,
so I will get more and more time. So, I can have, if I still further increase the tau, so it
will go like this. Suppose I have the, one of my example of this that suppose I have a
thermometer; thermometer, a simple mercury in glass thermometer, fine. It has some,
17
depending on the type of I mean liquid which you are using, the type of the thickness of
the glass and all those things it will have some value of the tau, because physical
parameters will control the value of tau.
Now, what will happen, you see here that if the, if I for protection suppose that mercury
and glass I put on a steel sleevings that means I have mercury in glass manometer, sorry
mercury in glass thermometer, so I put on a, for protection I put on sleevings, so that it
will not break. For that type of situations we will find that the time constant of the system
will increase. Later on you see that we have various types of, I mean thermometers. We
have thermocouple, we have thermistor, we have RTD. You find that in the case of RTD,
because of its large, huge size you will find it is the, time constant of the system is very
large compared to the thermocouple or thermistor. So, time constant will control that how
quickly your output will reach the input. So, in this case you will find that the tau
increases that means your time constant increases, right.
Error in measurement at any instant of time is defined as e m. We have used the subscript
m that is to tell that it is an error in the measurement x minus y by K. So, it will be
ultimately X s minus X s multiplied by 1 minus e to the power minus t by tau and it is for
18
a step input equal to X s into e to the power minus t by tau. We have seen that, you see
that as the time goes, so this error will be smaller and smaller, because it is exponential
term and there is negative, so obviously as the t will be large and large, so this will die
out. At the initial stage when t is zero, so it has large value. Its value is almost equal to, it
is exactly equal to X s, the steady state value. As the t goes, so it will die out.
So, it will be better visualized if you run in a flash. So, the normalized error if have I
normalized, if we make it unit less, so it will be e m by X s equal to e to the power minus
t by tau. It will be better visualized, I mean in a flash it will, you can see here.
So, x-axis we have plotted t by tau instead of t. It does not matter, because it is the
normalized …... Because t is a dimension of time, so we have now plotted normalized
value. It does not matter, because it is t just multiplied by some constant factor of 1 by
tau. You can see here that at a very low value of t or at very low value of t by tau or when
it is zero, initial stage we have just given step input, my output is maximum. This is our
error, error is maximum. So, as the time goes, so obviously the error will fall down. It is
exponential decay and you will find that as the time goes, so my error will get decreased
or the normalized error is getting small and small and ultimately it will die out. So, this is
19
very important in the sense you will find that in a system that you must know that when
the error will die out.
In some cases I cannot; my entire reading will be erroneous, because from a sensor, the
reading from the sensor, because if I take reading, suppose if I take the reading of a
sensor after one of t by tau, so it is totally error. So, I cannot accept. So, I should take the
reading when the t by tau is 4 or t is 4, when 4 seconds, suppose if it is, I mean if it is, I
mean if I plot it in, in terms of t instead of t by tau or if it is suppose tau is 1. So, after 4
seconds I should take the reading, okay.
Now, we should consider the ramp input. Ramp input is not very usual. I mean in the
instrumentation system, it is very difficult to give a ramp input, but electrical system it is
very necessary. You find in some of the cases you have to give the ramp input. One of the
good examples of the ramp input is saw tooth waveform.
See, you are aware of that in the case of, always in the, this is the saw tooth waveform
you give in many electrical circuits, this type of waveform. So, one of the common
examples of the saw tooth waveforms when we give in the, in the oscilloscope, in the
20
cathode ray oscilloscope to measure some voltages. Voltages, amplitude we are giving
between the plates y, particle plates and the saw tooth waveforms we are giving in the x
plates, so that to synchronize or to make the wave stationary.
In instrumentation system, saw tooth, I mean raw, I mean ramp input, you will find it is
not very usual, even though we will consider the ramp input. In many cases we will find
that instead of ramp input, the input which I will give is basically a terminated ramp
input. In the case of step input also, sometimes very difficult to give in instrumentation
system. So, it will appear actually the terminated ramp input like this one instead of step
input. So, in this case, you must consider this portion which looks like a ramp, okay. So,
for that reasons we consider here the ramp input.
Now, the ramp function input is defined as X equal to 0 for t less than 0 and X equal to X
s dot into t for t greater than equal to 0, right. It looks like this that if the time is varying,
so it is going on increasing like this, right. This is t. So, this is at any step time, what will
be the value of X that can be multiplied by t. So, this is our X s, X s dot. Slowly it is
increasing, right. So, this is our ramp input.
21
Recalling the characteristics equation, equation 1 of the first order systems, we call that
first order systems here, we just replace X by X s dot t. X s dot means dX s by dt into t.
So, tau y dot plus y equal to X s dot into t. This is our first order differential equation for
ramp input. We will solve this equation and see how the output will come, how the
response will come and how my, how my error in the measurement system will come?
22
The initial conditions are x equal to y equal to 0 for t equal to 0. That is quite obvious.
That means we may assume that the ramp input, it does not matter that it will give an
input like this one. So, at t equal to 0, so quite obviously I can say that the x is equal to 0
and y is equal to 0.
And my output will look like y equal to KX s dot tau e to the power minus t by tau plus t
minus tau, right. Now, error in the measurement at any instant of time will be given by,
see, e m equal to X minus y by K equal to X s dot into t minus X s dot tau e to the power
minus t by tau minus X s dot into t plus X s dot into tau equal to minus X s dot tau e to
the power minus t by tau plus X s dot into tau, right. So, let us go back. So, this is our,
you see, there are two parts. One is the transient part and there is a steady state part. Why
it is transient? You see that, this as the time goes, this function will be, this function is
going to reduce. As the time goes on, so this will become, almost approaches to zero, but
this will remain as it is. So, this is our steady state error and this is our transient error,
right?
23
You see, if I plot it, it will look like this that steady state error at any instant of time,
okay, hopefully it is finalized, it looks like this. So, x-axis we have plotted time and this
is our output. These I have plotted. This is my input, so this is equal to actually X s dot
into t, right and this is y by K. I have plotted the output. Actually this should be parallel, I
mean it is, I cannot draw it nicely. I think it should be like this then. It should be parallel
to the input.
Here the steady state time lag you can see; it is the steady state time lag of the system.
Initially time lag is very small, but as the time goes, so we have a steady state time lag
which is tau and steady state error also you can set. The initial stage it is, steady state
error is very small.
24
Why? If you look at, if you go back, if I go back to the previous slide, you see here, you
have a negative term here. So, that will contribute, so this will become almost zero for
some value and as it goes, as the time goes, so this will become insignificant, so this will
be dominant. So, that is the reason we got the characteristics like this. Here,
characteristics, this is my output and this is my input and this is, my output will look like
this. So, error initially is very small and as the time goes we have a steady state error
which is X s dot into tau.
25
And interestingly, you see this steady state time lag is always tau. So, it means that steady
state error will be at any instant of time, any instant of time X s dot into tau and so it is,
time lag is tau.
What is the remarks of a, I mean for a system with a step input? It is apparent that the
smaller the value of tau, the faster will be the disappearance of the transient error. This is
26
everywhere I found that the, if the, in the case of step input also you find that if the tau is
small, so I will get, immediately I will reach the final steady state value. If the tau is
large, it takes long time, right? Long time means what? The steady state error will take
long time, to make the steady state error zero. It is our goal for any instrumentation
system or any sensor that the steady state error should be zero as quickly as possible.
Now, it is apparent that the smaller the value of tau, the faster will be the disappearance
of the transient error. Moreover, once the transient has disappeared, the instrument lags
behind a constant value which is again the time constant of the instrument, tau. That is
already we have seen, right?
Therefore, it is obvious that the lag of the instrument is dependent directly on the time
constant tau. For an example, a temperature sensor having a time constant of 5 second
will ultimately lag behind a ramp input by 5 second. Initially lag will be small, but after
sometime we will find the lag will have steady state value, say it will be 5 second. So,
whatever the time constant of the system, our lag also will be the same. So, the
measurement error is directly proportional to the ramp input and time, okay. This is also
very important.
27
We have seen that the measurement error depends on the ramp input as well as the time
constant of the system. So, if the time constant is small, the measurement error will be
small. If the time constant is large, the measurement error will be large and you see,
unlike the step input this measurement error will remain forever. If you look at the, in the
case of step input, I have an input, I am giving an input like this, right, I am getting output
like … So, after sometime, there is no steady state error. This will be zero.
But, whereas in the case of ramp input, you see here, if I take a white page, in the case of
ramp input, so there is always a steady state error, initially small. So, it is like that, so
there is always a steady state error. This is our error. So, it will remain forever in the case
of ramp input, since it is a time varying input function, right.
28
If low measurement error is required, the instrument must have a low value of the time
constant; it is quite obvious, if the measurement error low is required. It is not necessary
in the case of the step input and all these things, because ultimately it will reach …. Only
time constant will control the value that means after what time the steady state error will
be zero. Whereas, in the case of, I mean ramp input, I mean if you want to reduce the
steady state error, it will be, always you have to choose the value of the time constant
smaller and smaller.
Now, I will consider the sinusoidal input. Sinusoidal input is common that the ambient
temperature variations or some, in some particular applications like the biomedical
applications, you will find, when the patient is infected with malaria parasites, we will
find the temperature usually for the first two three days that that particular time only the
particular temperature appears, so very periodic. So, ambient temperatures also you will
find the daytime there is some temperature, nighttime some other temperature and
interestingly, the time constant of the system is obviously very large, right? So, sinusoidal
input, we will face a lot like vibration analysis and all those things.
29
You see, the periodic signals are encountered in many process such as vibration analysis,
ambient temperature variation, then, etc. When a periodic signal such as sinusoidal input
is applied to a first order system or instrument, the frequency of the input signal influence
the response of the measurement system. So, the frequency will control a lot of
parameters; we will find that. So, frequency is of prime importance in the case of
sinusoidal system.
30
The sinusoidal function is defined as, we can write the sinusoidal function like the X
equal to X s sin omega t; very simple, we all know. That means if I have a sinusoidal
function, sorry, so this is my X s, okay. So, this is total time period, T where T equal to 1
by f and omega equal to 2 pi f, circular frequency equal to f and T is the total time period
and time period 1 by inverse. So, this is X s value on both side, this is also X s, so for T
greater than equal to zero and for T less than zero, it is zero. So, this our sinusoidal input.
So, the characteristics equations will be as follows. Obviously, I have just replaced on the
right hand side x by Kx s. Actually it was there, so I have replaced tau y dot plus y equal
to Kx s into sin omega t. Initially our basic equations, do you remember in the very
beginning of the lecture, it is tau y dot plus y equal to Kx s. In the case step up, in the
case of step input we have replaced with K equal to x s. In the case of ramp input, we
have replaced x equal to x s into t. In the case of sinusoidal input, we have replaced x by
x s sin omega t, right?
31
So, the solution to this differential equation yields, ignoring initial condition that y zero
equal to y zero, get Kx s upon under square root 1 plus omega square into tau square sin
omega t minus phi, where phi equal to tan inverse omega tau.
This we can write as equal to A sin omega t minus phi; phi is the phase shift of the input
signal. There will be, there will be some phase shift that means that my, if my input is
32
like this, so my output will look like some other, so it will look like a, phase shift like this
one, right? So, there is a shift in phase. Sometimes, in many instrumentation systems you
will find that this phase shift is a nuisance. That means we have to kill this phase shift. In
the case of leading phase angle we have to use a lag network and in the case of lagging
phase angle we have to use a lead network.
Even though the circuit is very simple with one resistance and one capacitance, I can
make a lead lag network, right? So, this is my output, you can see here. So, the phase
change is coming and as I told you earlier, you see, this omega is very important in the
case of sinusoidal signal. That means omega equal to 2 pi f. What is omega? Omega is, if
I draw it here again, so if my input signal is like that, so this is my T, total time period, so
f equal to 1 by T and 2 pi f equal to circular frequency omega. So, this omega will
influence a lot. It depends on, the phase shift also influenced by both omega and tau.
You can see that the phase shift is equal to tan inverse omega tau and the amplitude of the
signal which is A, which is I mean K into x s under the square root 1 plus omega square
tau square, we will find that if that omega is large, obviously your value of y also will be
small, value of A that means the amplitude also will be small, right?
33
Where A equal to Kx s upon under the square root 1 plus omega square into tau square
that A represent the amplitude of the steady state response and phi is the phase shift of the
output response with respect to sinusoidal input, right?
It is apparent that the amplitude of the output response of a first order system depends on
the frequency of the input periodic signal, right? The delay in the measurement is given
by D equal phi by omega and is expressed in seconds, where phi is the radian. I am sorry,
this will be, this will be, where phi is the radian and omega is in radian per second, right?
Now, delay in a system is important. In many cases we will find we cannot allow delay.
In some systems, we will find that we have to, I mean we have to accept this delay,
because this delay is a natural phenomenon; like a phase shift, it is a natural phenomenon.
Only thing we have to make the delay possibly to, so that the delay will be same for all
the frequencies. We cannot make delay zero, but if the frequency, for all the frequency
range the delay is same, so that will satisfy our goal.
34
The amplitude ratio is given by A by Kx s equal to 1 upon under the square root 1 plus
omega square tau square, we can see here.
Now, we have plotted here, you see that, plotted normalized value of the output, okay
versus omega tau. Instead of omega, we have plotted omega tau. We want to make it also
unit less that is the reason we have plotted. It is dimensionless. Omega is in radian per
35
second, this is seconds, so it is dimensionless we have plotted. So, this is our response of
a first order system or the frequency response of a first order system, okay with
sinusoidal input, right? So obviously, since it is normalized, this output will be 1 and this
will be 20 log 10 of A by Kx s will be zero degree, right?
You can see here that if we increase the frequency, so our, there is a roll off and for the
first order system, this will be minus 20 db per decay, right? So, this is a roll off and that
is the phase. So, it is the flat response, after that it will start to fall.
Now if I, if I plot the phase, you will find of a first order system with, I mean, sinusoidal
input, it looks like a phase shift of 45 degree. You see the phase shift of 45 degree at
omega t equal to …, because you see, you remember that phi, phase shift is tan inverse
omega tau. So, if omega tau equal to 1, so phase shift will be obviously 45 degree. So, it
is 45 degree at omega equal to 1, right and at very low frequency it is zero phase shift as
the frequency is becoming high. We have a, this omega is high …… I mean, tau is
constant. Omega is getting higher and higher, so if it is infinite, obviously I will get a 90
degree phase shift, right. So, this is about the amplitude and the phase response of a, of a
first order system with sinusoidal input.
36
Now you see, this is important, because that I told you that earlier that omega plays a key
role in the case of response, in finding the response of the first order and second order
instruments with the sinusoidal input. So, the frequency will determine or the tau will
determine at what frequency you can use your instrument, because in other way, because
if the, if you fix the omega, the tau will be determined and if you know that my signal has
some particular frequency, I can tell that what should be the value of tau, for that reason
that I will get a reasonable response of the input to the system.
Now, remarks will appear like this.
It is obvious that both the time constant of the system, both the time constant of the
system and the input signal frequency influence the system response. So, the system
response will depend on both omega and tau. We have seen, you see, why because you
see, the amplitude also, you remember A equal to Kx s root over 1 plus omega square tau
square. Isn’t it, we have seen; so, this is my response or if I take a normalized, so it will
be A by Kx s equal to 1 upon root over 1 plus omega square tau square. So, as we say that
both the time constant of the system and the input signal frequency which is omega,
influence the system response, because it will control the value of the amplitude or the
37
normalized output as well as the phase shift, which basically depends on tan inverse
omega tau, right?
For those values of omega tau for which A equal to A by Kx s is almost equal to 1, the
measurement system will have almost no attenuations of the input signal amplitude and
there will be very little time delay. That is always we want that we have shown the
response. Isn’t it?
38
If I go back and see, our response we look at, for the frequency when I am getting the
response here 1, so that is this desirable property. I cannot make the measurement at this
region. I want to make the measurement in this region only, right, isn’t it? So, this will be
my frequency response of the operation, so that the omega tau has a lot of influence.
39
Now, if the measurement of high frequency signal is desired, then the system having a
small tau will be necessary. Quite obviously, because our response we have seen again
that the response falls down. So, if I want to take the response at the higher frequency, I
must use a small value of tau, so that the omega tau will be, I can accommodate omega
tau, will not be large, so that I can accommodate my frequency in this region, right?
A large time constant system will result in the removal of the high frequency component
from the output signal. Quite obviously, I mean if the frequency is high, we have seen
that the, if we look at the plot, right, so that what will happen? That means I have a plot
like this. So, you see, here it is falling down. So, if the omega tau is large, so if the tau is
constant, what will happen? That the frequency component of the output signal, so it
falls. If I want to take the reading here, so obviously my output attenuations will be quite
large. So, I won’t get any output, right. So, that is we are telling; a large time constant
system result in the removal of high frequency component from the output signal, right?
Now, you see that in measurement systems, so the first order instrument is, even though
is not very common, but there are some instruments which you will find first order. But
suppose, if I take a mercury in glass manometer, I mean thermometer, it is a first order
40
instrument and the other one if we consider that the mercury in glass manometer, then we
can consider as a second order instrument. Now, first order instruments and second order
instruments the difference is we will find that as I told you that if you look at the output,
if you know the outputs of the systems and if I can or if I have a calibrated output, in that
type of situations I can tell that what is my output and if my differential equations I have
approximated as the first order systems, I am getting the desired output or the output
which is supposed to I get, because if it is calibrated, I must know what is the output. So,
I can tell the system is first order.
See, if i am not getting, say I have to go for higher order instruments, for a second order
instruments or third order instruments, but in the entire course we will find that we will
concentrate only on the first order and second order instruments. If the system is non-
interactive, so I can go for a higher order instrument. Suppose I have a thermometer
which is cascaded with a second order recorder, so in that case it will be a, I will, suppose
I have an example I am giving.
Suppose I have a, suppose I have a first order instrument, what happened? I have a first
order instrument cascaded with a second order instrument. Suppose it is a thermometer
41
and I have a recorder. So, in that case it will be, if it is non-interacting, I can consider it as
a second order third order instrument. But mostly we will consider these two separately,
first order and second order instrument as a, separately.
So, this we have considered the, here in this we have considered the dynamic
characteristics and the first order instrument and these dynamic characteristics of the first
order, I mean of the instrument, not for the first order, for the second order instruments
will be considered in the lesson, lesson 3 of this course, industrial instrumentation. This
ends the lesson 2 of industrial instrumentation.
This is lesson 3 of industrial instrumentation and we will continue with the dynamic
characteristics of, of a system especially second order systems. Already we have
discussed about the first order system. Now, we will consider the dynamic characteristics
of a second order system.
42
Now, contents of this lesson are the dynamic characteristics, the response of a second
order system for step, ramp and sinusoidal input, where we did it for the first order
instruments we will do for the second order instrument. Also, we will solve some
problems on the first order and second order instruments.
43
If I look at the ramp input, ramp input we have seen, characteristic equations of a second
order system with ramp input can be expressed as follows. Already we have discussed the
ramp input in the case of first order system. So, we are taking same three inputs, so that
we can compare putting these two side by side. D square upon omega n square plus 2 psi
D by omega n plus 1 y equal to Kx s dot into t. y equal to dy by dt equal to 0 at t equal to
0. This is initial condition, so we have discussed this before also. What does it mean? ….
x s dot, but my input looks like this.
This is our ramp input. We have discussed thoroughly these when we discussed the first
order system, right? Now, this we can, I mean do it in the case of second order systems.
So this is my input. This constant K, as it happened before also, the x s dot into t, right? It
is continuously changing with time, so x s dot, so I made it unit less quantity. So, these
are our initial conditions.
Now, the last problem; we will find in the second order system, so there is no question.
So, if it is second order system we have to take both plus and minus of the dynamic error.
That is the reason we have given. So, you have to take both. That means you have to take
both plus or minus of 6%, so accordingly you will keep the value of the omega n. So, the
44
natural frequency of the sensors you have to find here. You have to find the value of
omega n here. So, you try to solve these problems and the solutions will be given in the
next class and I will remind you that, here you see that the problem number 3.4, the
instrument is first order, problem number 3.5 also it is a first order system and whereas,
problem number 3.6 is a second order system. So, with this I come to the end of the
lesson 3.
45
Prof. Alok Barua
Lecture - 3
Dynamic characteristics (Continued)
This is lesson 3 of industrial instrumentation and we will continue with the dynamic
characteristics of, of a system especially second order systems. Already we have
discussed about the first order system, now we will consider the dynamic characteristics
of a second order system.
Now, contents of this lessons are - the dynamic characteristics, the response of a second
order system for step, ramp and sinusoidal input, where we did it for the first order
instruments we will do for the second order instrument. Also we will solve some
problems on the first order and second order instruments.
1
At the end of this lesson the viewer will know the characteristic equation of a second
order instrument, responses of a second order instrument.
Now, second order system - a sensor that is modeled by a second order differential
equation is called a second order system, quite obviously. How will I know whether it is a
first order or second order? If I, if I, if I could look at the response and if the, if the
2
characteristics, if the output is coming which is supposed to come, because for some
calibrated instruments I can tell whether how the response should look like. I f it, if
it cannot be modeled by the first order instruments we have to model by the second order
instruments or higher order instruments.
Now, accelerometers, diaphragm pressure transducers, all these things will be discussed
later; mercury in glass manometers are few examples of second order system. Now,
second order measurement system can be written as a 2 y double dot plus a 1 y dot plus a
naught y equal to b naught x.
Now, if I go back to our differential equations or the nth order differential equations
which looks like this, the …. differential equations looks like a n d n y by dt n plus a n
minus 1 d n minus 1 y by dt n minus 1 plus so on … plus a 1 dy by dt plus a naught y
equal to b naught x.
3
Now, you can see here that if I put n equal to 2, so we will get three terms which looks
like a 2 y double dot a 1 y dot plus a naught y equal to b naught x. Quite obviously as we
know, all of us know that y double dot is nothing but d square y by dt square. Similarly, y
dot is equal to dy by dt.
4
Now, this equation can be written as or rewritten as y double dot upon omega n square
plus 2 xi upon omega n y dot plus y equal to b naught upon a naught into x or I can write
y double dot upon omega n square 2 xi upon omega n y dot plus y equal Kx. K, we know
that in the case of first order systems also K is basically the static sensitivity of the
system, which is equal to b naught by a naught.
Now, we have introduced some new terms, let us look at all these terms. First of all
omega n; omega n is equal to under root a naught upon a 2 which is the undamped natural
frequency of the system. We have xi equal to a 1 upon 2 under square root a naught into a
2 which is damping ratio of the system and K is the static sensitivity of the system. Now,
we will see that this damping ratio of the system that means xi will play important role
for the second order system. Depending on the value of xi, we will find that how much
harm will be there in the systems. Typically, you will see in future that in the second
order system, in a system, that xi usually we put the value of .65 to .7, it lies between,
because we cannot make xi more than 1. Then, in that case, it will be over damped
systems and we cannot make xi equal to 1 also, it is critically damped systems; because,
if it is critically damped due to aging, because ultimately you know that all omega n, xi,
5
K depends on some physical systems that means some resistance value, some capacitance
value, some spring constant.
Now, due to aging problem, this will change, obviously. So, I do not want my system
should be over damped, no, I mean we never want until and unless we are forced to. I
will give some examples. In future we will find that sometimes even though we are
saying that the system is under damped it won’t be under damped due to the change of
the value of xi. So, always we will try to make xi at least .65 to .7, so that I will get at
least 1 over should then we will come back to the rest position.
The homogeneous solution of the equation 1 which you have written will describe the
natural and intrinsic response of any system. The output response will depend on the
roots of the characteristic equation, because the response actually will give, give you the
roots. How the, where the root likes, I mean lies for the sigma j omega plane that will tell
you the output response.
6
Now for step input, equation 1 can be written as D square upon omega n square plus 2 xi
D by omega n plus 1 whole multiplied by y equal to K x s. This is equation number 2.
Now, we have introduced a new term which is called differential operator which is equal
to d by dt of this one. That means in this case it will be first term for, suppose for the first
time it will be d square y by dt square by omega n square. This will be our first term.
Refer Slide Time: 8:08)
7
Now, we should have some initial condition, being as you know that for solving any
differential equations we need some initial condition to, to solve the equations. The initial
conditions are y equal to zero at t equal to zero, dy by dt equal to zero at t equal to zero.
So, with this we can solve this differential equation.
Now, the particular integral of the equation 2 is equal to Y p equal to K x s, but we

always, I mean look for the transit response also. So, only a particular integral will not
give the complete picture of the response of a system. So, the homogenous solution of
this equation will be given by particular integral plus complementary function solution.
So, particular integral already we have given and this is the particular integral. So, we
have to find the homogenous solution. We need to find a complementary function to give
the complete solutions of the homogenous solution for the system.
8
The complementary function, we will have three possible forms. Case 1, when xi greater
than 1 our over damped system real unrepeated roots. If you look at the sigma j omega
plane, it will look like this. I have sigma j omega plane. So, roots will be real unrepeated
roots, fine.
9
A normalized output will be expressed as y upon K x s equal to 1 minus xi plus square
root of xi square minus 1 upon 2 square root of xi square minus 1 multiplied by e to the
power minus xi plus root over xi square minus 1 into omega n t plus xi minus under the
square root xi square minus 1 upon 2 xi square, square root of xi square minus 1 e to the
power minus xi minus square root of xi square minus 1 into omega n t. You can see here
this is our, this is our particular integral and this is, that this complete this one, along with
this one will give you the complementary function.
Now, if you look at very carefully, you will find this will depend on time. So, at, after
sometime when this because of this, because of this term negative so this will die out.
Only this Y equal to K x s will remain. But, we should be interested in the transient
response of the system also. That means because in all the cases you will find what will
be the settling time of the systems, if I have to find the dynamic error of the system. So,
this, this, this complementary function is important, because in many of the cases we will
find that I cannot allow the system to get the complete, I mean steady state response
before it reaches to steady state response. Suppose 5% or 1% of the value, of the final
value I have to take some actions. But in some system it is very sluggish. But, I have, if it
is allowed that within 5% of the final steady state value that reaches it will take some
actions, because the, in instrumentation you will know all these measurements and
whether it is you are measuring the temperature, pressure flow, ultimately you have to
utilize this to control some final control element which mostly are some valves or some
heaters, which will heat the system or some boiler temperature, this, that. So, this is very
important.
10
Now, in case 2 when xi is 1, it is a critically damped system, the real repeated roots, how
does it look? It looks like that sigma j omega plane, so real repeated roots, one roots over
the other. So, it is critically damped system.
So, in that case, my normalized output will be y upon K x s equal to 1 minus 1 plus
omega n t multiplied by e to the power minus omega n t.
11
Here also you will find that, in the last cases we have, in the case of, suppose in the case
if I go back to our original previous slide you will find some sort of oscillation will come
in the case of under damped systems, because of the, you see the omega n t term is there.
I am sorry; so, omega n t term is there, so some oscillation will come.
Whereas, in the case of, in the case of critically damped system you will find if I look at
the response here, so I have 1 minus this one. So, ultimately the response should look
12
like, so it will go like this and it will go like this one. So, there is no oscillation, but it will
take some time before it comes to the steady state conditions or it is 1, because ultimately
since it is minus omega n t, you will find this term will die out, only this term will
remain.
Now case 3, when xi is greater than zero and less than 1, we call it under damped systems
when the roots are complex conjugate roots. What does it mean? It means that in the
sigma j omega plane, so this will be our complex conjugate roots. Please note that if it is
complex, it must be conjugate, otherwise I cannot realize it by any system, any physical
systems. So, if the roots are complex, it must be conjugate. That means it will be, suppose
it can be a plus minus j b. This is a real part and this is the imaginary part.
13
Now, the output in this case y upon K x s equal to 1 minus e to the power xi omega n
minus xi omega n t 1 minus xi square. Here you can see, I am sorry, that in the case of
second order system, if it is over damped or critically, then there is no question of
oscillation. Only in the case of under damped system because of this term you will find
the oscillation will come. There will be simple oscillation, after that it will die out.
How much will be the oscillations? How, how many, I mean humps you will get that
depends on the value of xi. If the xi is .1 or .01, we will get large hump, otherwise you
will get 1. If the xi is .7, .6 to .7, so it will be, it will be always just one hump and …. will
come.
14
So, if I go back, so it will, I am sorry, so if I go back and have a look in the case of over
damped systems, yes, you see, here there is no sine term. So, it is over damped system,
no sin term, so there is no hump; only it will take large time to settle down.
But in the case of, similarly in the case of critically damped system also there is no
question of oscillations, there is no sine term, sinusoidal term. But, in the case of, in the
15
case of under damped systems because of this sinusoidal term there will be, I am sorry,
there will be oscillation because of this sine term, there will be oscillation. Also, there is a
phase shift and the phase shift will be given by sin inverse, phi equal to sin inverse under
the root 1 minus xi square. So, phase shift will be there in the case of second order
system, when it is in under damped, when working in under damped conditions.
Now, mostly we will try to bias our system as second order system, as an under damped
system.
If you look at very carefully in the analog meters, especially the voltmeters and all those
things we will find, if you look at very carefully to the needle, we will find that when the
needle I have the voltmeters and it is coming. See, the needle will be, if you apply some
voltage your needle will go down here. They will make a slight …., then will come back
to the original position. That means that is where I will take the measurements?
If you want to make exactly whenever it is crossing that that is you will be wrong,
because that is overshoot. So, when it will come back to the original position that is the
position where you take the reading. That is the actual reading, because the overshoot we
16
have done purposefully due to spring and all these things that it is under damped systems
and the value of the damping coefficient is .6 to .7.
Now, if I look at the response of the system, you see, let it come, there is animation, okay
it has come properly. Now, see this is the, I made xi less than 1 and this is xi equal to 1
and this is over damped system. So, this is our over damped system, this is critically
damped system and this is our under damped system.
Now, how much will be the overshoot that will depend and number of overshoots also
depends on the value of xi and we have plotted the normalized output. Obviously, if it is
normalized it will, maximum value will be always 1. So, there is no question about this
one. Now, interesting thing if you look at the origin of these systems, here we have
normalized output omega n t. It does not matter; we can plot t also, because omega n is a
constant. So, I mean, just it is the scale factor we have plotted and here we have plotted
also unit less, because we want to plot here unit less that is the reason we have plotted
like this and this is also unit less.
17
Now, look at the origin of the systems. If you look at the origin you see there is a s curve,
distinct s curve in the system and this will tell you that even if you do not know whether
the system is first order or second order, by looking at the origin I can tell whether it is a
first order system or second order system. Well, if the system is under damped there is no
question; there will be some overshoot, it will come back. So, because if it is a under
damped system I will get overshoot and it will come back to original position. It may
happen like this also, so I have a normal …. line, it will go like this.
That means suppose I have a second order system, there can be several overshoots, so it
will come back, right. So, if it is a under damped system, always I can tell whether it is a
first order systems or second orders, because in the case of first order system there will be
no resonance, so there will be no overshoot. But in the case of second order system there
will be always overshoot.
18
Now, suppose the system is critically damped, right; suppose the system is critically
damped, in the case of first order system, I will get a response like this, right normalized
output, this is omega n t and in the case of second order system, if it is critically damped
I will get a response like this. So, look at the origin in the two cases. So, since it is, it is
under damped, I mean it is not under damped it is very difficult to tell, but if somebody I
mean does not tell you what is the system whether it is a first order or second order, you
can tell by looking at the response. There is a distinct s curve. At the initial you look at, if
you zoom it, it will look like this one, whereas in the case of first order it will be always
like this, right?
The reason, there is a distinct reason. If you look at the expressions you will find, I will
leave it to you; you look at the expressions of the first order and the second order system
you will get your answer. Initially it will go like this, then it will go below like this,
whereas here it is initially very high change, then it will go like this.
19
So, this is our response. So, we will try to make, so you can see that the over damped
system is taking long time, it is taking long time to get the final steady state value,
because any instrument what I want that its response should be 1. So, any instrument if
the response is like this, okay, if I am giving the step input, these are all due to step input,
is not it? I am giving a step input like this, I want that my output also will look like this.
But, in any physical system, no physical system can have an input like this one, output
like this one. There is a lag. So, obviously what will happen? So, it will take some time.
Now, you see this will, later on you will see these are plays very important role that this
may, how much the error, sorry.
20
See, this is the errors of measurement. If you look at very carefully, error means that
some sort of error has arrived. If suppose at the, at the time t 1 that means omega n t 1, so
some error has arrived, right? Now, so this is important, because sometimes we cannot
wait for a longtime, that the, when the, it will, it will reach that particular position. So, it
will also tell the settling time. We will see the definition of the settling time at the end of
the lesson, because always not all necessarily that always you will find that I will allow to
have this type of, I mean long time to settle down our system. Why it is important? You
see, it is also important in the case for under damped systems.
21
Suppose a under damped systems also what will happen? I will get over, I mean,
oscillation. So, in that case, suppose if I say that when it will reach 5% of the final steady
state value, you take the reading or take the action, open the control bar or close the
control bar. For many industrial process please note that the 5% of the final steady state
value is quite good for the control. Ultimately this temperature control or your flow
control or pressure control will be reflected in the quality of the product which you are
getting. In the case of fertilizer plant, it is the quality of fertilizer which is coming out, in
the case of steel plant the quality of steel it is coming out. So, it hardly matters or I mean
all the management people will not accept that if you make some sensors till it will reach
the final steady state value. I can take the action if my, if my quality of the product does
not change even if when it reaches 5% of the final steady, because I know what is my
steady state value. So, if it reaches 5% of the steady state value, I can take my action,
right, because ultimately it is the quality.
But, in some process we will find, especially in the bioprocess we will find, the precision,
the temperature should be very, very tightly controlled. So, 5% of tolerances cannot be
allowed there, because cell will die in that cases. So, temperature can be, suppose it is 31
degree centigrade, I have to maintain exactly 31 degree centigrade or it might be only
22
30.5 to 31.5. If the temperature goes beyond that range, cell will die. In that situations, I
cannot allow this 5% of tolerance band. But, in many practical process or industrial
processes like plants, these fertilizers, petrochemicals, steel, we can allow this 5%. That
means what I am saying that within when it reaches 5%, suppose this is our 5% of the
final steady state value, you take action. That means you take the reading, sent to the
control valve and whether it is to be opened or closed you take the decision or stop it,
whatever it may be, whatever may be the case.
Now, if I look at the ramp input, ramp input we have seen; characteristics equations of a
second order system with ramp input can be expressed as follows. Already we have
discussed the ramp input in the case of first order system. We are taking same three
inputs, so that we can compare putting these two side by side. D square upon omega n
square plus 2 xi D by omega n plus 1 y equal to K x s dot into t. y equal to dy by dt equal
to zero at t equal zero. This is the initial condition; we have discussed this before also.
What does it mean? The x s dot, my input, it looks like this. This is our ramp input. We
have discussed thoroughly this when we discussed the first order system, right?
23
Now, this we can, I mean do it in the case of second order system. So, this is my input,
this constant K as it happened before also, x s dot into t, right. It is continuously changing
with time, so x s dot, so I made it unit less quantity. So, these are our initial conditions
.
The solutions are found to be for the case 1 when xi is greater than 1, is over damped
system.
24
So, this is y by K. I have not put x s on the left hand side as it happened in the case of, I
mean, the step input we have seen that we have done like this - y by K x s. These we have
not done. The reason is that we have not done because of the fact that we will find there
this is unit less quantity. I have to make it unit less quantity. If I put x s dot here, it won’t
be unit less, because any value when it is normalized is supposed to be unit less. So, we
put x s dot t on the right hand side. So, we put just y if it is output divided by the, divided
by the static sensitivity on the left hand side and other things on the right hand side. So,
this is, as you can see x s dot t 2 xi x s dot upon omega n 1 plus in the bracket 1 plus 2 xi
square minus 1 minus 2 xi square root xi square minus 1 upon 4 xi under the square root
xi square minus 1 multiplied by e minus xi plus root over xi square minus 1 into omega n
t plus minus 2 xi square plus 1 minus 2 xi under the square root xi square minus 1 upon 4
xi square root of xi square minus 1 e to the power minus xi minus square root xi square
minus 1 into omega n t, fine.
Now case 2, as it happened before also it is critically damped system, when xi equal to 1.
In that case the output will look like this. y by K x s dot into t minus 2 x s dot by omega n
1 minus e minus omega n t 1 plus omega n t. You can see here these are all unit less
because x s dot by t, omega n is also radiance per, I mean per time, so these all becoming
25
unit less. Nowhere you will find that these, all these terms you can see this is, this steady
state, this transient term will die out of the factors, of the minus factors there; here also
for this, whereas in the case of, in the case of under damped systems you will find it is
something different.
Now, case 3 is a under damped system that means xi is greater than zero but less than 1,
as it happened before also, which is more sought after systems, because always we make
the system the under damped systems. So, it is y by K x s dot into t minus 2 xi x s dot by
omega n in the bracket 1 minus e to the power xi omega n t upon 2 xi under the square
root 1 minus xi square sine under the square root 1 minus xi square omega n t plus phi.
You can see here that it is, xi is 1 minus xi square because xi is now less than 1,
otherwise it will be an imaginary quantity.
26
The phase shift, I got a phase shift here and the phase shift is given by tan inverse 2 xi
root over 1 minus xi square upon 2 xi square minus 1.
Now, ramp response of the second order instrument, there is an animation, let it come.
You can see here that this is my input. This is input; it is going x equal to x s dot
and we have plotted for different values of xi. This over damped system is taking large
time, so the error is also large and this is for critically damped systems, we can see here
27
and this is for under damped system. We have put xi equal to .5. So, there is a oscillation.
Quite obviously there is sine term, sinusoidal term, there is oscillation. After that this
will, it will die out.
Now, this is our steady state error. The, we have plotted the, we have shown the steady
state error for an under damped system, which is 2 xi x s dot upon omega n. This is the
steady state error of the system, right and a time lag of the system that also we have
shown for the under damped systems 2 xi by omega n. This is our time lag for the system,
this is for this one. So, in the case of over damped system, so this will be our error and
our time, our time lag will be this much in the case of over damped systems, right. So,
obviously we can see that if it is under damped, after few oscillations it will come, it will
start to follow the input. Actually, in fact, this and this should be parallel to each other
and asymptotically these, all these two also should be parallel to each other, because at
the steady state response all other, I mean, all the transient part will die out, only the
particular integral will remain. So, this is our, the response of a second order system for a
ramp input. Ramp input, I mean ramp input you will not see that we already discussed
also in the case that usually the ramp input will not come.
28
You will not face the ramp; unlike electrical systems, always you will keep ramp input
like saw tooth waveforms. Always we know that it is a ramp input which we give in the
oscilloscope to synchronize, because in the two vertical plates we give our voltage which
we want to measure and to make the signal stationary we give the saw tooth waveforms
on the two horizontal plates. So, the signal appears to be stationary on the screen,
whereas in this case, the ramp input has not much of significance in the instrumentation,
except in some cases, where because in many of the cases you see, because of the large
inertias and the spring constant in many of the times you will find that even if you are
considering the input as a step input, so it won’t be a step input to the systems; it will be a
ramp input or terminated ramp input. That means it looks like that I am giving a step
input to the system, but the system, to the system it will not appear as a step input, it will
appear as a terminated ramp input. This is a very good example.
Now, you see that all of, all of you must be knowing of the load cell. Load cell we will
discuss later on in our, of this, of this course. Might be, I don’t know what is the exact
number of the lesson, but we will cover this in details, these load cells. Those who move
in highways also you see the load cell also there, because there is a weigh bridge, because
on the states the restrictions of that, the how much the load they will carry in the
highway; there is some state restrictions and highway restriction that you cannot. So,
usually they want to see the some certificate from the way side road bridge, I mean,
weigh bridges where they will and weigh bridge is nothing but the load cell.
Now, when you put on the load cell some weight, so it will never appear as a, even
though it is supposed to be a step input it won’t appear as a step input, it will work as a
ramp input, even though terminated ramp input, rather. Even though we are not much
concerned with this transient there, we are more concerned with the steady state response,
but the transient is important in the sense that we must know that how long it will take or
when it will die out, the system, this transient response will die out. For that reasons we
must analyze the transient response of the system also.
29
Now, last one is the sinusoidal input. That means the, you are considering the sinusoidal
input to a system. The sinusoidal input is equal to x equal to x s sin omega t, say, all of us
know from the first year classes how the sinusoidal input. So, this is my sinusoidal input,
so this is our x s and this is our x, this is omega t, right and y by K the output, normalized
output, rather y by K by x s equal to sin omega t plus phi 1 minus omega by omega n
square upon 2 square plus 4 upon square, sorry, whole xi square omega square upon
omega n square. Here you will find that is in this case that this omega n plays very
important role in the case of sinusoidal input and the frequency is very important,
because any system we must know that what is the frequency limit of the system? That
means up to what frequency I can operate our, I mean in what frequency it will accept?
We cannot put like an anti-analyzing filter that type of things which we usually do in the
case of electrical signals, but here always we want that what should be the input? We
must determine how much is the input in the system, what should the frequency or the
highest frequency of the input to the system which can be tolerated?
So, this is our normalized output of the system. So, in fact what will happen if you take
the amplitude? It will be y by K by x s. It will be y by K by x s. Only this part will be the
amplitude output.
30
So, if you look at the normalized output y by K x s, so this will be 1 upon, this will be the
amplitude output, because this will not go, right. So, may be you will find that in futures
and while when you solve the problems, so I need this type of, these values. So, you must
remember or you must go through the, this notes while you are solving the problems.
31
Now, phase shift in this case, in the case of sinusoidal input is tan inverse 2 xi upon
omega, upon omega n minus omega n by omega. Phase shift is also important, because
any sinusoidal input if you are measuring some sinusoidal signal how much the phase
shift of the system you must know. We will see the, how the phase response will look
like.
Now, if it is the magnitude response of a first order, I mean second order system for
sinusoidal input, it looks like this. Now, in sinusoidal input even though it is very, it is
not very common in nature, but you will find the sinusoidal input in many cases. In
vibration analysis we will find that we are giving there periodic input, not necessarily
always sinusoidal, we are giving some periodic input. In biomedical applications you will
find also sinusoidal input. Suppose the variations of the temperature of a patient which is,
who is suffering from malaria parasites and it is almost the sinusoidal, even though time
period is quite large for 24 hours, you will find that particular temperature will arrive,
when temperature will get at a particular time of the day, so that is also sinusoidal input
and some other more applications we will find in the sinusoidal; some ambient
temperature variation that is also sinusoidal in nature.
32
So, in the daytime and some temperature night, some other temperature in the morning
some other temperature, like this, so it is sinusoidal input. So, it is not very uncommon,
but even though the period in most of the cases is very, very large. Unlike electrical
systems, any physical systems as mechanical system, has a very large time constants of
the sine …. I mean sinusoidal input. Now, if you look at here, this is the magnitude
response of our sinusoidal input.
We have plotted here y by K by x s, so maximum is 1. You see here interestingly there is

large overshoot here. This happens because of the resonance. So, you have to avoid this
thing, because this is important. That is the reason you see we have plotted here in this
case omega by omega n on the x axis side and we have normalized the x axis also and
you see that here this is the one that means here is the omega equal to omega n.
Interestingly when xi equal to zero that means damping coefficient is zero, it is very
impractical thing. I mean you cannot make a damping coefficient zero. For academic
interest we should consider these things.
You so see that when the damping coefficient is zero, the magnitude becoming infinite
asymptotically infinite, infinite cannot be, because it has some supply. Suppose if we say
33
that if I using an op amp, so you cannot get infinite value. See, it will be limited by the
nonlinearity that is the power, plus minus power supply of the op amp and if I slowly
increase the value of xi from zero to .1, my hump is getting reduced. This is the xi equal
to .1. You see, I am getting a hump like this one, then it is coming back. It is xi equal to
.7, we are interested in xi equal to .7. We have seen, we told you that in the case of
second order systems any instruments we make usually .65 to .7, so I make a xi equal to
like this one, so there is one overshoot, then it comes back and when it is over damped
systems, xi equal to 2, so it will look like this one.
Now, see here I have not plotted for xi, the critically damped system. It does not matter,
because it will be the response for xi equal to 1. It will be, the response should look like
this one, something like this and this roll off, you will find always minus 40 dB per
decade for the second order system. In the first order system it will be minus 20 dB per
decade, it is 40 dB per decade. So, what does it mean? Actually what is the significance
of this curve? The significance of this curve it tells that what is the frequency limit of
your instrument of a second order instruments?
Suppose in the case of over damped systems you can see that there is a large attenuation
if you go far off from the, from the origin. Whereas, in the case of if it is xi equal to .7,
you will find there is a very little, there is a single over shoot, then it comes back, so that
almost horizontal or flat frequency response I am getting. We are not supposed to use it
after the omega n, because at the omega n suppose if it is, xi is .1, there is a large
overshoot. In many instruments it is not allowed, it may damage the instrument. So, I
will, if I know the omega n, so I can limit my frequency response of my instrument. That
is the reason it is necessary we must discuss this point, right.
34
So, as you can see, this way I can see there, let me see, so here I can, I can tell that here it
will look like this that the xi decreases means response or here in this case xi increases, so
I will get a response like this one. So, I am more interested at omega equal to omega n.
If I look at the phase response of a second order system for sinusoidal input, let it come
all the animations, it will take some time, hopefully it is completed, no; you see here that,
35
here that if xi equal to zero that means the damping coefficient is zero. We have already
discussed also when the damping coefficient is zero how my magnitude response will
look like. My response will go like this, then go, go like this one, total discontinuous
curve, it will go like this, come like this; let me go ahead, take it, again it will go like this,
xi equal to zero, come like this and response will be this - from zero degree, this is zero
degree to minus 180 degree it will follow and if I slowly increase the value of xi, suppose
xi I made .1 my response will start to look like this one.
Similarly, if xi equal to 1, it is critically damped systems, so response will look like this,
response will look like this and if xi equal to over damped systems, our response will go
like this. How? This will go like this. Interestingly you see, for, this is omega equal to
omega n, it will, it does not matter what is the value of xi, it will always pass through
minus 90 degree at omega equal to omega n.
You will be more clear if you look back, if I go back to our, sorry, you see here, you see
here, what happen that omega equal to omega, I mean when phi equal to tan inverse 2 xi
upon omega by omega n minus omega n by omega. You can see here that when omega
equal to omega n, always it is minus 90 degree. That …. depends on the value of xi.
36
Mathematical fallacy will come, all these things, and what will be zero by zero and all
this thing, I am not going to, I mean go into details of that thing. Let us go back; already
we have seen this thing.
Now, two more terms we will introduce. This is called dynamic error of the system,
because we need this while solving the problems. In our future slides we will see that we
will solve some problems - first order, second order instruments. Dynamic error, d of a
system can be defined as d e equal y by K by x minus 1 and it represents a measure of the
inability of a system to adequately reproduce the amplitude of the input signal for a
particular input frequency, right. So, its inability that means if you have to look at that
what should be the error? Obviously, it depends on time.
If the time goes by, then you will find the error will be getting reduced and reduced, but
after certain time if I say that how much is the dynamic error, because this will be utilized
to solve some problem, you will find that is very important definitions for any physical
systems, a measurement system with a magnitude ratio close to unity over the anticipated
frequency band of the input signals is preferred to minimize this. Obviously, we will
37
always make, always try to make d e equal to zero, but that we cannot, so we will take
some minimum value of this one.
Now, settling time is another important, the characteristics that is used in characterizing
the speed of response of an instrument is the settling time. How does it mean? It looks
like, you see here, I am sorry, settling time looks like this, I have this one, I am not very, I
should draw it very nicely. We have discussed this before also, suppose the 5% of the
value, so this is our value t 1, so this is the settling time. If you take infinite settling that
does not make any sense that means after the 5%, when it reaches the 5% of the tolerance
band whatever the time required that is we call it settling time.
38
This is the time after application of a step input for the instrument to reach and stay
within a tolerance band around its final value.
Now, let us start some problem. So, let us look at the problem 3.1. It says, a first order
temperature transducer is used to measure the temperature of the oil bath and if the
temperature goes above 100 degree centigrade, heat supplied to the oil bath should be
39
stopped within 5 seconds after reaching 100 degree centigrade. Determine the maximum
allowable time constant of the sensor, if the measurement error of 5% is allowed, right?
So, let us solve this problem.
I do not know, yes, it has chosen properly. Problem looks like this, so here, let us go
back. So, it looks like this. That measurement error, error will be equal to e to the power
minus t by tau. Now, for the given problem, problem e to the power minus t by tau equal
to .05, for t less than equal to 5 second, tau will become less than equal to 1.66 second
that means the time constants of the system should be less than equal to 1.66 seconds,
right? If it, the time constant system is less than equal to 1.66 seconds, so the 5%
allowed, I mean after 5 seconds, 5% of tolerance band can be allowed or 5% for the
dynamic error can be allowed, right?
Now, let us go to problem number 3.2.
40
It tells, I will solve here from problem number 3.1to 3.3 and I will give the problem 3.4 to
3.6 and will explain to some extent 3.4 to 3.6, but the solution will be given at the
beginning of the lesson 4. So, I will give some hints, so you are supposed to solve it of
your own. So, the problem number 2 or 3.2 of the lesson 2, lesson 3 will tell that, like this
I mean a first order instrument has a time constant of .5 second, it is measuring a process
parameter that is sinusoidal in nature and having a frequency of 3 hertz. Determine the
dynamic error of the system, fine.
So, the system is known, it is the first order system, time constant is also known, but the
signal is now sinusoidal in nature, instead of step and all those things; frequency is given,
so you have to find the dynamic error of the system. So, there will be a constant dynamic
error in the first order, because since there is a phase shift, so there is a constant dynamic
error in the first order system. Here it looks like this.
41
You see here that in the case of first order systems, I can write the dynamic error equal to
d e equal to y by K by x minus 1. Here, y by K by x equal to 1 upon root over 1 plus
omega tau whole square. Here omega is given. Omega is 2 pi 3 and tau is equal to .5.
So, y by K by x will be equal to 1 upon root over 1 plus 2 pi 3 into .5 whole square,
which is coming, I mean showing with the complete calculations, which I have
calculated, root over 89.7364, which is equal to 1 upon 9.472, which is 0.105, clear?
42
So, the dynamic error, error equal to .105 minus 1 equal to 0.89 or minus 89%. That is
the error, right? This is your answer.
Now, let us go to the problem number 3, which is telling that the first order temperature
sensor has been suddenly immersed in a liquid that has a temperature of 100 degree
centigrade and if after 3 seconds the sensor shows a temperature of 80 degree centigrade,
calculate the instrument time constant. Also calculate the error in the temperature reading
after 2 seconds.
43
So, here you see that y equal to, let me take the white page, so y equal to 100 1 minus e to
the power t by tau. So, t equal to 3 second and y t equal to 3 second equal to 80 degree
centigrade. So, 80 equal to 100 1 minus e to the power minus 3 by tau. So, after solving, I
can show you the detailed calculations, so it is coming .8 equal to 1 minus e to the power
3 by tau. So, it is coming around 3 by tau or 3 by tau 2.303 into log of 5. So, it is coming
around tau equal to 1.86 second.
44
So, now if I put, this is a time constant of the system, now if i put y equal to 2 second to
100 1 minus e to the power minus 2 by 1.86, so equal to 100 1 minus e to the power
minus 1.0732, so it is equal to 100 1 minus 0.341 is equal to 65.8 degree centigrade. So,
this is the temperatures and at after 2 seconds, so the error will be, you subtract it from
100 or 100 should be subtracted from this. It is minus 34.2 degree centigrade. This is the
error, right.
Now, problem 3, I will not make the solution to the problem number 3, which you have
to do.
45
So, a pressure transducer is to be selected to measure the pressure of a vessel and the
pressure variations can be considered as a sinusoidal signal or the frequency lies between
1 and 4 hertz. Several sensors are available, each with known time constant. Select a
sensor if the dynamic error of plus minus 1% is acceptable. Now, you see, it is first order
instrument. So, there is no question of taking the plus 1%, because always it will be, it
will go below, if you remember. So, it is the first order systems always, the sorry, so it is
a first order system. It can never over shoot like this one. It will always go, so it is,
always go up to minus 1% you should take and you have to find the time constant of the
system that is your question or what is the range of the time constant, rather? That means
that it will, because it is 1 hertz to 4 hertz, so this is 1 hertz to 4 hertz, so within the, if it
lies between this, what should be the range of your time constant. This I will solve in the
next class.
46
Now, let us look at 3.5. The periodic signal is to be measured with a first order instrument
having a time constant of 3 second. If the dynamic error of plus minus 5% can be
tolerated, find the highest frequency of input signal that can be measured by the
instrument. So, in this case, you must see that this is a, basically we have considered this
as a first order instrument and you have to find the value of omega that means the
frequency range for which the instrument can be used, right, highest frequency.
47
Now, the last problem, you will find a second order system. So, there is no question, so if
it is second order system, you have to take both plus and minus of the dynamic error that
is the reason we have given. So, you have to take both. That means you have to take both
plus and minus of 6%, so accordingly you will keep the value of the omega n. So, the
natural frequency of the sensors you have to find here, you have to find the value of
omega n here. So, you try to solve these problems and the solutions will be given in the
next class and I will remind you that here you see that the problem number 3.4 the
instrument is first order, problem number 3.5 also it is the first order system, whereas
problem number 3.6 is a second order system. So, with this I come to the end of the
lesson 3.
48
Dr. Alok Barua
Lecture - 4
Strain Gauge
Welcome to the lesson 4 of industrial instrumentation. In this lesson, I will cover strain
gauge - one of the very, I mean important sensor in instrumentation. However, before,
before I start the, this lesson I must solve because as I promised that at the end of the
lesson 3 that I have given the lesson 3, I mean problem 3.4 to 3.6, which I will solve in
this class.
Now, problem number 3.4 as you know is the magnitude of the dynamic error. Dynamic
error is or I should write d e, mod of d e should be less than equal to 0.01, because you
know that it is a first order instrument, even though I told that the error can be plus minus
1%, but in the case of first order instruments we have seen that there is no over shoot, so
error should be always less than, can only have, it should be less than that means it
should be always .99 instead of 1.01. So, if it is there, so the magnitude ratio we know
equal to y by K by x. So, it should be 0.99, because .01%, 1%, so it is 0.01 should be less
1
than equal to y by K by x, right and as you know this is equal to 1 upon under square root
1 plus omega tau square.
Now, omega is given, let us calculate for omega equal to 4. So, if I calculate for omega, I
mean frequency of 4 Hertz, so it will be 2 pi into 4, because f equal to 4 Hertz, as it is
given. So, it will be, omega square will be equal to 631. So, it will be 0.981 into 1 plus
omega square tau square equal to 1. So, it will give you omega square tau square equal to
0.01936. So, tau will be equal to under the square root 0.01936 divided by omega square
is 631, so it will give you tau equal to 5.6 milli second.
2
Similarly, if we take f equal to 1 Hertz, because we say that the frequency will lie
between 1 hertz and 4 hertz, so we will get tau equal to 0.0226 second. So, I should say
that the tau should lie between 5.6 milli seconds and 0.0226 second that means 5.6 milli
second, tau 0.026 seconds. If the tau lies between this, the time constant lies between
these values, so I can handle the signal for the first order instrument of 4 Hertz to, 1 Hertz
to 4 Hertz.
Now, let us solve the problem number 3.5.
3
3.5, you see we have to find the, the highest frequency, right? That was the problem. It is
also a first order instrument. So, problem number 5, solution to problem 3.5, so it will
look like, d e equal y by K by x minus 1. It is here, the error is plus minus 5%. So
obviously, so it will be .95 less than equal to 1 upon root over 1 omega tau square, sorry
that was the given. So, I can write that, from this omega tau square will be less than equal
to 0.108. So, I can write omega will be less than equal to 0.01095 radian per second. So, I
have not solved the intermediate step; that you can do. But, this is our principle, how we
can find the value? This is our answer. So, if omega is less than that, I mean that means it
can handle the frequency which has a circular frequency value of 0.01, .01095 radians per
second.
4
Now, problem number 6, 3.6 rather, so the solution to problem 3.6, it looks we have a
dynamic error of plus minus 6%, right? We are given, error constant zi you are given .6
and since it is a first order instrument, so it will lie between, sorry it is a second order
instrument, so it will lie between .94, so it can over shoot. So, the plus minus, both plus
minus 6% will be valid here. So, y by K by x equal to 1.06, right. So, we have two
equation for this that means 1.06.
We know in this case 1.06 greater than equal to 1 upon, if I take 1 minus omega by
omega n whole square square plus 2 zi omega by omega n whole square square half. No,
I am sorry; I think there will be no square here. So, this will be, so it will get, this is 1.
5
And another one will be .94. It is .94 greater less than equal to 1 minus omega by omega
n whole square square plus 2 zeta two zi omega by omega n whole square to the power
half. See, if you solve these two, so I will get omega n greater than equal to 730 radians
per second. This is our answer, right? Now, let me go back to the, our today’s lecture.
It is strain gauge.
6
Now contents, as I told solution to the problems of 3.4 to 3.6 of lesson 3, application of
strain gauge, derivations of gauge factor, composition of the strain gauge material,
bonded metal foil gauge, gauge length, backing material, adhesives, semiconductor strain
gauge, strain gauge in a Wheatstone bridge and temperature compensation scheme and
problem on strain gauge. We will solve some problem later on also.
Now, you see that strain gauge is basically a device, is a, when it is subjected to some,
subjected to some force that means if there is a stress there will be a change of strain.
That means what will happen that if there is a change of resistance? So, that change of
resistance can be calibrated, that or measured and can be, in future can be calibrated in
terms of, in terms of either load or displacement.
Now, directly this strain gauge is used for the measurement of load; indirectly it is used
for measurement of displacement, small displacement as well as for measurement of
pressure. We will see later on that it is used for the measurement of the pressure, when it
is used in conjunction with the diaphragm gauge, right? Now, let us look at the, so that is
the basic application of strain gauge. Now, let us look at the derivations of the gauge
factor of the strain gauge.
7
Now, I mean, it is the, at the end of the lesson the viewer will know the derivation of the
gauge factor of the strain gauge, strain gauge composition and temperature compensation.
Let us take a wire of length L and cross sectional area A and if this wire is stretched or
compressed, its resistance will change due to dimensional change and because of the
property of the materials called Piezo resistance which indicates a dependence of
resistivity on the strain.
8
Now, resistance R is given by, as you know this is very common equations, rho L by R
equal to rho L by A, where rho is the resistivity of the material of the wire, L is the length
of the wire and A is the area of cross section of the wire, right and if I take differentiation
of this, so dR equal to A rho dL plus L drho minus rho L dA upon A square; all are
variable, so in this case we will get equation like this.
9
Now, see the axial strain we defined as this one that means axial strain we have defined
epsilon a equal to epsilon into dL by L. The transverse strain we have defined as epsilon t
equal to dD by D, which if you take derivative it will be 1 by 2 or half equal to dA by A,
where A is the area of the, area of the cross section of the strain gauge. Now, Poisson’s
ratio is defined by, you see, nu equal to minus epsilon t by epsilon, you know or epsilon t
by epsilon, simple epsilon. Again the volume of the wire V equal to AL. As you know V
is the volume, so it is A multiplied by the length of the wire. If you differentiate we will
get dV A dL plus L dA.
Moreover, we can prove that del V equal to, almost equal to AL epsilon 1 minus 2 nu.
Now, I will combine these two, I mean, so I will get after combining AL and epsilon,
after combining AL and epsilon we got dL, so 1 minus 2 nu. So, this can be further
simplified. Again we know that del V equal to A dL plus L dA, so A dL 1 minus 2 nu
equal to A dL plus L dA, so minus 2 nu A dL equal to L dA.
10
Now, dR equal A rho dL plus L drho minus rho L into dA upon A square. So, it will
further be simplified like this, right, because we are using the Poisson’s ratio, so we make
it positive. So, I will get the equation, the change of resistance equal to dR rho dL; rho is
the resistivity of the material of the strain gauge, multiplied by dL 1 plus 2 nu, where nu
is the Poisson’s ratio divided by AL drho by A. It can be further simplified, if I write like
this, dL equal to, because as you know that R equal to, because you know that, we know
that R equal to rho L by A, so we have applied this here. So, it will be dR into R, sorry
rho dL equal to, then R if you replace rho then it will be R into A. A, A will cancel out,
so we take the L in the denominator and R on the left hand side. So, change of resistance
per unit value of the resistance that we have actually done.
11
So, the gauge factor is given by, it is usually defined as lambda, is very standard notation
for the gauge factor, into dR by R upon dL by L equal to 1 plus 2 nu plus drho by rho
upon dL by L. Now, you see this, this one is actually resistance change due to change of
length. This is the resistance change due to change of area. Typically its value lies
between zero to .5 for all materials and finally the, that is the resistance change due to,
due to the effect which is called Piezo resistance effect, right?
Now, in the case of semiconductor this will be very high, whereas in the case of metal, I
mean this is predominant. In the case of metal you will find 1, whereas in the case of
semiconductor, it is, this will be predominant. You can almost neglect these two parts,
because it is so high, right?
12
Now, further we simplified. We have introduced drho by rho upon dL by L equal to pi 1

into E, where pi 1, where E is the modulus of elasticity and pi 1 is the, capital pi 1 is the
longitudinal Piezo resistance coefficients, which can be positive or negative, right. So,
with this that gauge factor, expression of gauge factors we have seen. Now, we always
want, for in the instrumentation systems or for the strain gauge that the gauge factor
should be as high as possible, because if the gauge factor of high, then for unit strain I
will get more change of resistance, my signal conditioning circuit will be simplified. That
is requirement for all sensors, right.
Now, let us go to the instrumentation lab and see one practical strain gauge, how it looks
like and which is used as a displacement sensor to make a small, I mean displacement
sensor that means for the small measurement of small displacement. We will come back
after sometime.
13
It is a displacement sensor based on strain gauges. We have already seen that in all strain
gauge measurements we usually, we use a bridge, Wheatstone bridge with four arms and
for temperature compensation as well as for the larger output, we usually use at least two
strain gauges and sometimes you will find there are four strain gauges. In this
displacement sensor, we have two strain gauges and one will be on axial elongation, other
will be on axial compression.
14
And if we put this two strain gauges on the opposite arm of the bridge I will get larger
output.
You see here that there is a strain gauge on this side; you can see here, so and there is
another strain gauge on the backside of this brass plate and how it works, you see, when I
give the displacement like this, you see here, the displacements I am giving here. So, this
strain gauge will be under axial elongation and the strain gauge on the back side that
means on this side will be under compression. So, if I put on the opposite arm of the
Wheatstone bridge, obviously I will get double the output as well as you will find that I
will receive the, I will get the temperature compensation in this particular circuit and this
particular displacement sensor is used for the displacement of, small displacement like
less than 1 centimeter and that type.
Welcome back to the class room. You know the strain gauge is implemented in several
different ways.
15
Number 1 is unbonded metal wire gauge, number 2 is a bonded metal wire gauge and
number 3 is a bonded metal foil gauge and bonded, number 4 is bonded semiconductor
gauge and 5 is diffused semiconductor gauge. Now, in the case of bonded metal wire
gauge, I can have any unbonded metal wire.
16
Suppose I have a two pillars, sorry suppose I have a two pillar that means two pillars
here. So, I have put a wire, so if there is, if this, if this pillar is tilted like this one, so some
stress will be developed, because this wire will be under tension. In that case, so the
stress will be developed. So, it is the unbonded strain, because any wire, a simple wire if
it is under, if you pull on both side, so it will have certain amount of stress and strain,
right? So, this type of ….. measurement was, I mean people used sometime back, so we
can call, I mean it is some sort of academic interest, because unbonded wire strain gauge
is very uncommon nowadays.
Now, next we will find the most widely used gauges, bonded metal wire gauge, right and
here in this case that the strain gauge will be bonded on a backing material. It looks like
that means it looks like I have a backing material, so over which this strain gauge is put
like this one, right. So, I have a backing material there. So, we have electrical connections
we have taken out. So this is a, backing materials are various types which will be
discussed after sometime. So, this is the strain gauge which is called bonded strain gauge.
So, it is actually with the adhesives, so it is to be, this is our vacuum material and it is to
be placed on a body on which we are interested to measure the stress or strain, right?
17
Third comes the bonded metal wire gauge which is most widely used metal wire gauge
nowadays. Because we will concentrate on this lot, because it has some advantage. We
will see that the transfer strain will be very less in this case, because the same technology,
whatever they are using for making the PCB is also used for making the bonded metal
wire, metal foil gauge and interestingly you will find, you will find the shape is
something like that that this end resistance effect will be very much less in the case of
bonded metal wire gauge.
Now, bonded semiconductor gauge is also there. You will find the semiconductor itself is
used to bond on a material. So, it is a, instead of I mean simple wire type of gauge I am
using the semiconductor strain, whereas the diffused semiconductor is something
different. You will find that in the, one of the example of the diffused semiconductor
gauge is the, is a pressure gauge, diaphragm pressure gauge, where the diaphragm
material is of, made of semiconductor. It looks like this.
Suppose I have a diaphragm, circular diaphragm, you can see here. So, I have 4 strain
gauges here, suppose two on the, so this material is semiconductor and where I want to
make the strain gauges that portions will be doped. So, if I dupe, I mean dope like this
18
that portions, so I will get some gauge there and this type of strain gauges is called, is
called the diffused semiconductor gauge. It is used for, it is a very, because of the
advantage of a semiconductor strain gauge, as I told you earlier, is a high gauge factor
whereas in this, almost you can say some 60 to 70 times the gauge factor is more than
that of the wire type of gauge or gauge made of metal, right?
So, composition of strain gauge material that is very important. You will find that there
are many strain gauge are available in the markets and you will find here the different
gauge factors we are given. Lambda equal to 2 is a very common type. It will be cheap,
low cost strain gauge is called constantan or advance. Here, the composition is 55%
copper and 45% nickel. Then you have isoelastic, which is 36% nickel, 8% chromium,
4% manganese, silicon and 52% iron.
Then, you have isoelastic 2. These are all trade name, because you see the strain gauge
was first developed, I mean commercial strain gauges which are developed mostly in the
industry, so they have given some trade names. Later on they gave, gave the composition,
but that is the reason the funny name, they have given constantan, isoelastic, like that.
19
Then, we have molybdenum, isoelastic 2, which is 36% nickel, 8% chromium, .5%
molybdenum and 55.5% iron or steel. So, lambda is 3.6, not very high.
Karma, which is lambda equal to 2, because you see, what is the use of this 2? You will
find the applications will be different if you have that 74% nickel or 20%, because strain
gauges are used for various environments. Somewhere you will find the reducing
atmosphere, somewhere it is oxidizing atmosphere, somewhere it is very corrosive
atmosphere. So, in all these, I mean, features you will find that, in all these different
applications you will find that you have to use different type of strain gauges.
Then, we have a armour D, which is 70% iron, 20% chromium, 10% aluminum; lambda
is same. Platinum - tungsten rather I should say it is a very good gauge factor. It is 92%
platinum, obviously quite costly, 8% tungsten and lambda equal to 4. Gauge factor is 4,
because of the gauge is higher, gauge factor is higher, obviously I will get the better and
better output. Because, as you know that lambda means dR by R by dl by L, so whenever
the lambda is high, so I will get more change of resistance, right? For the same amount of
strain, I will get more change of resistance. So, that is very much necessary, okay, to have
a, simplify our signal conditions.
Then, we have a Nichrome V 80% nickel, 20% chromium, lambda equal to 2.1.
Semiconductor strain gauge i s lambda equal to 130, right? This should be equal, I do not
know why it did not come, so lambda equal to 130, exceptionally high. You see that
compared to all these this is almost 60 to 70 times more, at least 60 times more; 62, 16, I
mean 5 times more than the conventional wire or please you note one thing that whether
it is a wire type of gauge or a, I mean bonded wire gauge or bonded metal foil gauge, the
value of lambda does not change. It is a physical property; it depends on the compositions
of the materials.
People for the, over the years they have tested this; you have, they have seen, they made
some funny composition to get the value of lambda higher and higher. Whether you are
using in a bonded metal gauge or unbonded gauge, bonded gauge or bonded metal foil
20
gauge, it does not matter, because ultimately the value of lambda depends on the
composition of the material which is used to make the strain gauge.
So, bonded metal foil gauge as I told you earlier, it is the most widely used transducer for
stress analysis. It consists of a metallic foil pattern similar to the process used to produce
printed circuit board. So, the as you know, the printed circuit board, because it is printed
circuit board what happens that as you know that it is a bakelite type of material is taken
and it is covered with copper. Then, a mask is put on this one and it is a, it is ultraviolet
light falls on that and whenever the, that falls whenever it is obstacle, so that portions
when you put later on in the liquid, some special liquid you will find that that will go out.
The same principle is also used for making the strain gauges, the reason I will tell you
after sometime. So, photo etched metal foil pattern is installed on a backing material.
21
You see, here it is a typical metal foil gauge where in the case of wire gauge this two
thickness will be same, whereas you see here all the end resistance or end portions of the
strain gauges are made very thick. The reason is, you see, when I am subjected to some
stress, so I am supposed to, that suppose in this case suppose I am giving it some tension
here. So, all these supposed to have an axial tension, is not it? Whereas, what will happen
you see here on these portions? These will have a transverse, right? But, while I am
making the calculations I have not included this part. So, what I have to do?
I have to do in such a way if I make it thick, so the contributions, so the value of R in

these portions will be very, very small, is not it, compared to this portion. So, the
contribution to the transverse strain or the transverses strain error in this case will be
very, very low. That is the reason metal foil gauge purpose, because in the case of wire
gauge I cannot change the dimension of the wire in between, but in the case of metal foil
gauge we can do it, because it is the principles of the, principle of the printed circuit
board is utilized to make this type of gauges. So, this end portions are very thick. So, the
resistance small, r small, here r small, so the error, transverse error will be minimum,
transverse strain error will be minimum.
22
You may immediately say, sir, what is the harm if I take a simple stretch of wire like this
one. This is not possible. You see that I have to make a sufficient length of the wire, so
that the value of r is sufficient. If the r is large delta R will also be large, please note.
So, that will ease our, simplify our measurement. If the r is small, r is small if I take
suppose a small length of wire, if I take a full length of wire, this one, so my delta R also
will be small, R will be large delta R also will be large. So, that is the reason all the funny
shape and this type of shape you have to make in the case of strain gauges. This is very
simplified.
Actually, you will find there are various ….. It looks like, I have several, you will find
that there are several that means, so that the value R should be that means substantial. So,
this value should be very substantial, so that the delta R also will be large, right? So, that
is very much necessary.
23
So, is a bonded metal strain gauge. You see this here, this is a backing material on which
we have put, right?
Now, gauge length, the proper choice of the gauge length is important factor for some
specific applications. Strain is to be measured at the location where stress is maximum
and the strain gauge averages this measured strain over the gauge length and the
24
measurement can be erroneous due to improper choice of the gauge length, right. So,
gauge length is very important.
Backing material, it electrically isolates the metallic gauge from the test specimen, right,
because you see in all the cases you will find strain gauge we will put on a Wheatstone
bridge. So it is a electrical circuit, so that electrical suppose I am interested to measure
the, measure the strain or a load on a beam, steel beam. So, in that type of situations, so I
cannot connect directly on the steel beam itself. I have to insulate, because we are
interested only on the strain we have.
I do not want to make electrical connection with the, I mean, steel beam itself, right. So,
that is it will isolate, first of all it properties, electrically isolates the metallic gauge from
the test specimen Then, it transmits the, and applied strain to the sensor. It provides the
surface that is used for bonding the gauge with the specimen surface where the strain is to
be measured. The backing material should have a wide temperature range. Polyimide and
the glass reinforced phenolic are two commonly used backing material. These are the
most commonly used backing material.
25
You see, these are polyimide. So, these are polyimide and glass reinforced phenolic and
two commonly used backing material. Now adhesives, because you know that this
bonded metal gauge, I mean wire gauge or foil gauge it does not matter, it cannot, I mean
you cannot, you have to use some adhesives. So, usually you have to put some adhesives
on which and there is, at least there should be a some time of one 24 hours you have to
keep for the curing or drying up process, right?
So, there should be some features, as you know, in the case that while you are installing
on a, suppose a pipe, suppose I have pipe like this one, so I put on a strain gauge over this
one. So, I will put a strain gauge. I want to ….. suppose there is a bending of pipe, how
much strain will be developed in that pipe? So, in that case, so I have to put the strain
gauge and I have to, whatever the strain developed on this pipe, after the, one, once it is
under tension or compression same should be transmitted to the gauge itself, right?
So, there should be a very strong bonding of the, first of all the, gauge with the backing
material, then backing material with the, this test specimen, right? If there is a some gap
that means between the test specimens and the, and the gauge, so the actual load which is
I have given to this, this pipe or test specimen cannot be ….., will not be fully transmitted
26
to the gauge. So, there will be some error which I do not want, right? So, the bond created
by the adhesive serves as a mechanical and thermal coupling of the strain gauges and the
test specimen.
This is very important, because you know the, all the strain gauges since it is a resistance,
it is temperature sensitive. So, I have to make all sorts of temperature compensation and
whatever the temperature compensation is there, so I have to incorporate that, because
you see the, if the, I am not interested in the resistance change due to the change of
temperature. I am interested into the resistance change due to the change of the strain
there, because if it is, there is a resistance change due to the temperature that will be
erroneous reading, because I am interested, I am measuring in the test specimen how
much stress is developed, how much load is developed there. So, in that case, if there is a
some strain developed due to the change of resistance, so that will give us erroneous
reading.
Adhesives should accurately transmit the strain given to the test specimen. It should have
a thermal conduction and the expansion characteristics same as the metal itself or test
specimen. The adhesives should not shrink or expand during curing process otherwise
the pseudo strain will be developed in the gauge. This is also very important, because
while in the curing process means you have to leave for at least 24 days hours to dry up
or curing process. Once you put the, suppose in my pipes I put a strain gauge here, so I
put an adhesive. So, I have to leave it for 24 hours before drying.
Now, during drying up, so during drying, during curing process or dry up process there
should not be any pseudo strain developed. If it is developed, then what will happen?
You know that the, that resistance, strain gauge will be under tension. So, it will give you
some false reading. Even though you can compensate for that you can use some other
resistance to balance the bridge and all those things, but that is not very desirable, right?
The epoxies and cellulose nitrate cement and ceramic based cements are the, some of the
adhesives used for making the adhesives, I mean for adhesives.
27
Now, semiconductor strain gauge is most widely used, as I told you, because of its high
gauge factor. Only problem with the semiconductor strain gauge is that the, it is very
sensitive to temperature. That means it is, temperature coefficients are very high, so that
immediately it changes the value of those temperature, I mean resistance changes as the
temperature changes and also the, its performance deteriorates in the presence of the
moisture. That is main problem in the case of semiconductor strain gauge.
Silicon is a basic material for making the semiconductor strain gauge. Usually silicon is
doped with boron to make p-type strain gauge. It will be n-type strain gauge if it is doped
with arsenic and the resistance of p-type gauges increases with the applied tensile
strength and that of n-type gauges decreases, right? Now, only problem with the strain
gauge you see that, you will see later on, I have to use some dummy gauges for
temperature compensation. We will see later on this thing.
Now, getting two strain gauges, two semiconductor strain gauges of exactly same value
of the lambda is a very difficult, because see, see in the case of I mean metal gauges, the
value of the lambda depends on the composition. If the composition is correct it does not
matter whether you make it wire, I mean metal foil. So, you will get always the same
28
value of lambda, whereas in the case of, if the slight doping changes in the case of
semiconductor strain gauges its value of lambda changes. So, the two, exactly two gauges
having the same value of lambda is very difficult to get, like beta of a transistor. Two
beta is always, we say that there is some, it lies between some value, two values. I cannot
have a two transistor exactly same beta. I can measure it and make, make it. Suppose in
the complementary symmetry amplifier and push amplifier we always, we try to make
beta of the two transistor same. We pick up and measure. If two are equal we say these
are pair.
Now, resistivity of a semiconductor material is given by rho equal to 1 upon q n mu,

where, I am sorry, this will be where, so this will be where q equal to charge of electrons,
number of charge carriers that depends on the doping and mu is the mobility of charge
carriers, right?
29
The impurity concentration typically lies between two values which will come at the end.
The resistivity of the p-type silicon with concentration, let it come first, then you will,
you see here the impurity concentration typically lies between 10 to the power 16 to 10 to
the power 20 atoms. The resistivity of the p-type silicon with concentration of 10 to the
power 20 atoms per centimeter cube is 500 micro mho meter. That is much higher than
that of the copper. The strain gauge material shows a change.
Now, interestingly you will see that if you increase this beyond 10 to the power 20, you
won’t get any, much change of, much change of the resistivity in the case of, if you apply
some strain there, whereas if it lies between these two values, you will get sufficient
change of resistance in the case of semiconductors, I mean that you, actually we want.
The strain gauge material shows a change in resistivity with strain and this change of
resistivity is called the piezo resistive effect and the mobility of the charge carriers
changes due to applied strain thus causing the large change in resistivity. This is the most
important, you see, in the case of semiconductor strain gauge this change of resistivity is
very high that is the reason I am getting the large value.
30
If you remember our main expressions, so it, our expression was lambda, sorry lambda
equal to dR by R, excuse me, dL by L equal to 1 plus 2 nu plus d rho by rho dL by L. So,
this term for the semiconductor strain gauges will be very high because of this factor.
This term is not predominant in the case of metal wire gauges, right? So, that is I am
saying in the mobility of the charge carriers changes due to the applied strain, because the
mobility, it depends on the mobility, we have seen, right, so thus causing the large change
in resistivity.
31
The advantage you see, the high gauge, gauge factor useful for measurement of micro
strain, very small strain I can measure. It is not possible with the case of wire type of
gauges whether foil type or it does not matter, foil or wire type. In the case of metal
gauge it is not possible. Small gauge length, so very small strain gauge sensor we can
measure. So, the dimension of the gauge I can make very small. As I told you earlier, that
you see that in the case of pressure sensors we have seen that the, that small size I can
make which is not possible in the case of, in the case of wire type of gauges.
Now, disadvantage is high temperature sensitivity. Temperature sensitivity we can

compensate. It is very easy to say that is why we can compensate by using the dummy
gauges or using four strain gauges, but having four semiconductor strain gauges, having
the exact value of lambda is very difficulty. Nonlinearity, so nonlinearity is another factor
in this case of semiconductor gauge. Mounting difficulty is also there, because
semiconductor is very, I mean if it is diffused that is advantage, but if it is not diffused
bonded semiconductor strain gauge, bonded difficulties will be there and performance
deteriorates in the presence of the moisture and instrumentation, not necessarily you will
measure, I mean without moisture, I mean sometimes you will have to measure in some
moist where super heated steam is there. Sometimes you have to use corrosive liquids,
32
sometimes you have gas which is very hostile in nature, so in that type of situations
semiconductor strain gauge is not very suitable.
Now, see that I will make a temperature compensation. Strain gauge temperature
compensation is there, so it is necessary. I will do the temperature compensation; it looks
like this. I have an active gauge, one dummy gauge and fixed resistors. You see here, I
have an active gauge. You see that in the case of strain gauge, I can measure with a single
strain gauge also. I, whatever the change of resistance I will put on a Wheatstone bridge.
Suppose there is all the gauges resistance, simple resistance, I put one active gauge, in
that situation there is no problem.
I can measure that change of resistance values when it is subjected to strain, but the
problem is that you see it won’t, there, there will not be any temperature compensation.
That means due to temperature change some change of resistance will come, so my
circuit will not take care of that, whereas in the case of dummy gauge, dummy gauge the
meaning is that this gauge will not be subjected to stress or strain. That means suppose I
am interested to measure a, I mean, I mean measure the stress developed on this pipe, so I
33
will put the actual gauge or active gauge on this one and dummy gauge I will not put over
this one. It is also a strain gauge. These are not strain gauge.
These two are fixed resistors, simple carbon or any other type of resistor, but these are
basically the strain gauge. These two are strain gauge, whereas this gauge will be
subjected to strain and this will not be, I mean this will not be subjected to strain. So,
whereas this is subjected to strain, this will not be subjected to strain. So, what will
happen to this? This will be very close to the pipe, but it is not subjected to any strain. So
whatever, the reason is why I am saying that this should be very close to the pipe,
whatever the active gauge have the temperature this dummy gauge should have the same
temperature that I want to mean. If I can do that, so there is no problem. So, always we
will measure this, I mean this, this change of resistance by a Wheatstone bridge.
Now, there are two types of measurement. One case I will use a balance bridge, in
another case we will use an unbalanced bridge. Now, we will first consider the unbalance
bridge, let us look at.
34
So, in the unbalanced bridge we will find that, sorry it is a balanced bridge, so I have
four, so this is my output voltage. This is, let us take a Wheatstone bridge. So, otherwise I
cannot write. Suppose this is I 2, all current different, R 2, this is I 4, this is R 4, this is R
3. Current is going in this direction I 3 and this is my excitation, E ex, right and I gave
some nodal point a b. This is a, this is b, this is c and this is d.
Now, let us, let us analyze and let us analyze this circuit.
For the bridge to be balanced, is to be balanced, points b and d must be at the same
potential, at the same potential. So, I can write I 1 R 1, is not it, equal to I 4 R 4 and I 2 R
2 equal to I 3 R 3. So, simply I can divide. If I divide these two equations, I will get I 1
by R 1 I 2 by R 2 equal to I 4 R 4 I 3 into R 3.
35
Now, if there is no current, there is no current in the detector, so I 1 equal to I 2 and I 3

equal to I 4. So, R 1 by R 2 equal to R 4 by R 3, right, sorry, right? Now, what will
happen? Suppose R 1 is an unknown resistance or our, that means our circuit looks like
this. If I redraw the circuit, so R 1, R 2, R 3, R 4, so this is our output voltage. So,
suppose now R 1 equal to R x, so R x unknown resistance or strain, strain gauge, I can
write R 4 by R 3 into R 2, is not it?
Now you see, I can use either R 4 or R 2 to make the temperature compensation, is not it?
I can choose either of this one, because in that case whatever the changes occur in the R x
due to the temperature, the same change will occur also in the R 4 and R 2. Then, either R
4 can be dummy gauge, this can be dummy gauge or R 2 can be dummy gauge. In that
case any of the resistance, if you check R 4 suppose for balancing the, I mean for, as
dummy gauge, so in that case R 2 be used to balance the bridge. So, this is the case when
there is a no current flowing through the detector.
36
That means I am assuming in that case that is I will vary R 2 that means if I take a new
page that means, so this is R 1, R 2 R, 3, R 4, so this is our strain gauge suppose and the
R x, so as I told you before R x equal to R 4 by R 3 into R 2. So, this is our active gauge,
this is our dummy gauge, right? That is this is used to for making the temperature
compensation. So, ambient temperature, because if the ambient temperature changes, my,
this gauge will take care. So it, sometimes is called ambient temperature compensation
also, ambient temperature compensation and R 2 and R 3 should be of similar type. So,
these are fixed resistors or it can be variable resistors, because this is variable resistor to
balance the bridge, because of, suppose I have subjected to strain, again I will make e o
equal to zero. So, this R 2 can be calibrated in terms of the load, right or displacement,
whatever you like, because it can be used for making the displacement sensor also.
So, in that case R 2 will be calibrated in terms of the displacements or R 2 can be

calibrated in terms of the, in terms of the load also. So, it will be …., so that we will see,
look at and then, we will find how much is the load, whereas this will be our makeover
temperature compensation, right? We will find that in, later that we will use 4 gauge
simultaneously. All are active gauges to increase the value of the, value of the unbalanced
voltage, because I, sometimes I need large unbalanced voltage, so I can do it.
37
Now, let us go to the problem, problem 4.1, a steel bar of rectangular cross section 2
centimeter into 1 centimeter subjected to tensile force of 20 kilo Newton. A strain gauge
is placed on the steel bar as shown in the following figure. Find the change of resistance
of the strain gauge, if it has a gauge factor of 2 and the resistance of 120 ohm in the
absence of axial load. The Young’s modulus of elasticity of steel is equal to 2 into 10 to
the power 8 kilo Newton per meter square, right?
There is no question of bridge or anything. Simple problem we have given, we have

shown this that it is a, you see here that the installation it is, it is subjected to tensile, is
not it? So, it is, tensile load here, so it will be axial, so there is no question of any
transverse. So, there will be axial load and you have to find that the change of resistance
in this strain gauge. It is a very simple problem, right? So, no question of bridge or
anything, we have to, with that we will do later on, right?
Now, let us solve this problem.
38
Let us take, I am sorry, I can take this one. Here you see, lambda is equal to 2, is not it
and E equal into 2 into 10 to the power 8 kilo Newton per meter square, P equal to 20
kilo Newton and R equal to 120 ohm. Area of cross sections we have given A equal 0.02
into 10 to the power 01 meter square. So, stress on the bar, bar, is sigma a equal to P by A
equal to 20 upon .02 into .01. So, this will give you 10 to the power 5 kilo Newton per
meter square.
So, epsilon a or epsilon equal to sigma by E or sigma a by E. E is the Young’s modulus

10 to the power 5 by 2 into 10 to the power 8. So, it will be 5 into 10 to the power minus
4. So, dR by R we know by dL by L equal to lambda, is not or I can write dR by R equal
to lambda into epsilon a axial. So, delta R or dR will be equal to R epsilon into lambda.
So, it is 120 into 5 into 10 to the power minus 4 into I want to take the resistance, 5 into
10 to the power 4 is strain, multiplied with the gauge factor, which is advance probably,
yes this is advance strain gauge, obviously 2, so this will give you change of resistance
.12 ohm, right? So, the answer is 0.12 ohm.
I have solved this problem to know that you see that how small this change of resistance.
It is not very easy task to see this, measure this and as I told you that just using a one
39
simple, you can use one resistance that means R 2 and calibrate R 2 in terms of stress or
strain, it is very difficult and you look at this change of resistance. It is very small. So,
mostly we will see that, we will use that unbalanced voltage and we will calibrate that
unbalanced voltage in terms of displacements or force or pressure or load, whatever it
and we will use, instead of one gauge we will use four gauge, try to get four gauge. In
that case, there will be more unbalanced voltage.
Then the, we have seen that in the case of, in the case of our displacement sensor, one
gauge will be under compression, another gauge will be elongation. I will get the
temperature compensation as well as my bridge unbalance voltage will be also very high,
so that this will help us to make the circuit, signal conditioning circuit simpler, because if
the small, higher the voltage, the more and more or less will be the error in the
measurement also. So, those things are to be, to be considered. So, strain gauges we have
covered. We will see later on in the future lesson that it will be extensively used for the
measurement of load, right?
So, this is a basic sensor which is used, but a strain gauge alone cannot be used. It is used
either in some pressure gauge or some load cell and all those things and this ends the
lesson 4.
40
Prof. Alok Barua
Lecture - 5
Load Cell
Good afternoon! This is lesson 5 of the course Industrial Instrumentation. In this

lesson, we will cover the load cell. Load cell is very important devices. It is basically
a mass measuring or weight measuring instrument or sensor, whatever you say, using
the principle of strain gauges. We have seen that in the lesson 4, we have covered
strain gauge in details. So, using the strain gauges, now you will see that how we can
make a load measuring devices.
Now, it is a very simple form of instrument. That are, the advantages of this
instrument that it is simple, it can have a wide range, it is a linear instrument and it
has, it can operate in a dusty and corrosive environment. These are the some
advantages of this particular instrument. You can see in a, in a, if you move in a
highway that there are load cells and weigh bridges basically you will find, where
they used to take the weight of the truck or how much truck , how much load the truck
is actually carrying, they can find in the load cell. If you know the tear weight or
empty weight of the truck, so that you can find how much load it is carrying.
It is also used in the measurements of the load in the wagon online, I mean continuous
measurement of the wagon, I mean how much weight they are in the wagon, all those
things, I mean it is possible with this type of instrument. Basically we will cover,
there are, there are two types of load cells. The first is the link type load cell and
secondly we will cover the beam type load cell.
Now, link type load cell, usually we will find it is used for measurements of the large
load, whereas beam type load cell is used for, for the measurement of the small load.
Now, there are other type of load cell like ring type load cells, all those things, but due
to the restrictions of the, because same, principle is basically same that is the reason
we taken, we have taken two different load cells and cover it in the, in this particular
lesson.
1
Now, so far the disadvantages are concerned there are only, there is only one
disadvantage, which is the phenomena called the creep. Now, creep is basically that
the, it actually is a permanent deformation of the load cell. If a load is applied to a
particular cell for a quite long time, that does not happen in the case of, I mean in the
case of weigh bridges and all those ……. truck will move and after sometime the
truck will go away. So, but if by chance if you apply a load permanently there are the
chance of deformations, which will lead to the drift of the instrument, zero drift of the
instruments and all those things. So, you have to be careful, very careful about that.
The careful selection of the, also the material of the load cell also can prevent that
type of problem.
Now, there are two basic type of load cell. One is so far the ranges are concerned, one
is for the small load. Suppose from 500 gram to you can measure up to 5 kg and
others you can measure suppose 10 kg to even 1000 tons, that type of load cells are
also possible.
Now, contents are, we will first cover the unbalance Wheatstone bridge because we
will use here the unbalanced Wheatstone bridge. That is the reason we have discussed
the unbalanced Wheatstone bridge here. Then, we will cover the column type load
cell. Instead of link type load cell, I mention it is a column type load cell. Sometimes
people, some people call it even the axial link type load cell, some people call column
2
type load cells. It looks like a column that is the reason it is more appropriate if I call
it as a column type load cell. Then we will cover the cantilever beam type load cell,
because it is basically a cantilever on which the sensor are used. That is the reason it
is called cantilever beam type load cell.
At the end of this lesson, the viewer will know installation of strain gauges on load
cell, how the load cells will be, I mean installed and this is very important to make the
sensitivity high, also to get proper output. You will see that if the orientation is
something different, whatever we have prescribed here in the case of column type
load cell, you will find there is no output. So, orientation is very important for making
a load cell, because strain gauges are basic sensor there and it will make the complete
instrument of load cell.
Sensitivity of the load cell, every time I mean in both the cases, both the, in the case
of column type load cell as well as in the beam, I mean in the cantilever beam type
load cells, we will compute the sensitivity of the load cell. Sensitivity means the load
cell as well as the Wheatstone bridge combination, because ultimately we will
measure this unbalanced voltage by a means of Wheatstone bridge and that
unbalanced voltage will be calibrated in terms of load.
3
Range of a load cell, all that we will find that there are different load, I mean different
range of the load cell that also and fatigue strength, all these things is very important
to find the range of the load cell and you will see that in the case of column type load
cells, the range is much more than the beam type load cell.
Now, as I told you that, you will use basically Wheatstone bridge in the load cell.
Now, Wheatstone bridge is commonly used to convert a change in resistance to an
output voltage; that we have seen in the case of simple, I mean balanced Wheatstone
bridge. So, here I will use unbalanced Wheatstone bridge. You see, this is our basic
Wheatstone bridge circuits and we can see here that that we have used 4 resistance. It
is R 1, one resistance, then we have R 2, then we have R 3 and R 4 and we have used
an excitation E ex. So, we have written E ex that is excitation voltage.
We can use DC voltage as well as AC voltage, it does not matter, because ultimately
you will get a resistance. It is, so far it is a pure resistance then it hardly matters
whether, whether you are using a DC or AC voltages. Here, we have used DC
voltages and this is unbalanced voltage of the, of the bridge. So, this voltage e naught
will be calibrated in terms of the load which actually we are going to measure by this
instrument.
4
Now, with an initially balanced bridge, we can say that output voltage e naught
develops when the resistance R 1, R 2, R 3 and R 4 are varied by the amount delta R
1, delta R 2, delta R 3 and delta R 4 respectively, we will get. With the new values of
the resistances, change in the output voltage can be expressed as e naught equal to R 1
plus delta R 1, R 3 plus delta R 3 minus R 2 plus delta R 2 multiplied by R 4 plus
delta R 4 upon R 1 plus delta R 1 R 2 plus delta R 2 whole in the parenthesis
multiplied by R 3 plus delta R 3 plus delta R R 4 plus delta R 4 whole multiplied by
the excitation voltage E subscript ex.
5
Substituting for initially balanced bridge, we can assume that if the bridge is initially
balanced, usually in the standard load cells we will find that we will take all the four
resistance equal; it is not necessary you have to, I mean you have to make it equal,
because as you know the basic principle Wheatstone bridges are the bridge will be
more sensitive, when it is, all the resistances were almost equal. Otherwise, you can
say that when R 1 into R 3 equal to R 2 into R 4 this is also the initial, the balanced
condition of the bridge.
Now, E naught output voltage or unbalanced voltage can be given by small r divided
by 1 plus r whole square, whole multiplied delta R 1 plus R 1 minus delta R 2 by R 2
plus delta R 3 by R 3 and delta R 4 by R 4 into E ex, E subscript ex. This is equation
number 1. This equation is very important, because you see that by looking at this
negative term we can tell that, actually where I will put this R 2 and R 4, because that
will make our bridge output much higher, because we always want in any
instrumentation system that the output should be higher and higher, so that our
sensitivity will be high and high, sorry, where small r equal to R 2 by R 1. So, if all
the resistances are initially equal that means R 1 equal to R 2 equal to R 3 equal to R
4, so in that type of situation, this will be simply 1, right?
Now, if I say that if I use, I mean we, we have used three bridges, I mean four bridges,
four resistances, I can use one resistance also in the case of Wheatstone bridge.
However, suppose I have a, I mean I can simply use a, so instead of using four
gauges, I can use only one gauge also. Suppose this is R 1, this is R 2, this is R 3, this
is R 4, so if these are the, if R 1 is the basic, I mean our strain gauges, suppose R 2, R
3, R 4 are the fixed resistance, in that situation you will find that all these terms that
means delta 2 by R 2, delta 3 by R 3, delta R 4 by R 4 all will be zero, right? So, our
output will be e naught equal to delta R 1 by R 1 multiplied by 1 by 4 into E ex, E
subscript ex. So, you can see that if I use 4 bridges, so I can make the sensitivity
almost four times. You will find that this is very much true.
Now, let us derive this equation that means the equation number 1. How we got it,
actually these equations, that is very important equations of unbalanced Wheatstone
bridge, so let us derive that.
6
Suppose these equations I am writing, I am first drawing the Wheatstone bridge. This
is R 1, so it looks like, so this is our output voltage e naught. I can use any simple
multimeter to measure that voltage and I gave the, this name to E ex and this terminal
node is a, this node is b, this node is c and this node is d. See, if you have this type of
equations, quite obviously I can write that E ab that is voltage between this resistance
R 1 is equal to R 1 by R 1 plus R 2 into multiplied by E ex. So, it does not matter
whether it is balanced or unbalanced. It is, so always this is true.
Similarly, the potential across the resistance R 4, which can be derived as E ad equal
to, this is separate, E ad equal to R 4 upon R 3 plus R 4 that means current flowing
through this path multiplied by it is the E subscript ex.
7
Now, the output voltage, I can take a new page, the output voltage, therefore the
output voltage, output voltage e bd equal to e ab minus e ad, which I can write simply
R 1 minus upon sorry R 1 plus R 2 minus R 4 upon R 3 plus R 4 E ex equal to R 1
into R 3 minus R 2 into R 4 upon R 1 plus R 2 R 3 plus R 4 multiplied with the
excitation voltage E ex, right? So, these two terms, I mean we will see that R1, R 4
and R 4 and R 1 are cancelled out.
8
So, if initially the bridge is balanced, if I say that the R 1, R 3, R 2 equal to R 4, as I
told you earlier not necessarily the, all the resistance will be, I mean necessarily it will
be equal, I mean if this condition is fulfilled at all, then also the bridge is balanced.
So, the output e naught will be zero, right. Now, suppose all the resistance are getting
varied, as it happens in the case of load cell, both the, in the case of column type load
cell and the link type cell, all the four resistances are getting varied. If it is, we assume
that the resistance R 1, R 2, R 3 and R 4 are varied by delta R 1, delta R 2, delta R 3
and delta R 4 respectively, so my output voltage will look like …
You see, e naught will be equal to R 1 plus delta R 1 multiplied by R 3 plus delta R 3
minus R 2 plus delta R 2 R 4 plus delta R 4 upon R1 plus delta R 1, sorry, this is, I am
sorry again, so there should be actually a plus sign, yes plus R 2 plus delta R 2, R 3
plus delta R 3, R 4 plus delta R 4. See, if I take R 1 into R 3 common in the, from the
numerator R 1, R 3 1 plus delta R 1 by R 1 1 plus R 3 minus R 2 R 4 1 plus R 2 by R
2 1 plus delta R 4 by R 4 bracket closed upon, all same R 1 plus delta R 1 plus R 2
plus delta R 2 R 3 plus delta R 3 R 4 plus delta R 4, right.
So, this almost we can write, because delta R 1 and delta R 3 will be very small. So,
the multiplication of these two terms we neglect. So, this equation almost will be
equal to R 1 R 3 1 plus delta R 1 by R 1 plus delta R 3 by R 3 minus R 2 R 4 1 plus
delta R 2 by R 2 plus delta R 4 by R 4, right? Here also we neglect the term,
9
multiplication term that is R 2 product of delta R 2 and delta R 4. We have seen this
will be quite small, upon same denominator R 1 plus delta R 1 R 2 plus delta R 2 R 3
plus delta R 3 plus R 4 plus delta R 4, clear? So, R 1 R 3 1 plus, okay, yes take a new
page again.
So, since the bridge is balanced, we can write R 1 or the initially the bridge is
balanced R 2 R 4 initially when bridge is balanced, I can write the expression output e
naught will be equal to R 1 R 3 delta R 1 by R 1 minus delta R 2 by R 2 delta R 3 by
R 3 minus delta R 4 by R 4 upon R 1 plus R 2 multiplied by R 3 plus R 4. So, if I take
from the delta R 1 by R 1 minus delta R 2 by R 2 plus delta R 3 by R 3 delta R 4 by R
4 if I take R 1 R 3 common from the denominator 1 plus R 2 by R 1 multiplied by 1
plus R 4 by R 3, like this one, we assume that the ratio …
10
If we assume ratio R 2 by R 1 equal to small r, then output voltage will be equal to

delta R 1 by R 1 delta R 2 by R 2 plus delta R 3 by R 3 minus delta R 4 by R 4 upon 1
plus r 1 plus 1 by r E ex. So, this will give us the equation r 1 plus r whole square
delta R 1 by R 1 minus delta R 2, I can erase this part again, so it will be delta R 2 by
R 2 plus delta R 3 by R 3 minus delta R 4 by R 4 into E ex. These expressions we
have, the first we have shown in a basic. These equations will be utilized for all
unbalanced Wheatstone bridge, right?
So, in both the case of the column type load cell as well as in the beam type load cell,
we will use this unbalanced Wheatstone bridge equation and it depends, as you know
that if you make only two gauges, in that case delta R 3 and delta R 4 by R 4 can be
cancelled out, I mean you can use a fixed resistance. In that case, our output will be
obviously less. But, if you use four gauges, you will get temperature compensation as
well as the output will be more. As you know, for in the Wheatstone bridge to make
the, because strain gauge is basically very temperature sensitive, since it is a
resistance, resistance changes with temperature, but we want that the resistance will
change only when the we will apply the load to the, that load cell. So, the temperature
effect should be nullified. For that reasons, we have used that ambient temperature
compensation scheme.
11
So, for the ambient temperature compensation, at least I need two gauges; one gauge
can be active, other gauge can be passive also. That means it can be dummy. Even
though it is made of strain gauges but you can, that may not be subjected to load.
In that case I will get temperature compensation, whereas in the case if I use two
active gauges, then you will get temperature compensation as well as my output will
be doubled. But if you use four gauges, I will get temperature compensation and my
output also will be four times. In the case of, you will see that in the case of mean, I
mean in the case of mean type load cell only that I will get the output almost four
times.
Now, let us go back to our, so this was our final expressions we have seen, right? So, I
can choose some other pen, yes.
Now, column type load cell - a simple uniaxial column type load cell is shown in
Figure 5.2. Load can be either tensile or compressive, because it depends on the
where you will apply the load. If you apply the load on the top of the load cell, it is, it
obviously the bottom of the load cell should be fixed. So, in that case it is a
compressive load, whereas if you suspend the load from the bottom of the load cell
that means our column type load cell, it will be a, it will be a load of elongation.
12
So, in that case, it is not … a compressive load. It is a, it is a load of elongation. So,
that is the reason we are calling it either tensile. It does not matter, our expression, in
a final expression of the Wheatstone bridge will remain same. The strain gauges 1 and
3 will show the, are subjected to the axial strain and gauges 2 and 4 subjected to the
transverse strain.
You see that this is a column type load cell, which is figure 2. Now, you see here that,
let me take a pen; yes, you see here this strain gauge, here this strain gauge, it is
shown like this one, whereas this is shown like, so these strain gauges and there is,
similarly there is another, because if you take there are four phases of a load cell and
this load cell, it appears that it is subjected to the tensile load or elongations. Now,
what will happen? You see that, in this case this load cell will be subjected to axial
strain.
Now there are, similarly there are, you will find the other side, because if there are
four phases and you cannot see the phases three and four of this strain, of this load
cell, see in the phases three also, that is the backside of this one also, there is another
strain gauges which is subjected to axial strain, whereas gauge number four will be
subjected to the transverse strain, right and this is axial. This is, that is the reason we
call it axial strain. These are the resistance R 1, R 2, R 3 and R 4. This is subjected to
13
axial strain, this is subjected to transverse strain. Gauge number 3 again is subjected
to axial strain and gauge number 4 will be subjected to transverse strain.
Now, orientation of the gauges, this is like this, this will be like this. So, other side
this is on the phase three. So, it will be axial. So, it will be again up. That means it
will look like this, whereas in the case of gauge, this is three. In the case of gauge 4,
so it will again, it will look like this, right? So, this will be subjected to transverse
strain, whereas this will be subjected to, I mean tensile strain, because these are all,
since as it is shown in the figure, so it is subjected to tension, so it will be tensile
strain, right and now we have put in a Wheatstone bridge, it looks like this one.
Now see, what is very interesting factor that if I put all the strain gauges in the axial
direction, then what will happen? If I put all the strain gauges that means this gauge
number 2, gauge number, gauge number 1, gauge number 2, gauge number 3, suppose
all are subjected to axial, then what will happen? Interestingly we will find the output
will be identically zero, is not it? You can apply in our, our Wheatstone, unbalanced
Wheatstone bridge equations, equation number 1, probably. So, we will find that
output will be zero; that we do not want. That is the reason we have changed the
orientation of the gauges, whereas in the case of beam type, cantilever beam type load
cell you will find the, all the, I mean gauges are in the same directions. All are
subjected to axial. Obviously, two will be compressive and two will be under tensile
load.
14
If it is there, so I can say that, now axial strain, you see it is given by epsilon ax. We
write P upon A into E, right? Now, see we have written like this, E ax, so that it can
easily divide, the E ax means axial. Now, transverse strain is given by nu P upon AE.
Now, what are those, I mean legends let us look at, where A is the cross sectional area
of the link. We have seen the link.
So, the link is, looks like this, is not it? So, this area, this is, the cross sectional area is
A, right?
15
E is the modulus of elasticity of the link material, right? Then, we have Poisson’s
ratio nu of the link material and P as you know is the load, basically. It is a load and
the ….
So, I can write delta R 1 by R 1 equal to delta R 3 by R 3 equal to lambda into epsilon
ax equal to lambda P by AE. This is coming from the, I mean simple our strain gauge
formula. Since delta R 1, R 1 and R 3 are initially same, we can assume that delta R 1
by R 1, because you see it very carefully, both R 1 and R 3 are under axial load, right?
16
So, whatever the value of R, delta R 1 by R 1 is, it will be similar to delta R 3 by R 3,
right? So, it is multiplied by the lambda which is, lambda is the gauge factor of the
strain gauge materials, multiplied by the strain will be equal to lambda P by AE and
delta R 2 by R 2 will be equal to delta R 4 by R 4 equal to lambda into epsilon
transverse that means transverse strain equal to minus, I mean nu lambda P by AE.
This is coming from the basic formula which we have discussed, when we, in the
lesson 4 actually when we discussed the load cell and when we discussed the strain
gauges, where lambda is the gauge factor of the strain gauges used in the load cell,
right?
So, the output voltage e can be expressed in terms of the load P, assuming R 1 equal
to R 2 in equation 1. So, you put all these values of delta R 1 by R 1, delta R 2 by R 2,
delta R 3 by R 3 and delta R 4 by R 4 in equation 1. We will get output voltage e
naught equal to lambda P multiplied by 1 plus nu E subscript ex, excitation voltage
upon 2 AE, right? So, you see, it is a very, basically very linear instrument. If you
look at very carefully, the P is load actually, lambda is the strain gauge, depends on
the strain gauge usually if you use advance it is lambda equal to 2. It is a Poisson’s
ratio, this is also constant. Supply voltage, DC supply voltage, this is also constant,
area of cross section of the link this is also constant, Young's modulus of the material
is basically …
17
So, you can, interestingly you see nu and e will depend on the link material, where
lambda will depend on the strain gauges type of, so these are all constants. So, you
will find that the output voltage ex also, if the supply voltage remains constant, if you
use a battery, so obviously it will remain constant. Output voltage is directly
proportional to P, right? This is very important, because it will make us very
simplified signal processing. It will be our, it is basically coming under the category
of the linear instrument or I can little manipulate, so I can write P equal 2AE upon
lambda 1 plus nu multiplied by E ex into e naught, Ce naught, where C is called the
calibration constant of the, of the load cell.
Therefore, load P is linearly proportional to the output voltage e naught and the
calibration constant C is equal to 2AE lambda 1 plus nu into ex. So, obviously you
can see that the sensitivity of the load cell Wheatstone bridge combination, there is no
use of having only load cell. So, if you combine these two until its sensitivity is given
by S equal to e naught by P is load, 1 by C lambda 1 plus nu ex by 2 AE; very
important equations if I look at, so you see this is our equation.
Now, you see that one thing is very important that sensitivity can be increased by
increasing the lambda. Poisson’s ratio, it is hardly, it does not matter much because it
is around very small. It is .3 in the case of steel. Supply voltage we can increase,
because if you increase the supply voltage, obviously our output voltage will also
18
increase or we can decrease the value of A and E, right? A can, we can decrease that
is area of cross section of the load cell we can decrease. I mean in ….. our column
type load cell. Also, we can decrease the Young's modulus of elasticity of the
material. So, we can choose material in such a way the Young's’s modulus of
elasticity is also lower. So, in that case our sensitivity will increase. But, one thing is
very important. You see that in the case of, I cannot E, increase the supply voltage or
excitation to my bridge indefinitely. Because what will happen you see, each strain
gauge has the resistance. It is basically, it is a resistance, so it has a current limitation,
how much current I mean will pass through?
So, if I increase ex, so please note that since it is two resistances are in series,
obviously what will happen that the current through the, each of the gauges will
increase. So, you have to look at that this current should lie within the limit or the, if I
square r if you take, so that it should remain within the range of the dissipation of the
strain gauges. So, I cannot increase it indefinitely as some prescribe. Usually what we
do? Usually if you find that the, if there, we know the resistance of the strain gauges
and if you know the current, allowable current through the strain gauges accordingly
we will choose the power supply, so that under no condition the, it will cross the
current limit, right?
Now, interestingly you cannot decrease 2 and E. You will see that if you decrease 2
and A, A and E you will decrease the range of your load cell and one important thing
you see that, you see that load cell looks like this. If it is on the compressive load, one
thing we should mention, because we have already discussed the tensile load. If it is
under the compressive load, so it should not buckle. Usually what we will fix this
point and give a force on this side. So, if the load cell should not, I mean get a shape
like this one, right. So, that will create problem, so that it should not buckle. So, so if
you decrease that A and E, you will find there is a chance of that type of formations
which will damage all our equations or whatever the equations which we have derived
are no more valid in that type of situations.
19
This arrangement gives a sensitivity 2 into 1 plus nu times that achieved with a single
active gauge in the bridge. We have seen that if you use a single bridge, because you
see here, if I go to the previous slide we will find that or you can see that this one that
this 2 nu 1, 2 into 1 plus nu times that achieved with a single active gauges it is true,
because in the case of single gauges, so you will find the sensitivity will be much
lower, right. It also provides temperature compensation, as I told you earlier, since all
the four gauges are at the same temperature.
The equation 2 indicate sensitivity of the column type load cell depends upon cross
sectional area A, modulus of elasticity and Poisson’s ratio of the column, gauge factor
of the strain gauges that is lambda and excitation voltage of the bridges.
20
Now, range of the link type load cell or column, I mean, type load cell, both way we
can say. Either we can tell column type load cell or we can tell link type load cell, is
given by P max equal to S subscript fs into A, where S fs, S fs, is called the fatigue
strength of material of the load cell, right? So, if you compare this equation with the
previous equation that means sensitivity as well as the range, if I compare side by side
sensitivity and range, because this will increase the range, P max is the range of your
load cell, you will find that either sensitivity is high or range will be low and vice
versa, right. That means low range of the load cell that means infer that the high
sensitivity are associated with the low range of the load cell, while low sensitivity is
associated with the high range of the load cell.
21
The voltage ratio at maximum load e naught upon E subscript ex maximum for the
column type load cell Wheatstone bridge combinations will be given by e naught by
ex maximum lambda S fs, I mean fatigue strength 1 plus nu by 2E.
Now, load cells are usually fabricated from steel with following specifications. So, E
Young's modulus is 2.6, 2.0684 into 10 to the power 11 Newton per meter square, nu
equal to.3, Poisson's ratio and fatigue strength equal to 5.516 into 10 to the power 8
newton per metre square. Lambda equal to 2 for the advance; if the, if the strain gauge
22
is made of advance, so the gauge factor will be 2. So, we have taken lambda equal to
2. In that situations we will find that e naught by ex maximum is equal to 3.47
millivolt per volt, right? So, this is our expressions.
Now, some important points; now, see under the deflections of the column type load
cell, under the full load is of the order of 0.025 millimeter to 0.38 millimeter,
extremely small, because it has high stiffness and therefore, natural frequency of the
load cell is not very important. It will be determined by the force carrying elements
such as weigh bridges, because that itself carries, I mean a lot of its mass is much
more higher than the load cell itself. So, frequency all this things will be covered, I
mean determined by that platform itself.
23
Now, some important features; you see that to achieve the high accuracy, additional
temperature compensations will be required as shown in the following figure. Some
additional circuits use modified Wheatstone bridge for additional temperature
compensation is necessary, because you will find it may happen that the, all the
resistance may not have the same temperature coefficient of resistance. So, in that
type of situations, we have added, actually you will find very carefully only two, three
resistance R x, R y and R R z, probably all will be the, I mean varying sort of
resistance.
24
R x and R y are the temperature sensitivity, temperature sensitivity resistors and R x is
the compensation, compensates slightly different temperature coefficient of
resistances. R y, it is a non temperature sensitive resistors and it will provide desired
sensitivity. I can control the sensitivity by that.
This is our cantilever beam type load cell. That ends the, previously ends the, now
cantilever beam type load cell, you see that this is actually cantilever, you can see
here. This is a cantilever and we have installed, we applied a force P, right, so that it
looks like that the, I mean you will find it previously like this, now it will bend like
this, is not it? It appears, you see, if I apply the load P here, this side is bending more
that free end, this is the fixed end. But, you see very carefully that the strain gauges
were installed very close to the hinge point. That means strain gauges are installed
here, it is here.
Also, in the bottom also, there are two strain gauges. The reason is stress is maximum
at this point, not at this point. The stress is not maximum at this point or stress
gradient is also maximum at this point, but not at this point. So, that is the reason we
install the strain gauges in this directions. Now, you see that here also we have used
four strain gauges, 1 and 3 on the top and 2 and 4 at the bottom of the strain gauges. It
cannot see from the top, so 2 and 4 at the bottom of the strain gauges, bottom of this
cantilever beam type load cell, right.
25
So, you see that if I apply a force here at this point, at the extreme end, the load, the
strain gauges 1 and 3 will be subjected to tensile load, whereas strain gauges 2 and 4
are subjected to compressive load, right and we have taken a dimensions that means
this is, width is w of the strain gauge of the, of the, this cantilever, height is h and I am
applying a force, which is almost from the midpoint if I take at a distance of X from
the point, midpoint of the strain gauges. So, this is called the cantilever beam type
load cell. Let us do the analysis.
So, orientations of all strain gauges are in axial direction; this is more important. In
the case of column type or link type load cell, we have seen that the gauge 1 and 3 are
in axial directions, whereas gauge 2 and 4 are in the transverse directions,
orientations, whereas here all in axial directions. The gauges 2 and 4 are on the
bottom surface of the beam. The resistance of the gauges 1, 2, 3 and 4 are assumed
that the R 1, R 2, R 3 and R 4 respectively.
Moreover in the, in the unstrained conditions R 1 equal to R 2 equal to R 3 equal to R

4. There is, there is no force applied on this gauge. So, R 1 equal to R 2, R 3 equal to
R 4 and gauge, interestingly gauge 1 and 3 will be, I mean, I mean subjected to
elongation, whereas gauge 1, 2 and 4 will be subjected to compression. Because it is
on the bottom, so it will be subjected to compression.
26
This is our strain gauge circuits. We can see E ex, you see here. So, strain gauge R 1,
strain gauge 1, actually it should be, so …. parenthesis, so parenthesis and this will be
like this one. Same excitation voltage we have given and I am getting output voltage e
naught, right? Now, you see that, we have already seen that gauge, gauge 1 and 3 will
be at the top. It will be subjected to the tensile load, whereas gauge 2 and 4, while the
bottom surface of the cantilever beam, it is subjected to compressive load.
27
If that is the situation so I can see that I will get, the load P produces a moment of P
into x. The strain developed at the gauges are given by epsilon 1 equal to minus
epsilon 2, since it is, I mean under elongation, this is under compression, so these are
same in magnitudes if the initial value is same, but opposite in sign, equal to E 3 equal
to minus E 4 6Px divided by Ewh square, where w is the width of the cross section of
the beam, h is the height of the cross section of the beam.
The change of resistance in the gauges will be given by, you see change of resistance
of the gauges will be given by delta R 1 by R 1 equal to minus delta R 2 by R 2 equal
to delta 3 by R 3 equal to minus delta R 4 by R 4 equal to 6 lambda Px by Ewh square
which is equation number 4. Substituting equation 4 in the expression for unbalanced
Wheatstone bridge, which you have derived sometime later, the output voltage can be
written as simply e naught equal to 6 lambda Px. Because already that is negative sign
in front of the delta R 2 by R 2, so it will, all will become positive 6 lambda Px by E
subscript ex by Ewh square that is the Young's modulus.
28
So, P equal to Ewh square equal upon 6 lambda x into E ex into e naught. So, the P is
proportional to e naught. Again, it tells that it is a linear sensor. You see here it is a
linear sensor. P is proportional to e naught, obviously it is a linear. Again, let us find
the sensitivity. Sensitivity of the cantilever beam type load cell and Wheatstone
bridge combinations will be e naught by P equal to 6 lambda x E ex Ewh square. So,
as it happens as before that if you reduce E, w and h, so obviously our sensitivity will
increase. But, if you decrease w and h, there is a little chance of reducing E, because
we will use some, I mean steel, so we can only manipulate w by h square, w and h
square. If I reduce wh, our range of the instrument will also reduce, right? Similarly, I
cannot increase ex, excitation indefinitely, because this will create the problem of
that, it, it may cross the limit of the, current limit of the strain gauges.
Sensitivity depends on the shape of the beam. Shape means wh that is the shape of the
beam and the point of applications of the Young's modulus of the elasticity of the
material of the beam and the gauge factor and the point of application of the load,
because it is, x is the point of the application of the load. So, it looks like this. You
see, this is our x, this is lambda, this is wh, sorry, this is wh and we can take, I do not
know, yes, Young's modulus elasticity. So, it actually, basically we will …., so it will
depend on, sorry, anyway it depends on the Young's modulus of the material.
29
The range of the cantilever beam for measurement of load is Pmax equal to S fs wh
square by 6x and this is the range. Obviously, it depends on the fatigue factor. The
voltage ratio at the maximum, so you can see that it depends on the, so many other
points like fatigue factors and all these things. So, you can see here and this will
depend on all this wh and 6x. So, voltage ratio at the maximum load is given by, I am
sorry, so voltage ratio at maximum load is e, e naught upon E ex max equal to lambda
S fs by E.
30
Now, there are, even though we discussed about the, all the, I mean electrical load
cells, there are, pneumatic and hydraulic load cells are also there. Even though not
much in popular, I mean there are intelligent load cells; we call it intelligent load
cells. If you have, add some factors on this one, it looks like that it can be formed by
adding a microcontroller to a standard cell. You know that microcontroller is
basically, there are two types of microcontroller available. In many a cases in
instrumentations we will find we use microcontrollers.
Now, microcontroller is nothing but a microprocessor, but with some additional

peripheral chips built in. It is a some sort of embedded systems also, because those
peripheral chips are there. Now, one of the popular microcontrollers, as you know, is
8051, which is basically an 8 bit microprocessor along with the, along with the certain
amount of RAM. It has a very small RAM, 128 byte RAM. Then, two counters, so all
those peripheral chips are inside these machines. So, we can use it for measurements
of, for many controlled mechanisms.
Now, this is very important, because if you need little bit of computations, suppose in
the case of, in the case we are no more use, nobody uses microprocessors, we use a
microcontroller. There are two types of microcontroller available; the 8 bit
microcontrollers and 16 bit microcontrollers. 8051 or 8091, is basically almost same,
but it has the features which is basically 8 bit machines, 8 bit microcontrollers and
because there is no use of using 8 bit microprocessor and you will find incidentally
that 8 bit microcontroller can function so much, that it can do so much computations,
we do not need 16 bit microcontrollers.
So, if it can be added as it say the ..…. load cells, I can make it. It can be formed by
adding a microcontroller to a standard cell, right. It can calculate, display total cost
from the measured weight using stored cost per unit weight information, because
ultimately those who will take the reading of their truck, so they have to, must pay, so
that it depends on how much load you are carrying, so depends on them, usually they
pay. So it can be, we can all the scale factors different things, if there is certain
amount of …. also can be solved with the use of microcontrollers, right?
31
So, this is called the intelligent load cell. This is not, not new for the load cell. I mean
you all intelligent sensors or intelligent, I mean machines are, it has some inbuilt
computations facilities. So, if you use a microcontroller, then we will call it intelligent
machines.
Now, cost per weight can be stored for a large number of substances making the
instrument very flexible in its applications. This is also a, it is not true for, only for, I
mean this type of load cells. You will find in many applications, you will, you can do
it that means, that means suppose if you, I mean you can make a machines that means
all the weights are, I mean selected all the weights of, I mean they are all, you, all the
cost of the per unit volume ….. per kg of the different materials are there, especially
in the grocery store and all these things, in abroad you will find these type of things
are there that you can choose the particular cost and can calculate and make the total
value of the, your material whatever you have bought.
So, this is all about our intelligent I mean load cells and it is very extensively used.
So, we basically discussed two basic load cells that means the, a link type as well as
your cantilever beam type load cell. As I told you earlier, the link type load cell,
sometimes people call link type load cell, sometimes people call it column type load
cell and other is a cantilever beam type load cell. You will, usually you will find that
32
the cantilever beam type load cell used for the less load, slightly lesser load, whereas
the column type load cells are usually used for the larger loads, right?
This ends the lesson 5 of Industrial Instrumentation.
33
Prof. Alok Barua
Lecture - 6
Torque Measurement
Good afternoon! This is lesson 6 of Industrial Instrumentation. In this lesson, we will

cover the torque measurements. As you know, the torque is very important parameter.
Even though it is not exactly the process parameter, but it is also a parameter which is,
these are paramount important for the, most of the, most of the mechanical measuring
systems, where there, wherever there is a shaft, rotating shaft, so there is a need to
measure the torque, I mean developed in that, in the, in the, in the shaft, because it is
necessary to know that, know the optimal force up to which that shaft can withstand.
So, it is necessary to measure the torque in the system, so torque in the shaft. So, for
that reason, we have to measure the torque in the shaft.
So, there are various techniques of measuring the torque. Now, we will discuss in
brief two measure, two techniques and we will discuss the details one of the most
modern technique and its signal conditioning circuitry, which are actually used in the,
I mean in the industry.
1
Its contents - the torque cell, data transmission by slip ring method. Torque cell means
it will cover all the three different cells and when the question of data transmission
comes, it is basically the third type of torque cells which we will discuss, which is
based on the strain gauges. So, it is concerned with the third type of torque cell or
torque measurement techniques and wireless data transmission. There is, one is the
slip ring method, then there is wireless data transmission. This we will discuss in
details.
At the end of the lesson, the viewer will learn the different toque measurement
techniques, details of strain gauge based torque measurement, torque cell data
transmission.
2
Now you see, the measurement as I asked you, I mean told you earlier, the
measurement of applied torque is of fundamental importance in a rotating shaft. It will
ensure that the design of the rotating element is adequate to prevent failure under
shear stress, because you see it is a rotating shaft, so always there will be some shear
stress will be developed. So, we must know that how much is of, what is the optimum,
I mean shear stress which the shaft will withstand and before, I mean failure. So, for
that reasons we have to measure the torque.
Now, torque measurement is necessary to know the power transmitted by the rotating
shafts also. These are, I mean need of the torque measurement.
3
There are three methods of measuring torque and these are measuring the reaction
force in cradled shaft bearings. This is one of the methods or I should say the prony
brake method. Then we have, sorry, these are the, this is basically cradled shaft
bearing and the second is the prony brake method and third is the measurement,
measuring the strain produced in a rotating shaft due to applied torque. This is
basically based on the strain gauges. This, the third one, we will discuss in details.
This one we will discuss, this one, third one we will discuss in details.
4
Now, reaction force in shaft bearing, you see the torque transmission through a shaft
contains both power source and a power absorber. There must be a power source, it
must be some motor through which the power is coming and it is transmitted to some
other, suppose somewhere in the some lath machines or some other devices, so that is
the power absorber. Now, the torque is measured by cradling either the power source
or the power absorber end of the shaft bearings and then measuring the reaction force
F and the arm length L.
Now, the torque will be given by F into L. So, you must know F and we have already,
L is fixed, so we will know the L, measure the L, we will measure F and this will give
you the torque.
Now, this is, you see this is our, this is our cradled system. You see, here we have a
absorber and we have a source and we have a cradle. You see, here this length is L.
So, if you know L and if I know the force F, this is the Force F, so I can measure the
torque. This is one of the oldest methods of measuring the torque, right?
5
Now, prony brake method is basically, a rope is wound round the shaft and one end of
the rope being attached to a spring balance and the other end carrying a load of mass
m, right?
So, it looks like this. You see that we have a spring balance here and sorry, we have a
spring balance and then, it is rotating. Actually the, this should be little thinner, so it
should be the very … like this one. It is wound like this and it is coming like this. So,
it is little bigger and mass m is installed in one end and the spring balance is installed
6
in the other ……. This spring balance is graduated, so it is moving on a scale like this
one; has a conventional spring balance system, right?
(Refer Slide Time 6:41)
Now, if the measured force in the spring balance if F sb, then the effective force F eff
exerted by the rope on the shaft is given by F eff that means F subscript eff equal to
mg minus F sb, where F sb is the measured force, right? If the radius of the shaft is R
sh and that of the rope is R r, R subscript small r, then the effective radius is R
effective equal to R shaft plus R r, radius of the, radius of the shaft and radius of the
rope. So, if it is there, then I can write, the torque in the shaft will be given by T equal
to F eff into R eff. Obviously, it is in Newton meter, right?
Now, one of the most convenient, because you see, in all the previous cases of, both
the cases prony brake and the cradled systems, so we do not have the, the output is not
electrical. It is a mechanical output. It is very difficult to, I mean convert that type of
mechanical signal in the electrical domain and in all, you know, in the instrumentation
system we always prefer the electrical signal, because it can be transmitted easily. It
can be converted to the current domain, it can be transmitted over a long distance, it
can be digitized; all those advantage is there in the electrical system. So, always we
prefer the electrical system, where in instrumentation system where the output is
electrical.
7
In this case, in the measurement of torque by the strain gauges, you see basically the
four strain gauges are used, right? 4, I will show the diagram just now. Four strain
gauges are used, because if the, if the torque is rotating, so there is some amount of
shear stress on the torque. So, we will measure that shear stress and find the torque,
right and then we will put the strain gauges. There are four strain gauges, as you
know, the four strain gauges is always necessary to make the temperature
compensation as well as the large bridge output. So, it will be, I mean we will, it is
converted to the electrical domain.
So, this will be connected to a Wheatstone bridge and we will get the unbalanced
voltage. That unbalanced voltage will give, will be calibrated in terms of torque. So,
basically whenever we are talking of the sensitivity and all these things, so please note
that we are considering the, this torque cell as well as the assembly, entire Wheatstone
bridge assembly. So, it is the torque cell Wheatstone bridge assembly we are talking
about, not the torque cell only or the Wheatstone bridge alone, clear?
Now, it is most common method used for torque measurement. As I told you, this is a
very common method. It is widely accepted method of torque measurement and it
does not disturb, this is the most important thing, this, I mean this third point is most
important, I think this point is very important; it does not disturb the measured system
by introducing the friction torque in the same way as do the last two methods
8
described. Because you see in the last two methods, basically we are …., we are not
getting the actual torque truly. If you look at, because we are introducing certain
amount of torque, because we are introducing certain amount of frictions, so the, so it
will introduce certain amount of torque. So, we are not getting exactly the value of the
torque in that above measurement systems, right?
It converts to an electrical signal; it converts the torque to an electrical signal. It is

similar to the load cells. We are, we have studied the load cells, load cells it consists
of a mechanical element. Usually we have seen the load cells of the link, unit link
column type load cell. We have seen the beam type load cells. In all these different
cells we will find that that we are using strain gauges. It is similar thing; on the wall
of the shaft the strain gauges are to be installed. It is similar to the load cells. It
consists of a mechanical element, usually a shaft with a circular cross section and
sensors, usually strain gauges.
9
You see, this is our typical strain gauge, this is our typical, I mean measurement
systems. It is actually a shaft. We have taken a section of the shaft, so this should
actually go like this one, a section on the shaft and the four strain gauges are installed
on a 45 degree helix, right? Actually, if I say, it will look like this.
If I take a blank page, you see here actually this, if I take the shaft like this one, these
strain gauges are installed on a 45 degree helix. It will look like this one, right? So,
the four strain gauges here, these strain gauges you cannot see from this side. It is in
10
the other side of the shaft, so we cannot see this. So, that is the reason I have ….
dotted. So, this will give you the four strain gauges. Let us look at that.
Yes, you see here that this is our, you see that this strain gauges 3 and 4 you cannot
see. This is the 45 degree helix and you see here this is the 45 degree, this is the 45
degree angle. It is coming up. This is 45 degree. This is also 45 degree, 45 degree
helix. This is 45 degree, this is also 45 degree and these 3 and 4, you cannot see from
this side, right? It is on the other side of the, that is the reason we have drawn it
different with different colour, right and so, I should draw here, fine, strain gauges 1
and 2 and 3 and 4, right?
Now, let us look at, so this is the, installation of the gauge is very important, because
it will, if it is, I mean not in the helix, we can write, cannot write the, actually the
equations of this output voltage for the torque cell.
11
The four strain gauges are mounted on two perpendicular 45 degree helixes, as I told
you and the strain gauges 1 and 3 are mounted on the right hand helix and these are
under elongation. Please note these strain gauges 1 and 3 are mounted on a right hand
helix. These are under elongation. Strain gauges 2 and 4 are mounted on the left hand
helix and these are under compression. So, we have seen also in the previous cases,
when the two strain gauges are under elongation and two strain gauges are under
compression, we can put a …. in such a way that the, our b output will be, will be
quite large. If it is not four types it can be quite large, because all the strain gauges are
active.
You see that I can measure in all the previous cases also. In that lesson we have seen
that we can use the single strain gauges, but we are using multiple strain gauges to
maximize the output voltage, right. So, strain gauges 2 and 4 are mounted on the left
hand helix and these are the compressions and the strain gauges 1 and 3 are mounted
on the right hand helix and these are under elongation.
12
The shearing stress tau on the circular shaft is given by tau equal to TD by 2J, which
is equal to 16T by pi D cube. This is equation number 1, where T is the applied
torque, J is the polar moments of inertia of the circular cross-section and D is the
diameter of the shaft, right? If you put all these things, I can get the shear stress and
we assume that the, all the normal stresses are zero in these cases.
If the shaft is in the pure torsion, the normal stresses will be zero. Then, for a circular
shaft we can write sigma 1 and sigma 2, sigma 1 minus sigma 2 equal to tau xz. I
13
think this will be equal to, sorry, this will be, this is will be equal to, so applying
Hook’ s law, the principal strains epsilon 1 and epsilon 2 are obtained as epsilon 1
equal to 1 by E sigma 1 minus nu sigma 2 16T pi D cube 1 plus nu into E, right?
Epsilon 2 equal to 1 by E sigma 2 minus nu sigma 1 equal to 16T by pi D cube 1 plus

nu divided by E. However, delta R by R, I mean R equal to lambda. This is actually,
delta R by R is the change of resistance. So, as you know the definition, from the
definitions of the gauge factor, because as you know this is the gauge factor. Gauge
factor is actually defined as delta R by R by delta L by L; we have seen this thing,
right, so which will give you, so this is epsilon. This is equal to epsilon. So, I can
write that delta R by R equal to lambda into epsilon, is not it?
14
So, observing the installation of the four strain gauges, if you, we look at the
installation, because we have seen that the delta, I mean strain gauges 1 and 3 will be
under elongation. So, it will, sense the positive strain and strain gauges 2 and 4 under
compression, so it will, I mean sense the negative strain. So, what will happen? You
will find that the delta R 2 by R 2 will be negative with respect to delta R 1 by R 1.
So, we have written the equation, delta R 1 by R 1 equal to minus delta R 2 by R 2
equal to plus delta R 3 by R 3 equal to minus delta R 4 by R 4 equal to 16T pi D cube
multiplied by whole 1 plus nu by E minus into lambda. This is pi, please note and it is
not in math mode, so actually it is the pi like this one.
15
Now, the strain gauges are connected in a Wheatstone bridge, as shown in the
following figure with all the strain gauges. You see, this is a strain gauge. We have,
you see, this is, that I mean connecting the strain gauges in the Wheatstone bridge is
very important. So, that suppose if the very common question regarding the case of
link type load cell that, if we, what will happen if you, if you connect, if you
interchange between R 2 and R 3?
So, it will to be a problem, because in that case what will happen? You see, in that
case, R 1 and R 3 both will increase and R, R if I interchange, I am talking about if I
interchange R 2 and R 3 what will happen? You will see that in that case both R 1 and
R 3 will be under elongation, R 2 and R 1, R 2 and 4 will be under compression,
right? So, the unbalanced voltage will not be there, so this will create problem. So, for
that reasons we have connected like this.
16
Therefore, the unbalanced voltage of the Wheatstone bridge can be expressed as you
see here; e naught equal to 16 T by pi D cube into 1 plus mu divided by E lambda E
into E ex. So, what are the relations? You see, the relations are important, I am sorry.
This is the sensitivity of the torque cell. So, if I go and S equal to e naught by T equal
to 16 1 plus mu lambda E ex, you see, if you look at, these are, these, this excitation
voltage of the Wheatstone bridge, this young’s modulus of the elasticity, D is the
diameter of the, diameter of the shaft, mu is the Poisson’s ratio of the ( ) of the shaft
17
and lambda is a gauge factor of the strain gauges, whatever the strain, I mean gauge
factor we will use and so, you see that these are sensitivity that means for unit change
of torque, How much the output voltage I will get in the Wheatstone bridge? These
are sensitivity of the torque cell and the Wheatstone bridge combination.
Now, obviously what we will see? You see that, (( )) you will, you can increase the
sensitive by increasing the excitation voltage, by increasing the lambda. That means
increasing the, choosing the type of strain gauges which is, which are the higher level.
As you know, the lambda varies little bit, when varies from 2 to 4, usually, from the,
for the metal wire gauges; whether it is does not matter and if suppose, if you take
advance it will be lambda equal to 2. If you take (( )), it will be equal to 3 by 5 and if
you take (( )) and all these things, you, you will get the lambda equal to four point
something, 4.1 or something.
So, it is not, does not vary much, until unless if we use the semiconductor strain.
Semiconductor strain is not suitable for this type of operation, because large
vibrations and all those things are there. So, I can increase. So, lambda do not have
much choice. So, either 2 or 4 and in the excitation voltage, as I told you earlier also,
we cannot increase the excitation voltage, because it will increase the dissipation in
the individual strain gauges. So, now all the strain gauges has some limitation that
means how much power you dissipate.
It is, ultimately it is a resistance. So, obviously the how much power it can dissipate,
you must be calculate, right? So, depending on that you can optimize the value of the
excitation voltage in the Wheatstone bridge. Obviously, if I can reduce the D,
obviously the and E the young’s modulus velocity, your sensitivity it will be
decreased, but reducing D and reducing E obviously we will decrease the range of the
torque cell, right.
Now, see the sensitivity of the range of the torque cell will be given by pi D cube.
This is the T max, the maximum value of the torque pi D cube by S tau by 16, right?
So, if you reduce, you see here, if you reduce D, obviously what will happen that the
value of the, value of the maximum torque also will be reduced. If you reduce E, at
the same time S tau also will increased. So, obviously T max also will reduced. So,
18
we have to make the compromise between the sensitivity and the range of the torque
cell, right.
Now, where S tau, which was in the previous equation, it is the yield strength of the
material, right? The voltage ratio at the maximum torque can be given by e naught
upon E subscript ex maximum, will be expressed as follows is e naught by E ex max
equal to S tau lambda 1 plus nu upon E, right? S tau is the yield strength, lambda is
the, your gauge factor 1 plus nu. Nu usually varies, it, it is around .3 and E is the
young’s modulus of velocity, right?
So, it will give you the maximum voltage for a particular value of the excitation, what
can be the maximum voltage output for the particular value of the excitation, right?
So, we will show one example by which you will know that how much will be there?
Suppose for 10 volt power supply, how much will be the value of e naught, maximum
value of e naught, right? So, here as we can see that if increase the, if I increase,
decrease the, decrease the E or decrease the S tau, so obviously what will happen?
The maximum voltage also will decrease.
19
Now, we have taken an example. If S tau is equal to 4.14 into 10 to the power 8
Newton per meter square, Young’s modulus velocity, it is 2.068 into 10 to the power
11 Newton per meter square, where nu is .3, is a Poisson’s ratio and lambda equal to
2, then e naught equal to E ex, I can write maximum, equal to 5.2 millivolt per volt,
right? So, it will give you the 5.2 millivolt per volt. So, you can say that if I have a 10
volt power supply, so if I have a 10 volt power supply, so I will get a 52 millivolt
maximum output from the, from the load cell.
Now, data transmission is most important. It is not very important in the case of the
other cell like your link type load cell or the load cell I am talking about the link type
load cell or the beam type cell, because these are stationary, whereas in this case, you
see, all the strain gauges, the power supply, especially strain gauges is to be rotated at
the same speed, because these strain gauges is to be cemented on the shaft and it is
rotating at the same speed as the, as the shaft of the, shaft of the machine. So, it is
very much necessary to give the power supply, because if whenever we are using any
Wheatstone bridge, we have to give the power excitation voltage to the bridge as well
as you have to take out the unbalanced voltage from the bridge. Whether you are
using an unbalanced, it is, does not matter, so we have to take out the unbalanced
voltage from the bridge. So, there are two methods - one is the slip ring method.
20
Slip ring method is very common and other is the, other is the wireless telemetry
system. Now, slip ring method is convenient, because people work on the slip ring for
quite a long time and the telemetry method is rather new in that sense, right? Now,
you see here that this is basically a slip ring method of the torque measurements. We
have the same three strains. These are all installed on the torque cell. Please note these
are all 1, 2, 3, 4 are all installed on the strain gauges on the, on the torque cells; with
the usual cement and all these things you must install the gauges and we have the slip
rings.
These are the basically slip rings; you can see this is, are the slip rings; four slip rings
are there. So, through this I am giving the excitation voltage. So, this slip ring
excitation voltage is there and through other slip rings, through this, ….. the slip rings.
Because this part is rotating and this part is supposed to be stationary, this part should
be stationary and this part should be in rotating, right? Because it is on the shaft itself,
so through slip rings we have to take out.
Now, there is a typical problem in the slip rings, because you see the contact
resistance varies and as I told, you cannot increase E ex, because that will control the
dissipations in the, in the sensor itself. So, we have to limit the value of E ex. So,
when the contact resistance varies what will happen? The current will also vary, so
which will lead to the noises in the system. This is one problem with the cases of the
21
slip rings. We have to use some special materials and we have to use some special, I
mean some optimum forces we have to apply, so that the, there is no variation of the,
no more variations of the pressure or because, because ultimately it is a brush and
springs, so that you know variations in the pressure, so all these things you have to
look at, only then you will get the value of the e naught which is noise free, right?
So, this basically, you can use another slip ring also like this one; so you can use
another one. So, this should be in contact actually, it should be like this one, like this
one, please note, right? So, this will go like this one, so I can have all the four, I mean
I can have all. So, it is the current in, in each ring, right? If the current is large, so we
can, usually in the DC machines and all these things, we give the power like this, so
same principle is used, see, because we have the same technology we know, so based
on that we have developed these type of systems, right?
22
Now, the slip ring assembly contains insulated rings. Now, the details of the slip rings
method - the slip ring assembly contains insulated rings mounted on the shaft and the
brushes mounted in a case. The case will remain stationary, since high speed bearings
are placed between the shaft and brushes, right? The drawback, as I told you earlier,
the contact resistances between the rings and brushes vary. Thus the noise will be
developed in the excitation of the bridge and the bridge output, because the, both the,
you see, whenever taking, I am sorry that is we are talking not only about the bridge
output, even your excitation voltage will also contain noise, right? There will be,
further noise will be added when you are taking out the unbalanced voltage. That
noise will be already present in the unbalanced voltage.
Noise present means whatever the noise entering the, even though you are supplying,
giving the supply from battery, so those slip rings will introduce some noise which is
going to the, as excitation to the bridge and when you are taking out the unbalanced
voltage from the bridge also you will get the, I mean noise; it will be added rather.
23
You know special rings made of the copper nickel alloy and brushes made of silver
graphite mixture are used to reduce the noises, right? So, there are different special
types of, I mean slip rings we have to use, so that the, made of silver graphite and
mixtures which is used to reduce the noises. Now, next is coming to the telemetry
systems are that systems, which is or a wireless telemetry system, because there is no
wire. To avoid all these, I mean because you see that whenever these types of things
are there, slip rings has another problem which we have not discussed.
It is prone to very wear and tear after long rotation, after rotation for few days. When
the shaft is rotating continuously in the plant, so what will happen that there will be
wear and tear. All the brushes will be, will be, I mean there is a wear and tear in the
process, so you cannot avoid that. But instead, if I use a telemetry systems or wireless
systems, which is totally contactless, so that type of things will improve the, the life of
the entire measurement systems. So, I will discuss now the telemetry systems.
Telemetry obviously means the wireless telemetry. It utilizes a split collar that fits
over the shaft, right? There is a split collar, which fits over the shaft and a power
supply to bridge, a modulator, a VCO; VCO means voltage control oscillator and an
antenna are placed in the split collar. So, what are they placed? It is the power supply
to the bridge, a modulator, a VCO, an antenna are placed on the split collar. The
24
unbalanced voltage is passed through a PWM pulse width modulator that modulates a
constant amplitude 5 kilohertz square wave. You see, this is 5 kilohertz square wave.
Now, what is this? Actually you see that unbalanced voltage is coming. Let us look at
some of the, you see, what will happen? So, there are different modulation techniques,
as you know that we have, I mean pulse width modulation, we have binary phase shift
modulation, so this is one of the modulation techniques.
Suppose I have the signal here, sorry, is not it? This is a wave and we can have
another wave of, like this, yes or we can have another wave, right? Is not it? You see,
here in all the three cases the average value will not remain the same. In this case, the
average value will come down here, somewhere here. If I take a different pen, so
average value will come down, yes, average value will come down here, you see here.
In this case average value will come down here and in this case, average value will
come down here, isn’t it?
I do not know, it might be not very much visible, so I can take some other pen. You
see, the average value is changing in all the three cases, is not it? However, you see
this is the time period of your signal. If I take this is the time period of your signal T,
you see the T remains constant, right? So, what is varying then? If you can carefully
25
note, in communication system we call it mark. In a wave this is called mark and this
is called space, right? This mark to space ratio is getting changed, right?
If mark to space ratio is 1, we call it 50% duty cycle clock, right? If the mark ratio is
not 1, so we call it something else, not the 50% duty cycle. M is mark and this is
space, right? Now, here you will see that mark to space ratio is 1. Now, in the case,
this is the case 1. In case 2, we made M by S greater than 1 that is the average is going
up, right? In case 3, you look at very carefully; M by S is less than 1. In all the cases
you see, I am sorry, this cannot be the, your, fine, right? So, actually time period will
come and it will come down here. Sorry, this should go away. How it will go? I am
sorry, so it will come like this.
Now, see one thing very important that here, I mean you will see there. Let me choose
the pen again; here that mark to space ratio is varying. If the mark to space ratio
varies, then what will happen? Obviously, you will find that the, that the average
value of your signal also will vary. This signal, this average value of the signal we
feed to a VCO. Now VCO, as you know is a voltage control oscillator. It is nothing
but, the input is a DC voltage and the output will be the, output will be the frequency
and which are varying frequency. So, depending on the, your input voltage your
output frequency will vary. This is typically the VCO, voltage control oscillator. It has
always a free running frequency, right? Usually it is very quite high. So, we make a
free running frequency. Suppose 10.7 megahertz or something like that, right and
along with that we can, we can do that thing, right?
So, this is actually, so if I choose this one, so it will go like this one, right? This is our
time period; this is actually the time period. You see the time period is not varying in
all the cases. You see, if I take this one, so time period is not varying, right? So this is
again time period and this is the time period here, right, isn’t it? But the mark to space
ratio is getting …, that is actually the PWM, pulse width modulators. I am modulating
the width of the pulse by which my signal will be, because you see, the DC signals
which I am getting that is a very low value, so we cannot transmit it; I mean the DC
signals cannot be transmitted, it is to be converted to the AC signal. That is the reason
we are making it pulse width modulated.
26
Once we got the pulse width modulations, we now again we give it to the, again we
are taking some DC signal here. Then, we are giving it to the, some the voltage
control oscillator, right? So, it utilizes the split collars and which I was discussing and
the amplitude of unbalanced voltage is passed through a PWM that modulate a
constant amplitude 5 kilohertz square wave. So the basic square wave is 5 kilohertz.
That means what is that?
That means what we are discussing that means if I take a, so this signal, so this is T.
So, f is equal to 1 by T equal to 5 kilohertz, right?
27
Now, if I go back, now square wave will vary the frequency of 10 megahertz of the
VCO, right? In the square wave, actually we cannot ….. smoothen it. We will convert
it into DC, I mean then it will convert, like what will happen?
You will see here that actually if I look at, so we have the unbalanced voltage which
is coming from the strain gauge or I should say it is a Wheatstone bridge. So, it looks
like this that is I have a Wheatstone bridge. Then, it is coming to the, this is the DC
voltage. This is the DC voltage; it is coming to a PWM. Then, we are coming to a low
28
pass filter, because you see, this I mean signal, I mean which is coming from the
PWM depending on the, this is to be converted, passed through a low pass filter to
average it or the signal can be like this, like any thing. Then, it is coming to a VCO.
Now, here this, the frequency which you are, we are modulating, it is 5 kilohertz.
What is 5 kilohertz? That means this frequency, that means if I say in time periods, so
T or f equal to 1 by T equal 5 kilohertz, whereas VCO frequency is quite high. It is
around 10 megahertz or something like that, right, so that this will control over this
one that means some DC voltage will come down here, which is given to the output.
So, the output voltage of the VCO will be proportional to the DC voltage we will give
and this DC voltage again will depend on the unbalanced voltage of the Wheatstone
bridge, clear?
Now, this square wave will vary the frequency 10 megahertz of the VCO. The VCO
signal is transmitted through the rotating antenna. Everything is rotating, I mean VCO
signal is transmitted through the rotating antenna and the signal is received by a
stationary loop antenna that encircles the spit collar. It is a collar like this one. See, if
you look at, it is, it is very difficult, so we have a shaft and it, it is split like this one.
So, it will be put like this one. Split, we can split it, so there is another antenna here
on the outside and it is stationary. Inside, it is rotating. There is another antenna in the
rotating shaft also, right?
29
Now the, how the, how I will give the power supply to the bridge? Next, that is;
So, signal is received by stationary loop antenna that encircles the split collar. So,
there are two antennas. One is rotating antenna, which is installed in the shaft itself
and that antenna is rotating as the same speed as the shaft. So, antenna, your
Wheatstone bridge, power supply, everything is rotating and stationary loop antenna,
the signal received by the stationary loop antenna. Interestingly, you will see that the
same stationary loop antenna can be given for both the excitation, given to the bridge
as well as taking out the unbalanced voltage. Obviously, the frequency will be
different, so there is no question of overlapping.
We can, at the receiving end we can have the band pass filter. The power supply to
the bridge you see here, the power supply to the bridge, PWM pulse width modulator
and the VCO, because we are assuming that the low pass filter is incorporated within
the pulse width modulator and the VCO is fed through inductive coupling by a 100 to
150 kilohertz signal transmitted by the stationary loop antenna. There is no battery we
are providing as such, because Wheatstone bridge needs some excitation, isn’t it, so
that entire, all those, I mean chip excitation, all everything, we are giving through a
stationary loop antenna, which has a frequency of 100 and 150 kilohertz and the other
frequency, the VCO frequency it is quite high. It is 10 megahertz.
So, if you are at the receiving end on the, if you have two different band pass filters of
the appropriate bandwidth, we can easily separate these two signals or you have, at
the receiving end if you have a low pass filter, obviously this signal will go out that
means the excitation signal will go out. We will receive only the signal, which is
coming out of the, coming out of the bridge which is very high frequency around, it
is, it, we have a 10, 10 megahertz is the frequency. So, around 10 megahertz, because
when it depending on that you see it is, what will happen?
30
You see here that I have a 10 megahertz signal. So, over this there is a overlapping, so
I will get a signal range like this. So, this is 10 megahertz. So, this is minus delta f,
this is plus delta f, I will get, so over this, depending on the unbalanced voltage of the
Wheatstone bridge I will get. This is the centre frequency of our VCO. This is the
VCO centre frequency, right? So, this is 10 megahertz signal, so over this I am getting
shift, right?
So, on both side you will get signal of, depending on the, whatever the and this, how
much the shift will be, because since it depends on the DC voltage, the input to the
VCO DC voltage, how much is the shift, it will also depend on the mark to space ratio
of the PWM and the mark to the space ratio of the PWM will depend on the
unbalanced voltage of the, your Wheatstone bridge. So, the frequency shift around 10
megahertz of the VCO will depend ultimately on the, on the, on the signal which is
coming from the unbalanced voltage, right? ,
Now, interestingly, you see the, usually how much this is? Suppose this is 10.5
megahertz, your unbalanced voltage, a frequency of the VCO corresponds to
unbalanced voltage and what is the frequency of the …which is given to the stationary
loop antenna? Suppose 150 kilohertz, so these two frequencies are totally different.
We can easily filter out ….These two frequency, we can filter out by a low pass filter.
31
If I, if I, if I use the high pass filter, obviously through this frequency we will get and
this frequency will be chopped out, clear? Because, our information about the
unbalanced voltage, because this unbalanced voltage or the shift of the frequency
from the VCO, from the VCO centre frequency will be calibrated in terms of torque,
please note that.
So, this frequency will be calibrated. This is, the shift of the frequency will be
calibrated in terms of torque. Ultimately, if the larger torque, the frequency shift will
be large; if the smaller torque, frequency shift will be small around 10 megahertz,
right? So, this frequency change is to be calibrated in terms of the torque and if I get
this signal by a high pass filter, simply if I can reduce the, if I can chop off this 150
kilohertz signal, so I will get 10.5 megahertz around that thing, fine, no problem. So, I
will get that signal. I will calibrate the, my meter according to the torque, right?
So, this is all about the power supply, all these things and this almost we came to the
end of this, I mean lesson before we are going to the some exercises, I will give you
the exercise and I will give you the solution also. But please note I will give the
answer, but not the solution as such, so you should compute of yourself. This is
nothing but, plugging in your all the expressions and you will get the answer. It is just
nothing great deal, but to know that what is the typical range and all these things, just
to bring the flavor of the or what is the torque measurements, what is the typical range
and all these things. You see here, let me take a blank page. The problem is like this.
32
Problem 1, determine the sensitivity of a torque cell if modulus of elasticity of the link
material, E which is modulus of elasticity of the link material is 2.0684 into 10 to the
power 11, 10 to the power 11 Newton, say Newton per meter square. Then, Poisson’s
ratio nu of the link material is .3, excitation voltage that means E ex is equal to 8 volt,
diameter of the shaft D is equal to 2.54 centimeter, right? The gauge factor of the
strain gauge, lambda is equal to 2; it is the advance strain gauge, obviously and the
resistance of the gauge, resistance, all are active gauge; resistance of strain gauge is
equal to 120 ohm. So, it is nothing, I mean to, just you plug in all these formula, so it
is 10 to the power 11. Please note, 2.0684 into 10 to the power 11, right? So, you
plug-in all these value, you will get the sensitivity, because we have already given the
expression for the sensitivity in our class, right? So, answer you will get. I will give
you the answer.
33
Answer you will get is, three point, just plug in all these values, 125. In which
equations you have to plug in? Because what I ask you? I ask you to find the
sensitivity. So, if I look at, so if I, the sensitivity expressions you see, our sensitivity
expression will look like this. S is equal to e naught by T equal to 16 1 plus nu lambda
E ex pi D cube E. All data’s are given, is not it? So, what, in what sense? You see
here, E is given, E is 2.0684 into 10 to the power 11, E ex is 8 volt, this is 8 volt. This
is 2.0684 into 10 to the power 11. D is given. D is given; although it is given in
centimeter, it has to be converted into meter. So, it will be 0.0254 meter. Then,
Poisson’s ratio is .3, lambda is 2.
34
So, if you calculate, so actually the value of S will come up as, S will come up as S
equal to 3.125 into 10 to the power minus 5 volt per Newton meter, right? Voltage by,
voltage by torque, so obviously it will, it will come like this. Now, interestingly the
second part is telling the, let me write the second part.
The sensitivity of torque cell can be increased if the input voltage or excitation is
increased, right? That you have seen, is not it? You have seen that if you increase the
exaltation voltage our sensitivity also increased, because in sensitivity expressions, we
35
have seen that it is in the numerator, is not it? All the sensitivity information, we have
seen the sensitivity expressions it is in the numerator, is not it, because sensitivity if
you look at, it is this 16 1 plus nu lambda E ex upon pi D cube E, is not it? So, if I
increase E ex obviously my sensitivity will all increase, right? Voltage E ex is
increased, so that is given as it is.
So, that means if I take this one, so it is increased. If each gauge, each gauge can
dissipate, each gauge means each strain gauge, it dissipates .8 watt of power, power,
determine the maximum sensitivity that can be achieved, that can be achieved without
damaging the strain gauge? You see, know I always the strain gauge has some
limitation that means how much power it can withstand? So, that is already given..8
watt of power it can pass. So, accordingly you have to, must calculate the current.
Once you get the current you know the voltage. So, once you get the voltage you can
calculate the, I mean sensitivity, right?
36
So, the sensitivity here in this case will be S equal to, S max will be given by 7.5 into
10 to the power minus 5 volt per Newton meter, right? So, with this, I come to the end
of this lesson that is of torque cell.
37
Welcome to the lesson 7 of Industrial Instrumentation. In this lesson, basically we

will discuss the thermistor. As you know, thermistor is a temperature sensitive device,
I mean if you increase the temperature, its resistance decreases and by using this
property, mainly it is used as the temperature sensor, but it has many other
applications in electronic circuits that we will discuss one by one.
Now, contents of this lesson.
38
Its resistance temperature relationship or characteristics, composition of a thermistor;
because it is a semiconductor material, so you have to look at what are the
compositions and how it is made? Manufacturing process, how actually a thermistor
is made and thermistor in a Wheatstone bridge. This is very important. We will show
that Wheatstone bridge is used to take an unbalanced voltage which is calibrated in
terms of temperature and its linearity of operation.
We will also see the thermistor in a potentiometer circuit. Either you can use it in a
Wheatstone bridge or in a potentiometer circuit, we will see both. We will analyze
both and what is the linearity of operation? Thermistor is very non linear sensor, we
will look at all these things.
Thermistor also can be used in a potentiometer circuit, right? We will see there are
various potentiometer circuits, used in the potentiometer circuit as follows. You see,
this is the, one of the thermistor which is used in the potentiometer circuit, right? You
see here that this, this should be the sign. So, this is the thermistor R T and R 1. In this
case, if R T equal to R 1, so R T dash will be equal to R T that means changed
resistance, where R T dash, as I told you earlier, will be equal to R T plus delta R T. R
T dash equal to R T 1 minus 2 e naught by E ex, excitation upon 1 plus 2 e naught by
E ex.
39
Now, interestingly the one common question arises. What will happen if I flip it? That
means if I put R T here and R 1 here that means the circuit, if the circuit looks like
this, then what will happen?
This ends the lesson 7 of Industrial Instrumentation.
40
Prof. Alok Barua
Lecture - 7
Thermistor

will discuss the thermistor. As you know, thermistor is a temperature sensitive device,
I mean if you increase the temperature, its resistance decreases and by using this
property, mainly it is used as the temperature sensor, but it has many other
applications in electronic circuits that we will discuss one by one.
Now, contents of this lesson.
Its resistance temperature relationship or characteristics, composition of a thermistor,

because it is a semiconductor material, so you have to look at what are the
compositions and how it is made? Manufacturing process, how actually a thermistor
is made and thermistor in a Wheatstone bridge. This is very important. We will show
that Wheatstone bridge is used to take an unbalanced voltage which is calibrated in
terms of temperature and its linearity of operation.
1
We will also see the thermistor in a potentiometer circuit. Either you can use it in a
Wheatstone bridge or in a potentiometer circuit; we will see both. We will analyze
both and what is the linearity of operation? Thermistor is very non linear sensor; we
will look at all these things.
At the end of this lesson, the viewer will know thermistor, its, all its characteristics. It
is used in the, how it is used in the electronics circuit, what are the signal conditioning
circuitry associated with the thermistor along with its application as a temperature
measuring element. This is the main application, but we will touch some of the
applications, even though it is not within the perview of the industrial instrumentation
like in the power supply circuits, in a Wien bridge oscillator circuits, there are some
uses of thermistors in those applications also, as I said other application of
thermistors.
2
Now thermistor, it is a temperature sensitive resistor. As you know the, all the
resistors are temperature sensitive. If we increase the, normally if we increase the
temperature, resistance increases. However, unlike metals, it shows a decrease in
resistance value with the increase in temperature. This is a unique feature which we
have in thermistor, which no other, I mean resistors have this type of property. This is
inverse, I mean its, its resistance is not increasing with the increase of temperature;
resistance decreases with increase of temperature; rather I should, I mean if I draw it,
it will look, the characteristics will look like, like this. You see that, suppose if I draw
here, temperature here and resistance here, the relationship is exponential like this
one, clear? This is the temperature versus resistance relationship of a thermistor.
3
It is made of oxides of nickel, cobalt or manganese and sulphides of iron, aluminium

or copper.
The resistance temperature relationship can be expressed as R equal to R naught

exponential to the power 1 by T minus 1 by T naught. What are those legends? R
equal to the resistance of the thermistor at T Kelvin, R zero is the resistance of the
thermistor at T naught Kelvin and beta is the material constant that ranges from 3000
4
to 5000 Kelvin. In fact actually, I am sorry, there is a mistake, so it will be beta
multiplied by this; the beta multiplied 1 by T minus 1 by T naught.
This is the thermistor, because this is unit less quantity that should be, so beta is a
material constant that ranges from 3000 to 5000 Kelvin. Now, one thing is very
interesting. You see that as the value of beta, always we desire that the value of beta,
it is not very difficult to find the value of beta, because I can measure the temperature,
I can measure the resistance at two known temperatures. So, obviously I, from that I
can find the value of beta. Now as the beta is higher, always people prefer beta to be,
beta should be higher.
The reason is that if the beta is higher, we will see that if the beta is higher, the
relationship will be sharper and sharper and beta increases in this direction, beta
increases in this direction. As we can see, if the beta increases, sensitivity that means
this temperature versus resistance relationship, sensitivity of the thermistor also
increases. That means for the same rise of temperature, I will get higher decrease of
resistance. So, that is always an advantage for making any instrument that the
instrument should be more and more sensitive, clear? I should go back, as I told you
this should be, beta should be multiplied, right?
5
The sensitivity S of a thermistor is given by S is equal to delta R by R by delta T. If

you can derive, then I will get minus beta by T square. If beta, for an example if beta
is equal to 4000 Kelvin and T is 298 Kelvin, S, we will find the S, the value of the S
to be equal to minus 0.045 per Kelvin and this is much higher than the sensitivity of
Platinum RTD. Platinum RTD is not that this sensitive, because both are resistance,
resistive sensors.
As you know, RTD means basically resistance temperature detector and I mean, if
you look at the sensitivity of the RTD, it is much higher. So, it helps to measure the
very small difference of temperature.
6
Now, manufacturing process if you will look at, two or more semiconductor powder
are mixed with a binder to form a slurry. Small drops of the slurry are formed over the
lead wires, dried and put in a sintering furnace. During sintering, the metallic oxides
shrunk onto the lead wires and form the electrical connection.
The beads are then sealed by coating it by glass. It looks like this, actually. So, two
wires we will take. I should, I will draw it here. I should draw it here, so two wires are
7
there. It is coming out, so the slurries are dropped on this one. Now, you put in a
centering furnace, so I will get a bead like this one, right? So, what will happen? So,
this will, sorry, this again I will draw it here. Let me erase first, so I can, so the
slurries are dropped on this one. So, it is put on a sintering furnace. Now, if you put in
a sintering furnace, what will happen? You know it will, it will shrink and this
shrinking process will make a very good electrical connections. Now, please note that
the thermistor are very sensitive to the, since it is semiconductors device, it is
sensitive to moisture also. So, there should be some protection, so that it will not
deteriorate with the moisture. Performance totally deteriorates with the moisture,
right?
The glass coating improves the stability by eliminating the water absorption. So,
obviously if we make a glass coating, so water cannot be absorbed there. Typical size
of a thermistor, it is 0.125 millimeter to 1.5 millimeter. As usual we compare the three
basic temperature sensors - one is thermocouple, then RTD, then thermistor. We will
find that the thermistor has the smallest size, I mean some of the size is so small, it is
sometimes very difficult to even to look at the naked eye. It can be made so small.
Now, making this small and there are that, there are two advantages of making such a
small size. Even though we have to put on a glass coating and all those things, first of
8
all its time constant of the thermistor is very, very small. If you compare to
thermocouple and RTD, the time constant of the thermistor is very, very small that is
the great advantage and making in this small size also will allow you to measure the
temperature in a very, I mean in a narrow location, very small location, where there
no other sensor can go inside, right?
Shape, it has various shape. It can be available in disks, wafers, flakes as well as in
rods. These are the different shapes of the thermistor.
Resistance of the thermistor may vary from few ohms to several kilo ohms. Usually, I
mean as you know, the thermistor, one application as I told you that we will find it
has, in electronic circuits also, we use it. Now, resistance of the thermistor may vary
from the few ohms to several kilo ohms. Now, one thing you should be very careful
about the minimum resistance, because the, as the temperature increases its resistance
drops, its resistance falls. So, it should not fall in such a way that it will make a, either
it will draw a huge current from the voltage source, from battery that means
prevention of short circuit should be there and also it should not load the subsequent
instrument, because if you have the very low resistance, obviously there is a chance of
loading the subsequent, I mean the signal conditioning circuit that is to be also
avoided.
9
Other semiconductor, conductor temperature sensors include the carbon resistors,
silicon and germanium. These are the some other, I mean sensors, even though which
are not much in use.
Now, silicon with varying amount of boron impurities can have either a positive or
negative temperature coefficient of resistance. So, you can have either positive
thermistor or a negative thermistor. We are not interested to have a positive
temperature coefficient, because we have many other sensors. But we need negative
temperature coefficient that also with a very high sensitivity, which is available in th
case of thermistor.
Now, germanium doped with arsenic, gallium is used for cryogenic temperatures,
when it shows a large decrease in resistance with increase of temperature. Because
this you know, cryogenic instrumentation is one of the very important instrument,
instrumentation especially in the measurement of temperature of the liquid helium and
all those things. We will find that in such a case of, because many a times it is
necessary to measure all these super conducting circuitry and all these things, we have
to measure the temperature. So, the cryogenic temperature measurement is, is again
another challenge. So, in that type of situations, I can apply the thermistor and it is
germanium doped with arsenic, gallium and it is used for cryogenic temperatures,
10
where it shows the large decrease in resistance with increase of temperature and
applications in electronic circuits.
It limits the large charging current of power supply filtering capacitor that we will
show later on at the end of the lesson that how it can be controlled. This, it has also
application of stabilizing the Wien bridge oscillator circuit. It is also used for
stabilizing our, I mean …. point stability of the, our analog amplifier circuits, I mean
transistor amplifier circuits. So, all these applications we should show later on.
Now, please note that the, in the case of thermistor, if I take a white page, the, it is
actually available; those are, because there are two types. Its resistance is drawn like
this one and in electronic circuit when they are using, we are using, making a dot like
this one. This is the case of thermistors, I mean dots, which is used in an electronic
circuit. In that type of the cases, it is available in the form like this one, disk like, so
two wires are coming out and it is colour coded. That means it happens, you see that it
looks like this.
So, colour code is put here, so that people will know the, what is the, at the room
temperatures or at some calibrated, particular calibrated temperature what should be
the resistance of the thermistor, right? Resistance is not very constant, as you know
11
that, because it is sensitive. Even though in the case of, suppose the carbon resistors
when you buy a, when you buy a carbon resistor there is a colour code on the carbon
resistor. So, obviously that colour code, I mean that value depends, I mean value
actually refer to some particular room temperatures. Whereas, in the case of
thermistor, the temperature is very important, because it is varying a large; it will vary
a large at the increase of temperature. So, colour code is there; whereas, in the case of
thermistor switch it is used in as a sensor, you will find it is like this one on a glass.
So, it is glass coating, so they put the thermistor there, right and wire is coming out,
right?
If I go back, now thermistor can be used as an active element in a Wheatstone bridge,

because you know, it is a resistance change. Ultimately the, if I want to measure at
any temperature, so what will happen that you have to calibrate the output voltage in
terms of temperature, if we use the bridge as an unbalanced, I mean if you use bridge
in unbalanced condition. So, I have to make the, that unbalanced voltage should be
calibrated in terms of temperature. That is actually done in the case of, if we use it as
a temperature measuring device.
There are various, I mean, I mean you can use a voltage to frequency converter and
frequency can be calibrated in terms of temperature. There are various …. people
12
tried with so many alternative. They have tried also with a thermistor as a frequency
changing elements of the circuits. Otherwise, as you know, very standard features that
unbalanced voltage of the Wheastone bridge can be given to a v to f converter, so that
the frequency will be directly calibrated, can be directly calibrated, which can be
measured by some frequency measuring devices and that can be calibrated in terms of
some frequency measuring circuits that frequency can be calibrated in terms of
temperature; that is also possible.
By all these means, actually people are doing all these different, I mean methods with
just to increase the resolution to increase the sensitivity, eventhough it is a quite
sensitive device. But in some situations, where the fraction of the temperature change
of measurement is important, in that case you have to think of that about change of
frequency with the increase of temperatures and all those things.
Now, this is a Wheatstone bridge, we have seen and you can see this one. You see,
here these are thermistors, so I can put a dot here. If I choose a yellow pen, so I can
put a dot here. So, it is thermistor. R’s are just simple carbon resistors. These are
bridge excitations, E subscript ex and this is our output voltage. Now, if the, initially
the bridge is balanced, then what will happen? Suppose all the R T, R 2, R 3, R 4, all
13
are same, all are same, then bridge will obviously be balanced at room temperature,
suppose.
Now suppose there is a temperature change in R T, so obviously the resistance value

of the R T will change. So, if the resistance value of the R T decreases, I will get an
unbalanced voltage. Previously unbalanced voltage is zero, now we get some
unbalanced voltage. That unbalanced voltage can be calibrated in terms of resistance
change and that resistance change, unbalanced voltage also further can be calibrated
in terms of temperature. This unbalanced voltage, I can feed to a, this voltage I can
feed to a v to f converter and this, a frequency measuring circuit, a counter or
frequency measuring circuit, right?
What it will do? So, it will simply do, it will measure the frequency and the frequency
will be calibrated in terms of temperature degree centigrade. The advantage of this
type of thing is, as I told you earlier, the resolution; sensitivity also can be increased,
because I have the signal conditioning circuitry, where I can make all this, all these
type of manipulation, so that I will get the frequency change more sensitive.
Now, if the Wheatstone bridge is initially balanced, we assume that the Wheastone
bridge is initially balanced, even not, not necessarily that the, all the resistance will be
14
equal. If this condition is valid that R T into R 3 equal to R T into R 4, then what will
happen is, you can see if R 2, R 3, R 4 are fixed value resistors, then the output
voltage e naught due to change in resistance delta R T for change in temperature will
be derived as follows.
If we redraw, let us redraw the Wheatstone bridge again. So, we will find, so we have
redrawn the circuits, we have labeled something a b c d. We have taken in, I mean
clockwise direction. This is R T thermistor, then R 2 carbon resistors, R 3 fixed
resistors. So, I should rather say that these resistors, R 2, R 3, R 4 is temperature
independent. That means if the temperature varies, this resistance, it is, their
resistance does not change, fine. We can accept that thing.
15
Now, e ab you can simply see that it will be equal to R T upon R T plus R 2
multiplied by E ex, because this is our excitation, isn’t it? This is our excitation
voltage. Then, e ad will be equal to R 4 upon R 3 plus R 4 multiplied by E ex, where
ex is the excitation, right? Now, the output voltage e naught equal to e bd equal to e
ab minus e ad equal to R T upon R T plus R 2 minus R 4 upon R 3 plus R 4 multiplied
by E ex, right?
So, again this is now, so this is our e naught. So, this is our e naught equal to this
voltage, fine, no problem.
16
Due to change in temperature, the new value of the resistance is R T. It is no more R

T, so it is R T plus delta R T; whether delta will be positive, negative that I am not
thinking, that we will discuss later on. There is a change and we said that the R T plus
it is delta R T. If it is there, then you see that I can write that e naught, the new value
of R T, let me take this one, R T plus delta R T, because it is a new value of R T.
What is this new value? This is due to temperature change. R T is replaced by R T
plus delta R T in these equations.
So, R T plus delta R T R 2 R 4 upon R 3 plus R 4 multiplied by the excitation E ex. If

I make all these cross multiplications, I will get, in the denominator I will get the R T
plus delta R T R 2 R 3 plus R 4. Numerator I will get R T R 3 plus delta R T R 3 plus
R T R 4 plus R 4 delta R T minus R 4 R T minus R 4 delta R T minus R 2 R 4. Now,
interestingly you see that what will happen that here these will cancel out, obviously.
If I take out, it will be easier. You see that this and this will cancel out, R T, R T R 4,
this and this will cancel out, isn’t it?
Then, R 4 delta R T and R 4 delta R T also will cancel out. Now, R T R 3 according
to the conditions, this and this will cancel out, is not it? So, if I look at, so what will
remain? It will remain delta R T into R 3. So, this is the only factor which will
17
remain. In the numerator R 1 R T plus delta R T plus R 2 multiplied by R 3 plus R 4,
right?
Because the condition is R T R 3 equal to R 2 R 4 that is the reason the first term and
the last term was canceled out. So, ultimately I will get the expressions e naught upon
excitation E ex will be delta R T into R 3 R T plus delta R T plus R 2 the whole
multiplied by R 3 plus R 4.
18
Now, assume that the, if R 2 equal to R 3 and R T equal to R 4, the equation 1
becomes, you see R T delta R T by R T the, I mean upon 1 plus delta R T by R T plus
R 2 by R T 1 plus R T by R 2. If you just, in the numerator it will remain same; it will
remain same in the numerator. So, in the denominator you can multiply a 2 will come,
because you see here this R 2 R T and R T R 2 if we multiply, we will get another
one. So, this will make 2, then R T plus by R 2 R 2 by R T plus delta R T by R T plus
delta R T by R 2. This is equation number 2, right?
So, finally I will get an equation. For the thermistors you see, the interesting term is
that in the, in the last equation, equation number 2 that the terms delta R T by R T and
delta R T by R 2 in the denominator are not small enough with respect to the other
terms. Therefore, they cannot be neglected to simplify the solution of the equation for
delta R T by R T. For special case of an equal bridge R T equal to R 2 equal to R 3
equal to R 4, rearranging equation 2 we get, e naught by E ex equal to delta R T by R
T upon 2 plus R T by R 2 R 2 by R T plus delta R T by R T delta R T by R 2.
19
If you do these things, so ultimately I will get, because if the R T equal to R 2, we

have assumed already, so I will get an expression, we have assumed that the R 2 equal
to R 3 and R T equal to R 4. So, I will get an expression which looks like, we have
assumed R 2 R T R 2 R 3 R 4 are same, so I will get an expression like this one. So,
this will lead to delta R T by R T 4 plus 2 multiplied by delta R T by R T, right? So, if
I make a little arithmetic manipulation, so this will be actually, this will be minus
sign; we have, minus sign should be there, 4 e naught by E ex 2 delta R T by R T
minus e naught by E ex is equal to delta R T by R T, right?
20
So, delta R T by R T 1 minus 2 e naught by E ex equal to 4 e naught by E ex. So,

delta R T by R T equal to 4 e naught by E ex 1 minus 2 e naught by E ex, equation 3.
Now, the thermistor resistance R T at any temperature T is then given by the simple
expression. Actually this is wrong, mistyped, so this will be equal, right? This will be
equal sign that means R T dash equal to R T plus delta R T, right?
21
So, I can write that if I take R T common, so R T 1 plus delta R T by R T, so this is

equation number 4. Substituting equation 3 and equation 4, we get R T dash equal to
R T 1 plus 2 e naught by E ex 1 minus 2 e naught by E ex. This is equation number 5,
right?
The value of R T obtained from the equation 5 is converted to a temperature by using

tables that list T as a function of R T for the specific thermistor being used, right? So,
22
we will get a table. So, we can find it; we can make a calibrator and as you know,
ROM and you can make calibrated ROM for also this type of purpose. So, if the,
ROM, the advantages is those linearity problem, nonlinearity problem can be easily
eliminated in the case of thermistor.
Even though it is an excellent device, very sensitive, very small in size, I mean you
can, it can, it can be used for the dynamic temperature measurement, everything is
there, but the problem is that non linearity. Even though thermocouple is also
nonlinear, but the advantage of thermocouple, for a short range you can consider as a
linear sensor. I am sorry.
Now, thermistor also can be used in a potentiometer circuit, right? You see, there are
various potentiometer circuits; also used in the potentiometer circuits as follows. You
see, this is one of the thermistors which is used in the potentiometer circuit, right and
you see here that this, this should be the sign. So, this is the thermistor R T and R 1. In
this case if R T equal to R 1, so R T dash will be equal to R T. That means change
resistance, where R T dash, as I told you earlier, equal to R T plus delta R T. R T dash
equal to R T 1 minus 2 e naught by E ex excitation upon 1 plus 2 e naught by E ex.
23
Now, interestingly one common question arises. What will happen if I flip it? That
means if I put R T here and R 1 here that means the circuit, if the circuit looks like
this, then what will happen?
If I take a white page that means suppose I have an excitation, let me take, first erase
this one. That means if I take a, that means actually our circuit, we got that like this
one, is not it? Thermistor here, R T we have connected here. This is excitation E ex. I
am connecting this output voltage, this is R 1, right? Instead, what will happen if I
make the circuit like this one? R 1, I will put the thermistor here. You remember in
the previous, some, some time back we have discussed the instrumentation engineers
must consider the loading effect of the, of the thermistor in case of temperature. Why
it is not used? You see, like this one, is not it, here.
Now, what will happen if the temperature rises? If the temperature rises, you see the,
the thing will happen, this resistance will fall and fall, right? So, the impedance of this
resistance, of the impedance from looking from this side will be different, right? In
some cases it will be, some value in some other cases, so it will make the problem in a
subsequent signal conditioning circuit. Then, if we have a signal conditioning
circuitry here, so that will, designing that type of circuitry will be different.
24
Whereas, if you make the circuit like this one, you see the, I will get a, some
impedance R 1 always same, right? So, that is the reason instead of connecting like
that I put a thermistor here, so this circuit should not be used. So, you should use this
circuit. So, what will happen if the temperature changes? This resistance will fall and
we will get more current through this resistance, which is in series and I will take the
output voltage.
So, obviously output voltage also will increase, because if the current is I, I 1 suppose,
so I will take the current e naught equal to I 1 by R 1. If R 1 increases, R 1 decreases
obviously, sorry, the I 1 increases because if the resistance falls, if the resistance falls
obviously I 1 will increase. This value will be, I mean value of the voltage will
increase. So, as the temperature increases, the e naught will increase, right because
this current will increase, because this has fallen down, right?
So, that is the reason we have got the expression R T dash equal to R T 1 minus 2e
naught upon E ex excitation upon 1 plus 2 e naught by E ex, right? This is one of the
circuits.
25
Some other circuits to improve the linearity of operation are as follows. You see,
some of the circuits we have shown. These various circuits to influence the
impedance, output impedance and all those things, these are the circuits; we will see.
So, this is the thermistor here. Instead they have put a circle; instead of putting a dot,
they have put a circle like this one and this is another circuit. So far, the thermistor as
we increase the temperatures I mean sensing devices, we have seen that, we can also
prove analytically that whether you use it in a bridge or in a potentiometer, its
sensitivity, resolution does not, I mean vary much. It does not vary at all.
But, however we will find that if I, if we use a sensor like this one, I mean if we use a,
suppose a circuit like this one, I will get always some advantages. I will show you
what is the advantages?
26
This is another circuit of the thermistor, which is used in a bridge circuit. You will see
here, there is typical reason why we have connected a resistance in parallel with the
thermistor. Now, as I told you, thermistor is a nonlinear device. So, linearity is a great
problem with the thermistor, so what they are doing? They always, there is a common
thumb rule that you use a resistance in parallel with the thermistor, right? So, what
will happen you see?
27
So, I have a thermistor. Any circuit, we have seen two circuits; so, one is a circuit like
this one, right? So, this is a thermistor and like this one we have seen, right? I can take
the output voltage from this position or we have seen a bridge circuit, excuse me. So,
we put a bridge, we put the resistors here, right? So, this is R T. So, I put a resistor in
parallel and this is R 2, R 3, R 4, clear?
Now, any resistance, if I take a new page, whenever you are connecting a resistance to
a thermistor, you know what will ………, you know what will happen to its
characteristics? Suppose I have a characteristics like this one, temperature versus
resistance characteristics of a thermistor, it looks like this, depends on the value of
beta, how sharp it will come down. As I told you in the beginning that the higher the
value of beta more sharper and sharper will be the fall that more and more will be the
sensitivity of the thermistor; all of us want that the sensitivity should be more.
Now, if you connect the resistor in parallel with the thermistor, you will lose first of
all sensitivity and if you lose the sensitivity, it will look like this one. You see, it will
look like this. For some small range, I can use the thermistor almost as a linear device,
if I use a thermistor and with the resistance in parallel. So, it won’t be fully linear. It is
not possible, because it is already a nonlinear device. So, it is exponential fall, but if
28
you use a resistance in parallel, for a small range it will become linear. So, by
sacrificing the sensitivity I am making the circuit or thermistor linear.
This is my circuit, as we discussed.
Now, when the, when thermistors are used to measure the temperature, the errors
resulting from the lead wires effect are usually small enough to be neglected even for
29
relatively long lead wires. Now, this is the typical problem of any temperature
sensitive devices. When you are making a temperature sensitive sensor, temperature
sensitive sensor, these are the typical problems, because the, I mean resistance of the
lead wires, because you see that you cannot, I mean put, suppose I have a, you cannot
put like this one.
That means suppose I have a, what I am saying is, suppose I have a, suppose I have a
furnace here, you, you cannot put, a thermistor might be inside; you cannot put the, I
mean you cannot put inside that very near to this meter. So, you have to take out.
Suppose if I have a furnace here, so you have to take out the wires and might be here
we have the meter. Here we have the sensors. Suppose these sensors are inside, so this
is the lead wires. It can be few meters also, right? It can be few meters also. All
temperature measuring, whether it is a thermocouple, thermistor and I mean or RTD,
it does not matter. So, this lead wire creates the problem.
Now, actually lead wire is not very critical in the case of, in the case of thermistor,
because of the reason that the thermistor resistance value is quite high compared to
the lead wire resistance, because there are many problems. Suppose there is a
temperature change along the lead wire, so obviously I mean the total resistance,
because if you look at the meter, meter from this side, what it will see? It will see that
the total resistance, it will not care, it will not, it will not see only the lead, only the
thermistor resistance, it will also look at the lead wire resistance plus thermistor
resistance. So, the lead wire temperature variation is very important in
instrumentation, right? So, it is not important in the case of thermistor, because of its
typical value. Its value is high, but it is very important in the case of the, in case of
RTD. We will see that in the later, I mean lesson, right and even though for a, I am
sorry …
30
So, we will see that when the thermistors are used to measure the temperature, errors
resulting from lead wire effect are usually small enough to be neglected, even for
relatively long lead wires, can be few meters also like this. The sensitivity of a
thermistor is high. Therefore, the change in resistance delta R T resulting from a
temperature change is much greater than the small change in resistance of the lead
wires due to the temperature variation. This is very important, right? That is the
reason we can have a long lead wire in the case of thermistor.
31
Also the resistance of the thermistor is large compared to lead wire resistance. We
have already, I discussed these things. It can be, it typically can be R T by R L can be
1000. Consequently, any reduction in sensitivity of the sensor due to lead wire
resistance is negligible, right? Now, there is another problem with the thermistor. It is
called the self heating error. Any resistance devices whenever you are using, the self
heating error, it is, it is common to both the thermistor as well as RTD, right? This
cannot be avoided.
This is the, I mean physical phenomena. You can avoid the lead wire problem, but
you cannot avoid, I mean the self heating error. Errors may occur due to the self
heating of the thermistor.
Recommended practice is to limit the current flow through the thermistor to a value
such that the temperature rise due to the I square R T power dissipation is smaller than
the precision to which the temperature is to be measured. So, actually the precision
will tell you how much current you will pass through the thermistor, right? So,
sensitivity and precision if you look at very careful that is going side by side. So, let
us look at, now suppose that R T that means resistance of the thermistor is 5 kilo ohm
is capable of dissipating a power of 1 milliwatt per degree centigrade above the
ambient temperature, right?
32
So, thus if the temperature is to be determined with an accuracy of 0.5 degree

centigrade, the power to be dissipated should be limited to less than 0.5 milliwatt,
right? This limitation establishes a maximum value of the current. What is that
maximum value, we will show this. So, you see that the maximum value of current
which will pass through the thermistor can be P by R T equal to 316 micro ohm,
micro ampere.
This will actually lead you to choose your bridge excitation or the potentiometer value
of the, I mean value of the potentiometer excitation or the value of the Wheatstone
bridge excitation, right because you can calculate how much the maximum current
will be allowed and accordingly … But please remember, if you do that thing,
obviously you will lose the overall sensitivity of the system, because your output
voltage also will be getting reduced, is not it? We are not measuring that resistance
there; we are measuring the output voltage. So, output voltage will be reduced. So,
adequate response can be obtained even at these low currents, because the sensitivity
is high. So, sensitivity is high, I can make the adequate response, but this is a typical
problem that sensitivity.
That is the advantage because, but obviously what will happen if you reduce, if you I
mean think of this limitation, obviously you have to sacrifice some amount of
33
sensitivity. That is quite obvious that you cannot avoid and I am sorry, it is not thank
you, I mean, let me go and look at some of the circuits for which thermistor is used.
As you know, the Wien bridge circuit, it is a typical circuit, it looks like this.
You see, the Wien bridge circuit looks like this, very familiar. It is a standard, I mean
circuit for making commercial oscillators. So, basically Wien bridge oscillators, now
there is R 3, R 4, like this one. So, these two voltages are coming to the, an op amp. I
will show you some of the other applications of the thermistor. This is the Wien
bridge oscillator circuit, right? This is Wien bridge oscillator. Now, see that always
we, we will try to make the, the gain of this amplifier very high. To, to oscillate we
will always make this gain of the amplifier quite high.
Now, there are some resistances. Suppose this is R, this should be R. If it is C, this
should be also C. There is no doubt about that thing and the frequency of oscillation is
omega naught will be equal to 1 by R C, right? Now, there are two resistances, you
see, R 3 and R 4. Now, I will not go into details of this Wien bridge oscillator, but you
see that in the, for stability of the system the gain should not be that high that it should
saturate the amplifier. It should not be that low also that the, your oscillation may not
start.
34
So, what they do commercially? They use either, previously they used R 4 as a
tungsten filament, because as you know, the tungsten filament and its resistance
depends on the temperature. So, as the current, as the time goes on, its resistance
increases or you can use R 3 as a thermistor. If you use R 3 as a thermistor, what will
happen? As the time goes by, the current will start to pass through this one and its
resistance value will decrease. So, that will control the amount of feedback you are
giving to this oscillator. Now, so instead of tungsten filament, I can use this thermistor
for making temperature, for making the stabilization of the Wien bridge oscillator.
This is one application.
In other application, as you know that we make usually that our power supply, is not
it? So, typically power supply looks like this.
All instrument needs the power supply, right? So, I am giving a supply here, right?
This is our AC supply. After stepping down and all this, this is the …….. I am getting
a DC here, positive this side, negative this side. This is AC this side. Now, any full
rectification as we know that, you see that our, it is not a DC, it is a ….. So, I must
smoothen it, because the pure DC should have a, suppose this is the time and this is
the amplitude, so this is the battery voltage e. So, it should be like this one, right?
35
Now, what they do commercially? They put a capacitor, a large value of capacitors,
right? Now, even for a, suppose 1 ampere power supply, 1 ampere, suppose 25 volt
power supply, so this is around, suppose 2500 micro Farad or 3000 micro Farad, this
is a typical value for the power supply. Now, according to the voltage, whatever the
voltage, across the voltage, suppose if it is plus minus 25 volt power supply, I will
…… a 50 volt power supply like this one, right? Now, the problem arises. You see
that when you switch on, now what will happen here? You see, if I look at, now what
will happen?
You see the, now my output voltage, suppose this is output voltage, right output
voltage will look like a slight fall, then again it will rise like that one, light fall, …,
because before the capacitor fully discharges, I mean charges, again it will fall like
this one, the output voltage. Charging and discharging of the capacitor, it will look
like this one, almost steady DC voltage. The best filter for this power line filter is a pi
section filter that means you use a, like this one, right? But, L is not, no more used,
because L has lot of problems, associated problems, magnetic field and it is costly, it
is bulky. So, to avoid that, people use the simple capacitor, so higher and higher
value. So, in this case, suppose if we use a 1000 micro Farad capacitor, 1000 micro
Farad capacitor, then we should have one, suppose this is a .1 or Henry or like this
36
one, so this will give you very good, but this is not used. We use it in very high value
of the capacitor 2500 micro Farad or 3000 micro Farad capacitor value.
Now, in this case, you know what will happen? You see that that initially when you
switch on the system, when you switch on the system, you will find there is a large
current is necessary to charge this capacitor, because you cannot connect the capacitor
like this one. There should be some leading resistors and all these things here,
available in resistors. So, this is always necessary, because otherwise the capacitor
will fully charge, right? So, what the people do you know, because what will happen
that if I charge this, I mean capacitor with large current, because if it is totally, I mean
discharge capacitors, it needs a large current.
Then, what will happen? This, there will be a large surge current to the diode, so
diode will burn out, right? To limit this, what we have to do, what we can do actually?
We can put a small value of thermistor in this position, right?
If I take a new page, if I take a new page, it will look like this one. So, I will put diode
as before. So, put a, now put a capacitor there, small value of the thermistor, I mean
AC. Now, initially what they do? You see it will, it will offer high resistance, this
thermistor and capacitor will charge and capacitor will slowly charge, because at this
37
time what will happen that the, since it is high resistance it will not charge fully.
Then, after sometime what will happen?
You see that as the time goes on, this, its resistance is, as the temperature …., because
the current will pass, there will be a self heating. So, what will happen to this
resistance? Its value decreases. That I want actually, because this is a wastage, this is
a loss of power; this will never be utilized. So, the efficiency of power supply will be
deteriorated just to protect four diodes. Because in semiconductor device you can say,
sir, I will use some fuse here. A semiconductor device is to be protected by another
semiconductor device, please note that, right? So, you cannot do that thing. So, I have
to use, if I use this one, so it will offer a large resistance. So, slowly it will start to
charge and as the time goes on, this temperature, resistance will increase, resistance
will decrease, sorry, temperature.
As the current passes resistance will decrease, because the temperature is rising and I
will get full. The loss within this thermistor also will be reduced, because there is no
utility of, because this is the total loss, because this voltage will be … So, this is the
typical circuit also used in smps also, because in smps also you know the switch mode
power supply or regulated power supply and all that, for giving the power to all the
chips also they do not, they need some power supply even though there is a switch
mode supply. But, for that reason they use a small series regulator. So, in those series
regulator, I have a diode rectifier like this one and it will utilize the thermistor.
Thermistor is also used in electronic circuits also for, I mean for stabilization. You
know, the …. point stability is very important in the devices. Now, once you switch
on, especially the circuit where there is a feedback, there is a chance of oscillation.
So, what they do? They purposefully put a thermistor there and this will limit the
current and this is a typical circuit I can show you.
38
You see here I have a, you see there is a current is going. There is a push pull circuit,
which is utilized for, for making amplifiers, right? I am taking a load from this
position and output, so I am giving a power supply of 25 volt here suppose. Now,
what they do actually? So, I have subsequent circuits like this one. So, there is a DC
circuit, DC circuit. So, this circuit is coming here, this circuit is coming, coming to the
collector like this one, like this one. See, this is used as a thermistor, right? So, this
can be do like this one. So, this is some resistance, we can come here also through a
capacitor. So, this 25 volt can be connected here. So, in fact this 25 volt should come
here. 25 volt, sorry 25 volt should come here. So, this signal, this is a typical
amplifier, so where the signal will come here in this, from this direction, it will go
down there.
Now, see the reason for there is also here, connection here. The reason for this type of
circuits is that, this is …. point stability, right? I need a stability of the thermistor.
This is another application of the thermistors in the electronic circuits, right? See, it
will make the …. point stability, because due to the resistance and all those thing it
may happen that it will go out. So to, because this is the feedback there, so actually
there is a feedback, so it should go like this one and it should go. So initially, so if
that, it may go to the oscillator region, so to stabilize this thing, so we will use the
thermistor and this ends the lesson 7 of Industrial Instrumentation.
39
Prof. Alok Barua
Lecture - 8
Thermocouples
Good afternoon! This is lesson 8 of Industrial Instrumentation. In this lesson, we will

study the thermocouples. As you, know the thermocouple is one of the most widely
used temperature sensors for the industry, because it is cheap and it is rugged as well
as it is easily replaceable. So, making all those qualities, I mean actually lead to a
very, I mean extensive use of this temperature sensors in almost all process industry
like steel, petrochemicals, fertilizers, everywhere.
Now, the contents of this lesson will be the principle of working of thermocouple that
means we will still go, go through the Seeback effect, the Peltier effect and the
Thompson effect. Then, the laws of thermocouple circuit; we will, I mean explain
why this law is necessary, because in many cases we will find the two, the tables for
these thermocouples, not for all thermocouples, are available. Different types of
thermocouples and their specifications - this is very important, because if we use any
dissimilar metals, obviously we will get the output voltage, but people over the years,
they have see that if we use a particular two different metals or the alloy or the alloy,
1
you will find that the output will be large. That means sensitivity will be high as well
as nonlinearity will be less. Also, the signal conditioning circuit which is to be
discussed like cold junction compensation and all those things, although we will
discuss, as well as we will discuss here the semiconductor temperature sensors.
Even though it is a thermocouple circuit, but that semiconductor temperature sensor

will be utilized to make the cold junction compensation. What is cold junction
compensation? That will be explained after sometime.
At the end of this lesson, the viewer will know thermocouple and lead wires, its range,
sensitivity, cold junction compensation as well as semiconductor temperature sensor
that I told you.
2
Now you see, it is a sensor which relies on the physical principle that any two
different metals A and B are connected together, an emf that is the function of the
temperature will be developed at the junction of these metals. In this expression, we
will find e equal to a 1 t plus a 2 t square plus a 3 t cube plus so on, right? Therefore,
we can see that this temperature emf relationship is clearly non-linear. We will see
the, most of the thermocouples is actually represented, its voltage temperature
relationship will be represented by 7 to 8 degrees, sometimes even 9 degree
polynomial. That means it will go on a 1, a 2 up to a 9.
There are some thermocouples obviously, which will have, which will obviously have
the second or third degree, it can be represented to second and third degree
polynomial. Because you see this is a non-linear term, that is the reason the
thermocouple, even though it is extensively used in industry, it is a, always that there
is a thermocouple charge available. That means for each thermocouple you need to
charge, so that if we get an unknown voltage, from looking at the chart you can find
what is the, what is the temperature, what is the unknown temperature.
But, please note that this is not very, I mean, I mean it is very serious problem,
because you see, nowadays you can have your, you can build your own ROM, where
3
you can correlate between the temperature and the emf, so the non-linearity problem
can be immediately solved.
The values of the constant, this a 1, a 2, etc., depend on the metals A and B, right?
It does not depend on the temperature. It depends on the metals and it is a constant;
we will find this is a constant. So far the metal is pure and the metal is the same type
of metal you will use, you will get the same value of a and b, same value of a 1, a 2, a
3 and so on. I have, you see here we have metal A, we have metal B.
Now, t 1 is one junction temperature, t 2; these two junctions, two different junctions,
metal A and metal B has a junction here, metal A and metal B junction here. t 1 is one
temperature, t 2 is another temperature. t 1 is greater than t 2 and we are writing t 1
minus t 2 is equal to t, temperature difference, right? So, I will get a voltage e, right?
4
Now, there are three emf present in this thermocouple circuit, right? We have when
two, Seeback effect when two dissimilar metals are joined together and emf exists
between the two points that is the function of the junction temperature. This is the
most prominent. Otherwise, we will find the other, like Peltier’s and the Thompson of
all, also will exist there, but it is not that prominent.
5
If two metals are connected to an external circuit in such a way that a current is
drawn, the emf may be altered slightly, because if you can use a meter which is, has a
very high input impedance, obviously it will not draw any current. But, if you try to
measure the voltage, obviously it will draw some current. So, this will affect, it will, it
will make a Peltier effect, so because the emf will be altered and if a temperature
gradient exists along either or both of the thermocouple wires or metals. Every time
we are saying metals, so ultimately you see these are the two different wires, right
which is actually used to make the thermocouple.
It looks like, I mean you see that if I take a blank page two different wires, so there is
a junction here, right? So, this will give you, I can draw it nicely, right, so this will
make the thermocouples. Suppose this is your metal A, metal B, so this is our
junction; please note, this is our junction. So, it is two different wires actually. Even
though we are repeatedly telling it is metal, but these are two different wires. If I go
back, so this is our Thompson effect, right.
6
Now, two important rules for analysis of thermocouple circuit. There are two different
laws, because you see that there are infinite number of, I mean combinations of the
metals you can make. But obviously, you won’t get the, I mean you can, you cannot
make the thermocouple charge for each and every, I mean combinations of the two
different metals. So, you have to use some laws of intermediate metals or laws of
intermediate temperature also. These two laws are very important for solving the
problems on thermocouples, which we will see later on, not obviously in this lesson,
but in some other lesson.
At the end of this course, we will solve some problems. Then you will find that this
type of, I mean laws of intermediate metals and intermediate temperatures is very
important. If a third metal is connected in the circuit, metal C, you see the metal C is
connected here, right? Metal C, this is the metal C which is connected. This is metal
A, this is metal B, right? Whatever the temperature, it does not matter, so it has some
different temperature, obviously. Here, I have some temperature. Here, you see that
what you are able to see that this is the temperature t 1 and this is the temperature t 2.
Now, we are taking the t 1 equal to t 2. If t 1 equal to t 2, it does not matter, our emf
will remain same as before, right? Because it is two dissimilar metals, we have
connected like this; two dissimilar metals, so these are junctions.
7
So, this is metal A, this is metal B, so I will get a current, right or you can show like
this one. I will get a current, I will get a current and this is the, this junction will be …
If it is, if I connect some other metal inside, so far this temperature remains same or
emf also will remain the same. It won’t be altered. As shown in the figure the, the net
emf of the circuit will not be altered as long as the new connections are at the same
temperature. These are law of intermediate metals, right?
Now, if a thermocouple produces emf e 1, when its junctions are at the temperature t
1, you can see here, if a thermocouple produces emf e 1, when its junction at the
temperature t 1 and t 2, these are the temperature at t 1 and t 2, so with two different
temperature, this is a metal A, this is a metal B, this is a whole ……. I got a voltage e
1, right and e 2, let that, for that we have to go for next slide.
8
Metal e 2, sorry, temperature t 2 and t 3 will be there; same metal, metal A and metal
B, I got a voltage e 2. Now, if I have the temperature of t 1 and t 3, two different,
because previously it was two metals and two different junctions are t 1 and t 2. This t
1 and t 2, like this one, like that this one, so this is t 1, this is t 2, right and now we
have connected another one, where this is t 2, this is t 3. Now, if now the temperature
is t 1 and t 3, then the net voltage will be e 1 plus e 2, right? That is, I am, actually
that is the intermediate temperature.
9
So, it is, if I go back to the previous slide, you can see, see here if a thermocouple
produces emf e 1 and when its junctions are at the temperature t 1 and t 2 and e 2,
when at the temperature t 2 and t 3, then it will produce an emf of e 1 plus e 2, if the
junctions are at the temperature of t 1 and t 3. This is the law of intermediate
temperature. These two laws, there are many laws, but these two laws will be very
important for solving problems.
Now, now we will discuss some of the important, most important thermocouples
which are used in industries. So, I call the specifications result, I mean retaining table
1, so this is important, because you know, you must know that you will, as I told you,
there are, you can make any combinations of two different metals or alloy. It will give
an output obviously, but that will not give you our desired purpose, because I need
large sensitivity, I need good stability, I need inertness, I need easy replaceability.
These, all these materials, I mean so depending on that we will find there are several
thermocouples and some of the thermocouples are very widely used in industry
depending on the temperature range, depending on the, whether the atmosphere is
oxidation, oxidizing atmosphere or the reducing atmosphere, so depending on the
output voltage, you will get from the thermocouple. Please don’t; thermocouple
sensitivity is very poor. So, per degree centigrade change of temperature, you will get
very small amount of voltage, right? So, you need signal processing for that.
So, it is not very that you will get a large voltage. Even though the voltage in the, are
of the order of millivolt, but the sensitivity is very poor. These are all microvolt per
degree centigrade, right? So, all these things are …, so large emf is always a desired
property of a choosing thermocouple. At the same time, you will find that the
linearity, stability, is all this very important. It should be inert like platinum-platinum
rhodium; platinum-platinum rhodium you will find it has a lower sensitivity. It is
around 10 to 12 microvolt per degree centigrade, but however you will find its
advantage is its inertness, it does not react with any other, because it can work very
nice in the, I mean in hostile environments also, right?
10
So you see this is the and mostly, another important thing is in the industry they do
not mention the thermocouple by name. They do not say chromel constantan, copper
constantan and iron constantan, platinum- platinum rhodium, they do not call. Usually
they give a type name. So, it is accepted for the, in the, in the any processing industry
over the years. So, we also give the typical name or type, I will say of this particular
thermocouple.
So, when you say that the type K that means you must know this is a chromel alumel
thermocouple. You see here this is the type K. So, this means it is chromel alumel
thermocouple. What is chromel? Chromel will be positive, Alumel will be negative.
That means what does it mean? That means if I, if I take a blank page it will be more
easy.
11
You see it is a, if I take like this one, this is our thermocouple, right? So, suppose this
is chromel and this is alumel, so this will give you positive voltage, the positive side
of the voltage. You see, thermocouple output is a pure DC voltage, right, so there
should be, some probability will be there, so which is positive, this is negative. This is
very important, while you will see in the lead wires, connections and all those things.
Now, please note another thing, while I will discuss the specifications, in the
thermocouples you need the lead wires. The lead wires means that you see that this,
how the thermocouples are very expensive, like platinum - platinum rhodium and you
cannot, I mean you cannot install your instrument or voltmeter very near to the point
of measurements. It might be far away, it might be 2 meters, might be 3 meters, even
sometimes more, right?
In that type of, say suppose I have a boiler which has temperatures of suppose 400
degree centigrade, I cannot install a meter very close to that, it is not possible also.
You have to transmit that signal over a long distance or even just if you have a
monitoring instrument, it must be, you must measure the voltage. Suppose it is a …..,
it is a, in the case of transmissionthat is different, but if it is simple monitoring
instrument, it might measure the voltage. If I want to measure the voltage, I cannot
install this meter, voltmeter very close to the wire or else the meter will be out of
12
order, meter will be damaged. So, there should be some wire, some distance of wire,
suppose 2 meter, 3 meter.
Now, the problem is that we will find that if you have a, if you have a, I mean a
thermocouple or platinum-platinum rhodium type, you cannot install a platinum-
platinum rhodium lead wires, because it will, it will cost, it will increase the cost of
the entire system, because the thermocouple is available like this one. So it is a, is a
terminal will be available here, right, so you connect the lead wires there. Now, it will
come to the, your voltmeter, so you will get a voltmeter, isn’t it? So, you will connect
a voltmeter there. So, this is certain distance here, this is certain distance we will find
around, suppose like some few meters. So I need, so people has suggested some, a
particular class of, for each thermocouple, thermocouple wires there is a particular
type of lead wires also have been recommended.
Now, you cannot change the lead wires. If you change the lead wires, you will find
your calibration also will go wrong. You cannot the flip also the lead wires. I will
show you what is that?
Now, you see here this is chromel alumel thermocouple and it can be represented by
eighth degree polynomial. What is that?
13
We have already told thermocouples e equal to a 1 t plus a 2 t square, so on. So, in the
case of these type of thermocouples, so a 1, 8, after that it will be insignificant the
coefficients, so I can ignore that thing. So, it can be represented by the eighth degree
polynomial, right?
Its application is basically minus 200 degree centigrade to 1300 degree centigrade and
the main application however is from 700 to 1200 degree centigrade in the reducing
14
atmosphere. That is I told that it should be a, you must mention whether it is reducing
or the oxidizing atmosphere. Total voltage swing that means for this change of
temperature that means from minus 200 to 1300 degree centigrade, I will get a voltage
swing of 56 millivolt. Now, typically you will see that the thermocouple, this
sensitivity of the thermocouple, you will find it varies. That means per degree
centigrade change of temperature how much the voltage change I will get at the
output of thermocouple that is called the …. sensitivity of the thermocouple.
It varies from, suppose in the case of platinum - platinum rhodium, 10 to 12 microvolt

to 60 microvolt per degree centigrade maximum, not more than that. Lead wires, see
as I told you, you see it is recommended that means iron copper nickel alloy, you see
here this iron copper nickel alloy that means in the lead wires iron should be
connected to chromel and copper nickel alloy should be connected to the, to the
alumel, right? Similarly, there are two options. You can use, instead of iron copper
nickel alloy you can use copper constantan that mean copper will go to the Chromel,
connected to the Chromel and constantan will be connected to the Alumel, right? So,
for each thermocouple we will find this type of tables, this type of, I mean
specifications on lead wires have been recommended. You see, this is t type of
thermocouple, this is also quite widely used thermocouple.
15
This is a copper constantan and copper is positive, constantan is negative. That means
if you connect a voltmeter, positive side of the voltmeter should connect to the copper
and negative side we will connect to the constantan. It can be represented by the
eighth degree polynomial. Application - minus 200 to 350 degree C; beyond this
temperature oxidation of the copper will occur, so we cannot use. Because, if you
oxidize, the copper is oxidized, your entire calibrations will be no more valid, so you
cannot use beyond 350 degree centigrade. We …., do not need to be, because there
are many other thermocouples which can be used.
Now, voltage swing, you see the voltage swing we have calculated 26 millivolt for
minus 184 degree centigrade to 400 degree. For these change that means around 584
degree centigrade, total change in the output voltage I will get 26 millivolt. You can
see that how much is the sensitivity. Now, you see another interesting point in the
case of, I mean low cost thermocouple you can use the same wire for the lead wires
also, because you cannot get a cheaper wire than this one. Suppose the iron
constantan, obviously we will use the iron constantan as a lead wires. Copper
constantan, you cannot have a, have a cheaper than this. You have also, you should
have the lead wires also of the copper constantan, right? So, copper should be
connected to, copper constantan should be connected to constantan, there is no
problem.
16
Now, J type of thermocouple, this is also quite cheap thermocouple, widely used
thermocouple. The positive is iron, negative is constantan. It can be represented by the
seventh degree polynomial. Application is minus 150 degree centigrade to 1000
degree. It is usable in the oxidizing atmospheres to about 760 degree centigrade and
reducing atmosphere to 1000 degree centigrade. So, in which atmosphere we will use,
depending on that you will choose the temperature range.
Voltage swing, you see, minus 184 degree centigrade to 760. So, you see, sensitivity
is not much high, because the, here the range is quite high, I mean quite large,
whereas in the previous case that copper constantan, the range is small. So, for minus
184 degree centigrade to 760 degree centigrade of voltage, temperature change of the
hot junction, so I will get a change of 50 millivolt, right and lead wires, iron will be
connected to iron positive and constantan will be connected to constantan.
This, as I told you earlier, this is the platinum. These all are noble metals, obviously it
is quite expensive. Now first one, the positive side is platinum 90% and 10%
rhodium, right and negative is platinum, please note. It can be represented by a second
and third degree polynomial, right? Now, main features of it is chemical inertness,
stability at high temperatures and its oxidizing atmospheres, reducing atmospheres
17
cause rapid deterioration and high temperatures and typical range is, I should, range
is, should be zero to 1538 degree centigrade. Voltage swing is 16 millivolt.
Now, copper, I mean in the case of platinum-platinum rhodium, I mean thermocouple,

you cannot use the same, it is expensive. We use copper and copper nickel alloy.
This is also the platinum-platinum rhodium, but here the rhodium percentage has been
increased. So, instead of 10% rhodium, because you, as you know rhodium is more
expensive than the platinum. So, it is 87% platinum and 13% rhodium alloy, which
will make the positive terminal and negative is the platinum; can be represented by
second or third degree polynomial. Application is zero to 1593 degree centigrade.
Voltage swing is 18.7 millivolt. So, copper should be connected to platinum rhodium,
copper should be connected to platinum rhodium and negative copper nickel alloy
should be connected to platinum. This will give …
18
Now we have one more, two more rather. Platinum rhodium, this is one positive
which is platinum 70%, 30% rhodium and platinum your 94%, 6% rhodium that will
become negative. It can be represented by eighth degree polynomial. So, no, I mean
advantage as we have in the case of 10% and 13% rhodium and application, 30, 38 to
1800 degree centigrade.
Now, you see application is almost same for all the, all this I mean, platinum -
platinum rhodium thermo …... That is the reason we have not represented and we
have not written it, I mean repeatedly, but it is same, right? As it happened in the case
of first, I mean slide of platinum-platinum rhodium thermocouple, so same the lead
wires are copper-copper nickel alloy.
19
Now, E Chromel constantan; it is actually rather, I mean it is a newer thermocouple. It

has a, its sensitivity is the highest, please note. It is very high, so it is zero to 980
degree, 82, but it is 9 degree polynomial and the 75 millivolt for this range, voltage
swing we get will be 75 millivolt. So, it is quite high that inside it is more than around
65 microvolt per degree centigrade, highest in fact. So, iron constantan are the lead
wires, so iron should be connected to the chromel and constantan should be connected
to constantan, right?
(Refer slide Time: 24:32)
20
Now, what is constantan? Constantan is an alloy with 55% copper, 45% nickel.
Chromel is 90% nickel, 10% chromium, Alumel is 94% nickel with 3% manganese,
2% aluminum and 1% silicon. Now, typical sensitivity is 10 microvolt to 60 microvolt
that I told you earlier.
Now, cold junctions you see the, the thing is that in the cold junction that we see in
the industry, you are, while you are, I mean making these, you, making a
thermocouple, hot junction will be the measuring junctions and cold junction is
supposed to be at zero degree centigrade, right? Well, I mean, if I had, suppose
instead of zero degree centigrade if it is something higher, suppose 15 degree
centigrade or 35 degree that is not a problem.
I can add, I am, accordingly add or subtract voltage from our actual voltage, so I can
get the, I can get the voltage, actually the voltage for 35 degree centigrade of the hot
junction, I mean the cold junction. But the problem is if the cold junction temperature
varies it is not possible, is not it, because you, you cannot maintain the cold junction
at a constant temperature. Summer will be one temperature, in the winter there will be
some other temperature, in the monsoon there will be some other temperature of the
cold junction. …… junction is something different, because you are measuring that
particular temperature in which you are interested. So, but I am not interested to know
21
the, there should not be any variations of the output voltage for the change of the
temperature of the cold junction. So, I must have some sort of mechanism by which
that I must correct the voltage output from the thermocouple, right? So, that is called
the cold junction compensation of a thermocouple circuit.
If ambient temperature variation of the cold junction can cause significant errors in
the output of a thermocouple pair, there are two alternatives, right? One is maintain
the cold junction at a constant temperature by some technique as an ice bath or
thermostatically controlled oven or subtract a voltage that is equal to the voltage
developed across the cold junction at any temperature in the expected ambient
temperature range, is not it? Because the and you see that the latter one, this one is
much easier.
The number 2 is much more easier than the number 1, because you cannot have a
industry, you cannot, if you have some 500 thermocouple you cannot have a 500 ice
bath, I mean thermostatically controlled oven. So, we have to make this type of the …
Some electronic circuit is necessary to make this; subtract the voltage is equal to the
voltage developed across the cold junction at any temperature. So, it will
automatically nullify the variations of the cold junction temperature.
Now, in this context, I must discuss one junction semiconductor sensor. Actually that
is utilized, widely used for, in the industry for making the cold junction compensation.
This is developed by the analog devices.
22
Now, you see, junction diodes are well suited for temperature measurement. The
junction potential of the silicon transistor and the diodes changes at about 2.2 millivolt
per degree centigrade over a wide range of temperature; 2.2 millivolt per degree
centigrade. This property can be used as the basis of an inexpensive sensor having fast
response. Even though it is used as a temperature sensor, but I will use this particular
property of the semiconductor junction diode to make my cold junction compensation
of thermocouple circuit.
23
Now, analog device AD590 is a two terminal temperature sensitive current sources,
right? It is a current source, which passes a current numerically equal to micro ampere
to absolute temperature when excited by a voltage of 4 volt to 30 volt at temperature
between minus 55 degree centigrade to 150 degree centigrade. You see, this is quite
wide range and obviously I can measure a temperature suppose of 700 degree
centigrade that is quite obvious, but obviously the cold junction, actually temperature
does not vary between, it is a quite, it, it won’t go below minus 55 degree centigrade,
neither it will go above 150 degree. This is a quite wide range I should say.
Usually what it happen in a tropical country that temperature will vary, suppose 7 to 8
degree centigrade which is very cold to suppose 40 degree centigrade, so with this
temperature range, I can make very nice cold junction compensation by these devices.
Now, this is the symbol of the constant current source. You see, its range is minus 55
degree centigrade to 150 degree centigrade, which is equivalent to 218 Kelvin to 423
Kelvin and it will give you, for this Kelvin temperature 218 micro ampere to 423
micro ampere. So, that means for 1 micro ampere, 1 Kelvin change, I will get 1 micro
ampere change in the current.
24
This is the simplified circuit of the cold … AD590. You see, the two transistors. Its, I
T is divided in two half Q 3 Q 4 and Q 2 Q 1 and there is a resistance R. The details is
coming next.
If the transistors Q 3 and Q 4 are identical, the current I T is divided into two equal
parts I C1 and I C2. Let us go back, so we can go, so I C1 and I C2, it is going in the I
C1 and I C2 dividing two equal parts I C1 and I C2. Now, Q 2 consists of 8 transistors
25
in parallel, so the current in Q 1 is 8 times the current in each of Q 2, quite obvious.
Again, the different between, difference between the VBE’s that means the voltage
emitted based emitted drop of the transistors, two transistors, of two identical
transistors with different collector current is proportional to the absolute temperature.
Now, I can write that V T equal to V BE1 minus V BE2 equal to that is base emitted
drop of the two transistors that is kT by q natural log I 1 by I 2 equal to k by q. If you
put all these, k, q, Boltzmann constants, then electronic charge, all, all the values, you
can put natural log, because I 1 and I 2 its ratio is 8, so natural log 8 kT, so putting all
the values of the Boltzmann constant and the electronic charge, I will ger 179 into 10
to the power minus 6 into T, T is the temperature into volts, right?
Now, the V T is the voltage across R and is thus proportional to the absolute
temperature. Therefore, the current through R that means that resistance I C2 must
also be proportional to the absolute temperature.
26
Since I T equal to 2I C2, the total current through the device I T must be proportional
to the absolute temperature and if R equal to 358 ohms, then obviously what will
happen? You will find the I T by T current for, by temperature if you want, micro
ampere per Kelvin, right? So, using this property, so people are making, so following
figure will show that a simple application in which the variation of the cold junction
voltage of the type J thermocouple, iron constantan, is compensated by a voltage
developed in series by the temperature sensitive output current AD590.
27
You see, this is our circuit. It is a very common circuit used in all laboratories, all
industries. The temperature range is minus, I mean sorry, 15 degree centigrade to 350
degree centigrade. That means you see, this cold junction is very close to the AD590,
right? This cold junction temperature may vary and this is the measuring junctions,
this is my measuring junctions and this is my cold junctions. Let me take that means
you see, this is our measuring junction and this is our cold junction. This is AD590
and it has a supply voltage seven point ….
These are just not, it is AD50. So, ….. analog device, but it will give us, I mean
stabilized power supply of 2.5 volt. It is necessary and this is by the voltage. This
iron, this is constantan, this is also constantan. This is iron constantan; this is, this is,
actually the scheme is for the iron constantan thermocouple. This is true for any
thermocouple. Only we will find, the value of R A will change. As the thermocouple
changes, this value of R A will change, right? Otherwise, it is the same circuit for
everywhere that means this 52.3 will be different for different thermocouples.
I will show you the different charts that means this AD590, so this will give you
current. So, obviously this will be my output voltage. So, this voltage will be, even we
can show you experimentally, if I change this cold junction compensation
temperature, cold junction temperature, this is the cold junction, because this is our
cold junction, this is AD590, even though, I mean if the measuring junction, if
measuring junction does not change, temperature does not change, we will see, we
will see that the output voltage will not change, right? Fine.
So, this is our, now we will see that it is, output voltage is equal to V T. You see, it is
independent of the temperature of the variations here. If the variation, it will, I A will
take care of that, because if the temperature increases I will increase, automatically
the voltage will be subtracted, so the V T will be there. So the, that is I want, actually
my output voltage should be equal to V T.
28
The circuit is calibrated by adjusting R for proper output voltage with the measuring
junction at a known reference, because you see that you have a resistance R. You can
see here, so this is our resistance R. So, for a known resistance with temperature the
….. is calibrated by calibrating this R value. Now, once it is calibrated, so we can go
ahead; so, voltage with the measuring junctions at a known reference temperature and
the circuit near 25 degree centigrade. If the resistors with low temperature coefficient
is used, the compensation accuracy will be within 0.5 degree centigrade for
temperature between plus 15 degree centigrade to 35 degree centigrade. Other
thermocouple may be accommodated with the standard resistance values shown in the
table II. You see this is the table II.
29
We will find that, as I said that the resistance R A, so in the case of J type of
thermocouple it is 52.3 ohm. For K, it is 41.2 ohm and E, 61.4. Every time it is,
obviously you have to calibrate it by, by, with the known temperature using that
particular, the resistance R. But, this is the fixed resistance or whenever we will
change the thermocouples, you, this resistance R A, R subscript A must be changed
also.
Already we have shown in the circuits of the, for J type of thermocouple with 52.3
ohm, we have given the chart here, right? So, S, R it is same 5.6 ohm.
30
Now, desirable properties of a thermocouple, please note it is for industrial uses. They
are relative large, the emf; everybody wants the large thermal emf. Especially the
change is more important, because you can amplify the signal. That is not a very
problem, you can amplify the signal the dc signal, but the change, sensitivity should
be high. Precision of calibration - you see, in the industry it is not possible, I mean
usually what you will, you see, the calibration you cannot afford to have the
calibrations every time.
What you will do? That you will take a large length of wire, suppose copper
constantan, Chromel Alumel or platinum-platinum rhodium, what the harm can
happen? Your, suppose for some, due to some contamination the junction
thermocouple has been, calibration has gone wrong or suppose due to the, I mean
welding I mean due to the problem in the welding, this thermocouples has ….., so you
just, what will you do? You will replace this thermocouple. So far you take from the
same pieces of wire, you do not have to recalibrate the instrument. That is the great
advantage of this type of thermocouple.
If I take a wire, large length of, suppose if I take two wires of the platinum-platinum
rhodium 10 meter, another platinum of 10 meter, we calibrate it, fine. Once you
calibrate, we make our calibration chart, right? What we will do? We will use a small
31
portion of this, suppose 30 centimeter portions of we will cut from each of the, each
of the wire, then we will make our own thermocouple. If it is damaged, then we will
just take the same wire, again from the, that the same largely, that that 10 meter wire,
again we will make up our own thermocouple, is not it? So, I do not have to
recalibrate the instrument; it is a great advantage.
Resistance to corrosion and oxidation, this is very difficult, because it is not, mostly
we will find this resistance to corrosion and oxidation is not available. You will have
to use the thermocouple in a covering material, which is called an industry sheath.
The interchangeability of the thermocouple is a principle reason for their wide use and
application. Best accuracy is obtained with the platinum thermocouple, which has the
accuracy of half percent of the temperature emf calibration curve. Thermocouples are
most commonly made in the form of wires insulated and welded together at the
measuring junction. As I told you, there are two types of weld. A twisted weld and a
butt weld. What is that?
32
You see here that your twisted weld means you will get a wire, you will get a wire,
you will get a wire. Now, this portion is welded or you can have a butt weld. This is a
twisted weld. You can have a butt weld, which will look like this. This is a butt weld.
So, there are two types of weld available. If the short the twisted weld is for wire, so
larger size and butt weld is for the smaller size. If the wire is small, so you can have a
butt weld, even they have a twist. Obviously, twisted weld will have more strength,
anyway.
33
In order to prevent the forming of a second junction, the wires of a thermocouple are
insulated from each other by being threaded through the porcelain insulator, which
will retain their shape up to 1500 degree centigrade, because the two wires might get
connected. If it is connected, you will get a second junction that is undesirable, so
what we will do? We will pass through the insulator and it can retain a shape of 1000,
up to temperature of 1500 degree centigrade. It looks like this.
That means you see that it will, what it will happen? You will find that wires are
there, sorry, see you will get a wire and a porcelain will be, it is passed through the
porcelain. That means it looks like, it looks like, sorry, here it will be there; now it
looks like this, right? So, you will get two different wires and a porcelain separator
will be there. So, you can make your own thermocouple like this one.
34
Now, multiple junction thermo, thermocouple output is very small, as I told you. You
will find sometimes it is necessary to make the multiple junction circuit, right? So,
multiple junction thermocouple is called thermopile. Thermopile is a term used to
describe multiple junction thermocouple that is designed to amplify the output of the
circuit. It is used in the, to measure the radiations, optical radiations; it very easily can
be measured. See, the thermocouple voltages are typically in the millivolt range.
Increasing the voltage output may be a key element in reducing the uncertainty in the
temperature measurement.
35
You see, this is actually a thermopile or a multiple thermocouple. There is a

measuring junction, several measuring junction, it can be ‘N’ such measuring
junction. Cold junction will be at one temperature, measuring junctions at one
temperature. So, this is the reference junction. So, what will be the output? Output
will be the ‘N’ times the voltage of the single thermocouple. This is because, they all
are in series, right? So, I will get obviously large output, right?
36
So, that is actually the thermopile or in a thermopile the measuring junctions are
usually located at the same physical location to measure one temperature. Above
figure shows the thermopile for providing amplified output. In this case, the output
voltage would be N times the single thermocouple, where N is the number of
junctions in the circuit.
It looks like this that means if I take a, we will find the thermopile, actually what they
do? They make a, like this one, so the, so that this, this centers they make the hot
junctions and these are all, the outside are called cold junctions and they put a
collimator lens over this one. So, when the sunlight falls what will happen? So, it will
get a, collimated here and all the measuring junctions will have the same temperature,
right? So, this type of photometer is available for optical radiation measurements and
the sun rays measurements and all these things.
37
In transient measurements, the thermopile may have a more limited frequency range,
because obviously you can see that that the, it will increase thermal capacitance. The
time constant of the thermocouple will usually increase. Now you see, thermocouple
is, it can be used for very fast measurements, because for dynamic temperature
measurement if you look at the time constant, the time constant of the thermistor is
very small. The thermocouple is obviously high, but the, but the, but the, it is
obviously, I mean higher than the thermistor, but you can utilize it for making the fast
measurement also.
But, the problem is that thermocouple you cannot use as a bare; you need some
covering material or sheath material. So, if you use any sheath material, obviously
what will happen? The time constants of the thermocouple will increase and the
response time, as I, we know that if the time constant increases the respond time,
response time will be increased. So, it will take more and more time to find the steady
state value of the temperature. So, thermopiles are particularly used for reducing the
uncertainty in measuring small temperature differences between the measuring and
reference junctions.
38
Thermocouple sheath, it is, the thermocouple junctions are prone to contamination by

gases, liquids and other metals. So, what you see, it look like this. So, we have, I have
a thermocouple, sorry, so, so I have a thermocouple. Entire things are put on a sheath,
right? So, this is our thermocouple T/C and this our sheath, right? So, obviously it will
increase. It will protect your thermocouple from any hostile environments. Suppose
any corrosions, so that type of things it will, I mean protect your thermocouple, but it
will increase the time constant of the system.
39
The metals contamination alters the thermoelectric behavior of the device, such that
its characteristics varies from the, that published in standard tables.
Because the calibration changes, everything will be changed. Some of the common
sheath materials, we see the mild steel it can withstand the temperature of 900 degree
centigrade; fused silica, it can withstand the temperature of 1000 degree centigrade,
recrystallized alumina 1850 degree centigrade, magnesia 2400 degree centigrade,
thoria 2600 degree. Effect of the sheath on time constants of the thermocouple,
obviously will increase, time constant will increase.
40
Now, grounded thermocouple, you see thermocouples are influenced by the external
effects. Prime examples of the external effects not always considered are the effects of
electric and magnetic fields, crosstalk effects, I mean effects connected with the
common mode voltage rejection.
Now, brief review of these follows that the voltage sources are capacitively coupled.
41
This all actually occurs when you are measuring the dynamic temperature
measurements. This is usually for the steady state temperature measurement. This is
not of, for very much important, but in some cases you may have two thermocouple
wires, which is going side by side. So, what is the influence of the measurements in
the dynamic measurements of temperature that is to be studied.
Parasitic capacitances and causes an alternating noise signal to be superimposed on

the desired signal. The noise is minimized by the, shielding the thermocouple
extension wires and grounding the shield, right?
Now, magnetic fields radiated from the current carrying conductors produce noise
current and hence noise voltage in the thermocouple circuit. So, magnetic noise is
minimized by twisting the thermocouple extension wires. Adjacent pairs of a
multipair cable tend to pick up noises when pulsating dc signals are transmitted, as I
told you. Now, crosstalk noise is minimized by shielding the individual pairs of
thermocouple extension wires.
42
Electrical connections made between the thermocouple and the grounded instrument
may introduce common mode noise, if different ground potential exists along the wire
path. Common mode noise is minimized by grounding the thermocouple and its
shielding at a single point as close practical to the measuring junction. Now, several
arrangements you can have for extension wire shield and ground combination
acceptable from the noise view point, as shown in the following figure.
43
Now, you see, that is one junction. You will find that this junction has been grounded,
right? Several …, now properly grounded circuit, when the measuring junctions are
grounded to the sheath. Measuring junction is not grounded, but it is grounded to the
sheath and you see that, this is the, this is the extension shield. There is an extension
shield, because to protect your environment, protect the, your thermocouple extension
wires from the external electrical circuits, electromagnetic fields, this is to be
grounded because any electrical wire was, once you have shielded it, whether you all,
almost all of you are using the oscilloscope; you have seen that the oscilloscope is a
very sensitive device.
When you are measuring the voltages, you always, we cannot give a bare wire like
this one. There is a shielded cable through, wire through which the signal will go,
right? So, that shielding is very much necessary, because the shielding and if you, if
you shield that wire and if you, two wires and if you ground that shield, obviously all
the problems of the parasitic capacitance can be easily eliminated. Any capacitively
coupled signal also can be grounded easily.
This is another alternative arrangement. You see, this is alternative grounded circuit.
You see here the, the thermocouple is not grounded while the sheath is grounded and
you see this, all the, one of the junctions either p-type of materials or n has been
44
grounded also. So, this is another grounding circuit which is used in the ungrounded
measuring junction. Now thermocouple, you will see that it is extensively used,
because of the, I mean because of, because it has a problem. Problem is, you see it is a
nonlinear device, obviously. All the temperature sensors, except RTD, we will find
that the RTD are nonlinear.
A thermistor is highly nonlinear. For a very short range, I think very, very short range,
you can assume the thermocouple as linear, but in all the cases we will find all this for
any unknown temperature I have to refer, in the case of thermocouple I have to refer
to the thermocouple chart. That means there is a temperature, I mean emf chart is
available, usually I mean available in some, published by some National laboratories
or something like that and sometimes you will find that the industry people, they
make their own chart.
You will find that in industry they make, because they are used suppose in the, you
are, they are using all Chromel Alumel thermocouple, so they make a chart of the
Chromel Alumel. Once they get the wires from the, your welder, they immediately,
they take a sample and make a calibration chart. It does not take much long, I mean
time. So, once you get and publish that one, so once you have that is distributed
throughout the plant. You will find that wherever there is an unknown temperature,
using Chromel Alumel, so you can immediately refer to that. Obviously with the cold
junction compensation and all those things, what will happen? You will find that that
type of problem you can immediately find the unknown temperature, right?
Obviously, you will find ...
Another problem you see that if you are measuring very high temperature, suppose I
am measuring temperatures of 1500 degree centigrade, right and in, in such a high
measuring, I mean temperature, you can have your, I mean cold junctions, you will
find even in a, in a bare that means without any cold junctions you can bare it; just we
can leave it in the atmosphere, your cold junctions. The reason is, you will see that
the, you have to calculate that how much error you are introducing. See, even if you,
if the cold junction changes the temperatures of 15 degree centigrade, I mean 15 to 35
degree centigrade, total change you will find is not long. So, in that type of situations
45
you will find it is not very difficult to make, you, I mean to leave these cold junctions
bare. That means leave it in the atmosphere, let the temperature change, let it change
the output voltage, but if we have changed the output voltage, you will find that it will
not change our, it will not introduce any significant error.
If the error is suppose 1%, if the error is, this error is just 1%, we can ignore that
error. We can ignore this, all the complexity of the cold junction compensation circuit,
right? So, this is the beauty of your thermocouple and you will see that in the
thermocouple, another I mean, another great advantage, as I told you earlier is the
time constant also. Its time constant is very small. It is easily available, all the RTD’s
we will find it is very bulky, its time constant is large. Thermistor is semiconductor
devices, its calibration is problematic. Two semiconductor, two different
semiconductor having the same amount of resistance is also very difficult to achieve.
So, all these problems are not there.
That is the reason over the years you will find it is most widely used, I mean
temperature range, I mean up to suppose 700, 1700 degree centigrade or 1800 degree
centigrade, I should say, I, I am not that, I mean I should not go to that high
temperature. Suppose a 100, 1500 degree centigrade, I can easily use thermocouple.
Obviously, if you have to go higher temperature, suppose above 1800 degree
centigrade or 2000 degree centigrade, we have optical pyrometer sort of temperature
measuring devices.
Otherwise, for the temperature range, typically industrial range is suppose from 500 to
800 or 900, 1200 degree centigrade, I can easily use thermocouple. But, please note
that all the things you have to, even though we are saying the thermocouple, what
about the associated circuit? That means compensation circuit, then your, the sheath
materials which must withstand the temperature. No, you will usually find no
thermocouples used in bare, always there will be a sheath, because you do not know
you might have been, you might be interested to measure temperature of the,
temperature of the hot sulfuric acid, you cannot use that above, you cannot use copper
constantan, iron constantan; it will immediately react.
46
Pure copper constantan, iron constantan thermocouple will immediately react with
the, that acid. So, I have to put the thermocouple in a sheath material like steam in, so
which will not react with the acid. That will lead to our safe measurements of the
thermocouple, I mean of temperature. So, with this I come to the end of this lesson
that is lesson 8 on thermocouple.
47
Welcome to the lesson 9 of Industrial Instrumentation. In this lesson, we will
basically cover one of the most important temperature sensor that is resistance
temperature detector.
Resistance temperature detector, even though it is not much, I mean used in the
industry, I mean in the plant, we will find the thermocouple is huge in number than
you will find the thermistor, but I should say for precision thermometry that means if I
want to measure the temperature with high accuracy, resistance thermometer is the,
resistance temperature detector is the only solution, because this, even though it is
resistance thermometer, now what you call nowadays we call it resistance temperature
detector and there are three basic classes of the resistance temperature detector.
We will find we have platinum, then nickel, copper, tungsten, all these things will be,
I mean discussed in details. Also, the signal conditioning circuit of the resistance
temperature detector, which are basically nothing but some bridges. Either, we have
seen in previous cases also, we can use it either in the, in the unbalanced voltage
mode or you can use it as balanced mode, right?
48
Let us look at the contents of this lesson. First, we will consider the theory of the
resistance temperature detector. Then, we will discuss the measuring bridges. What
are the different measuring bridges in the resistance temperature detector? Then, we
will see the construction. What is the basic construction of the resistance temperature
detector?
49
So, obviously at the end of this lesson the viewer will know details of platinum RTD,
then its signal conditioning circuits, its construction.
So, this is the circuit.
You see, this RTD is connected to the bridge by three wires. You see, three wires
means you see here there is one wire, there is another wire, there is another wire. Now
you, interestingly you will see this R 3 now is known in the bridge. R 3 is in the
bridge, but the contact resistance is not in the bridge circuit, it is in the galvanometer
circuit or it is all in the detector circuits. How do you know it is the galvanometers or
you can say it is just a multimeter or a voltmeter, is not it, right?
So, if there is little increase in contact resistance or little decrease in the contact
resistance, nothing will happen. What it will do? It will simply make your bridge
more sensitive or less; your detector more sensitive or less sensitive, you shall use it,
is it not? Suppose in the case, some cases I want to make a, use a milliammeter in
bridge circuit, which is carrying a current of suppose 500 milliampere. Even though
our detector circuits can only, can read 10 milliampere, even you can use a shunt, is it
not, right?
50
You use a shunt and when I am very close to the measurements, so I remove the
shunt, so that, because I know that bridge is almost balanced. There is also a little
chance of, I mean passing a large current through the bridge, through the detector. So,
that type of situation may arise, so any contact resistance will not, this contact
resistance will be more in the bridge circuit. So, no calculation, nothing whatever the
bridge balance equation will …… That will be absolutely correct, but it will be in the
detector circuit. So, it, make the detector more sensitive or less sensitive, it does not
matter, it is no way, I mean, I mean coming …...
So, I am denoting some current, because we have to calculate something. This I 2,

which is the resistance, which is the current through the resistance R 2. I 4 is the
current through the resistance R 4 and at balance, this R4 is, I 4 is also flowing
through R T and since it is, no current is flowing through a detector balance and also
this I 2 is, full is flowing through the, this R 3 as well through R 1 back to the battery,
right? This is called the, RTD is connected to the bridge by three wires ….., right? So,
this is all about your RTD, one of the most accurate temperature sensors.
51
Prof. Alok Barua
Lecture - 9
Resistance Temperature Detector

basically cover one of the most important temperature sensors that is resistance
temperature detector. Resistance temperature detector, even though it is not much, I
mean used in the industry, I mean in the plant, we will find the thermocouple is huge
in number. Then, you will find the thermistor, but I should say for precision
thermometry that means if I want to measure the temperature with high accuracy,
resistance thermometer is the, resistance temperature detector is, is the only solution,
because this, even though it is resistance thermometer, now what you call nowadays,
we call it resistance temperature detector and there are three basic classes of the
resistance temperature detector.
You will find we have platinum, then nickel, copper, tungsten; all these things will be,
I mean discussed in details, also the signal conditioning circuit of the resistance
temperature detector, which are basically nothing but some bridges. Either, we have
seen in previous cases also, we can use it either in the unbalanced voltage mode or
you can use it as a balanced mode, right?
1
So, let us look at the contents of this lesson. First, we will consider the theory of the
resistance temperature detector. Then, we will discuss the measuring bridges. What
are the different measuring bridges in the resistance temperature detector? Then, we
will see the construction. What is the basic construction of the resistance temperature
detector?
2
So, obviously at the end of this lesson the viewer will know details of platinum RTD,
then its signal conditioning circuits, its construction. So, these are the basic materials
all the people will be familiar with. Most important is the circuits, I mean Wheatstone
bridge, because from the user point of view, I mean we will get this resistance
temperature detector ready from the market, but we must know that how to use it,
how to make our signal conditioning circuitry, so that the lead wires errors and other
different possible errors can be minimized.
Resistance thermometer theory if I look at, resistance thermometer that is alternatively

known as a resistance temperature detector or RTD is one of the most accurate
temperature sensors. Using this, immediately people will raise the question that means
why do you need such an accurate temperature measurements? In many industrial
applications this is not necessary. Even if there is a temperature difference of 1 or 2
degree or 3 degree, it hardly matters. Especially at high temperature, suppose the
temperature is around 100 degree centigrade or 200 degree centigrade, the percentage
error is quite small and many process we will find, suppose in the, in a plant, in the
process plant, this much of difference hardly matters. You do not need that accurate
temperatures and detectors and resistance temperature detectors is not very, it is a
cumbersome, its electrical circuits are complex and all those things are there, lots of
precautions we have to take, but there are some applications.
3
One of the applications is the bioreactor application that means where the cell, you
know in the bioreactor the cell grows. So, in that type of situation, the temperature,
precise control of the temperature is very important, right? In that type of situation,
resistance temperature detector is the only solution, because in that type of situation if
the temperature varies, suppose if it is ……. temperature or set point should be 31
degree centigrade, it should be 31 plus minus .5 degree centigrade, not more than that
because if the temperature deviates above or below these, then the cell may die.
So, in that type of situations, I need the precision resistance measurements, because
once you make the measurement, in all instrumentation systems, once you make the
measurements, I can control also the temperature. The need of measurement is to
control it, is not it? That means heater power is to be controlled or some boiler
temperature is to be controlled, all those things will be there, until unless we can
measure it precisely. Moreover, you will find in some situations I need to measure this
temperature, small temperature differences. In that type of situations also, I need
resistance temperature detector, because I need a high accurate temperature difference
there, right?
The variation of resistance of a metal with the temperature can be represented by the
following relationship.
4
Let us look at the relationship. It is R t equal to R naught 1 plus alpha t plus beta t
square plus gamma t cube plus so on. This is equation number 1 of the resistance
thermometry. This is the basic equations, right, where this R naught is the resistance
at zero degree centigrade, R t is the resistance at t degree centigrade and alpha, beta,
gamma, etc., are constants, right? Now, the purity of the platinum can be checked by
measuring this R 100 by R 0, because whatever the, whatever the, actually the
materials you are using for making your resistance thermometry that should be pure.
Unfortunately if it is not pure, then what will happen? You will see that it will deviate
from the conventional resistance temperature graph, so, because alpha, beta, all things
will change.
So, I need a pure, pure platinum and moreover, you know that the, one of the, I mean
one of the basic need of using the resistance thermometer is the, especially platinum
resistance thermometer is its inertness. Platinum is very inert and whenever you have,
so that I can use it in very hostile environments, where there are chances of oxidation
and all those things that can be avoided. In the case if we use the platinum
thermometer or it won’t react with, in a, suppose if I want to measure the temperature
of the kiln or suppose if I want to measure the temperature of the sulphuric acid, the
sulphuric acid bath or sulphur, molten sulphur, so that type of situations I need some
thermometer which will not, even though we can put in a, in a, in a weld or in a shield
or in a sheath, so it does not matter.
It should not, it should not react with the materials of which we are interested to
measure the temperature, right? So, purity is the, one of the thumb rule to check the
purity, you, once you get the resistance thermometers, you check its ratio. Resistance
at 100 degree centigrade and upon resistance at zero degree centigrade, you measure it
and it should be higher. If it is higher than 1.390, it is better.
5
We can approximate R t as R t equal to R naught 1 plus alpha t plus beta t square; R t

equal to R naught 1 plus alpha t plus beta t square. This is equation number 2. For
pure platinum, alpha equal to 3.94 into 10 to the power minus 3 per degree centigrade
and beta equal to minus 5.8 into 10 to the power minus 7 per degree centigrade
square. Equation is nonlinear. You see, the equation 1, which we have discussed, we
have seen it is a nonlinear equation.
Obviously if you take, I mean above alpha if you take beta, gamma it is nonlinear. But
fortunately, you see the value of the beta here, whatever the beta value you are
getting, beta will be minus 5.8 into 10 to the power minus 7, so the value of beta is
quite small, right and we can see that we can measure also the, how much the
nonlinearity. So, you can see for the, most of the practical cases that is the reason I am
saying that the, however the nonlinearity of the platinum RTD we have calculated at
the 100 degree centigrade is .76% of the full scale deflection, right? It is quite small
for the, most of the ……. most of the applications, right?
6
Now, to calculate this actual temperature we do not have to solve the quadratic
equation. You see, the two basic scientist …….. and Griffith worked over a lot on the
resistance thermometry. They suggested that instead of solving that quadratic
equation that means second order differential, second order equation, because you see
in the most of the cases if I take, even if I take nonlinearity, if I take up to the second
order that means R t equal to R naught 1 plus alpha t that means like this one if I take,
if I take a blank page, so that is okay.
7
So, if I take for the, most of the practical purposes, sorry, if I take a, if I take R t equal
to R naught 1 plus alpha t plus beta t square that will suffice. It is t square, please
note, this will suffice, right? So, you can see that in this type of situations or I can
approximate like this, because higher terms like gamma and all these things will be
quite small and can be neglected. So, amount of nonlinearity will be introduced by
higher order terms. It will be very, very less. So, I can take it up to, if I take up to beta
or t square, it is more than, I have more than whatever we need or accuracy as I told
you, we have seen the .76% of the full scale. So, it is quite small. So, up to beta if you
take that means up to second order t square if I take that is enough, right?
So instead, the resistance temperature relation is rewritten as R t equal to R naught

multiplied by 1 plus C into t pt. This is equation number 3, where C is the mean
temperature coefficient of resistance between 0 degree and 100 degree centigrade.
8
We define the platinum temperature t pt, which is nearly equal to the true temperature
t by the relation t pt equal to R t minus R naught upon R 100 minus R naught into
100. This is equation number 4, where R t, R naught and R 100 respectively denote
the resistances at t degree centigrade, zero degree centigrade and 100 centigrade
respectively. The quantity R 100 minus R naught is called the fundamental interval of
the thermometer. The difference between the true temperature t and the platinum
temperature t pt was accurately given by the parabolic formula as follows.
9
t minus t pt equal to delta t by 100 whole square minus t by 100. This is equation
number 5, where delta, the delta is a constant for that particular specimen of wire.
Now, to derive the equation number 5, we proceed as follows. We have taken t minus
t pt t minus alpha t plus beta t square upon 100 alpha plus 100 square into beta into
100. This actually we got from the equation 2 and 4, right?
So, this can be equal to minus beta 100 square upon alpha plus 100 beta multiplied
with t by 100 whole square minus t by 100. Thus, delta in equation 5 is given by delta
equal to minus beta 100 the whole square alpha plus 100 beta.
10
The value of delta for the specimens employed lies between 1.488 and 2, 1.498. It
should be 2.498. The higher the purity, larger is this alpha and smaller is the delta.
Now obviously, delta I can find from the equation 5 by finding the platinum
temperature t pt, for the boiling point of sulphur which is usually 444.6 degree
centigrade. The reason that we have chosen the sulphur boiling point, because it is
available in the pure form and instead of taking at the boiling point of water, we are
taking little higher temperature
Usually, as you know that, I mean resistance thermometer can be used, especially the
platinum resistance thermometer can be easily used up to the temperature of 650
degree centigrade. So, we can take the temperature of 444.6 degree centigrade. It is
fixed for the normal temperature and pressure and whose true temperature is known
and then, substituting in equation 5, we can find it, right? The effective range of the
RTD depends on the type of wire used. There are various, as I told you, there are three
different, four different types of RTD and what are the ranges and how it will be
effective, can be looked at from this table.
11
Platinum, its RTD we have made in table. Its temperature range is 100 to 600, 650
degree centigrade, good linearity and chemical inertness. That is the most important
thing for the platinum RTD, its chemical inertness. You see here, this chemical
inertness is the most important factor, why the platinum is used typically for making
RTD or even I should say it is a ……. standard, industrial …… standard to use the
platinum RTD, even though we have the nickel and copper and all those things.
Nickel I can go for minus 180 degree centigrade to 430 degree centigrade and minus
200 degree centigrade to 260 degree centigrade and nickel and copper are susceptible
to corrosion and oxidation. This is the problem with the nickel and copper, but you
can use it in some situation. Even the platinum is quite expensive also. If you use a
large number of measurements, I can manage with the nickel and copper and in that
type of situations, I have to use in, in a sheath material or I have to put this entire
RTD in a, in a weld, so that it be protected from the environment, right or instead I
can use a tungsten also, which is rather newer, I should say minus 270 degree
centigrade to 1100 degree centigrade.
12
So, now this is the figure 1, where you can see the typical resistance temperature
characteristics of metals. We can see, here we have plotted R t by R 0 in y-axis,
temperature in the x - axis. Obviously, at zero degree centigrade R t by R 0 will be 1.
You have plotted in the, first one is of nickel. Then you have plotted for copper, then
you have plotted for platinum, then you have plotted for tungsten. You can see that
the platinum, sensitivity of the platinum is rather low.
Even though tungsten is even lower, but the tungsten advantages we are getting the
higher range, so that means per degree change of temperature. the resistor change of
the platinum is smaller than the copper and the nickel, right?
13
Now, RTD circuits, how will I, now actually what is in the RTD circuit? If the
temperature changes, the resistance will change. So, somehow or other, I have to
measure that resistance change, right, so that the resistance change can be calibrated
in terms of temperature. That is the very simple, how we can make the measurements
and how we can measure the, how we can make or how we can make convert this
resistance change in temperature. That is our goal in formatting RTD circuits.
Now, basically these RTD circuits are Wheatstone bridge circuits, right, because
Wheatstone bridge is, as you know, it is very unbalanced. It is very easy, we can find
the unknown resistance with the help of another known resistance. So, it is always
used for the measurement of the unknown resistance. But please note, the simple
Wheatstone bridge you cannot use. We have to use some modified form of
Wheatstone bridge, because RTD is a sensitive instrument. Its contact resistance, all
those things, we have to take care of, we have to count on that, right?
Now, RTD can be connected in one arm of the Wheatstone bridge as shown in figure
2. You see, this is the figure 2, a simple Wheatstone bridge circuit with RTD in one
arm, right? You see, the RTD , this is our, R T is RTD and these two fixed resistances
R 1 and R 2, right and this is our R 3, this is a variable resistance. We will balance
bridge with the, by varying R 3. Either I can take the unbalanced voltage D and
14
calibrate that unbalanced voltage in terms of temperature or I can calibrate this R 3 in
terms of temperature. It can be, in that case it will be a slide wire.
Now, I am telling, potentiometer in that case, it will be simple slide wire, where I can
shift this. That means it look like this. The wire will be like this one and there is a
contact. It will move like this one, right and this length will be calibrated in terms of
temperature. Now, your conventional potentiometer, school level potentiometer, so
where I will have a graduated scale here, so if I move this one, so I can balance it and
this will be calibrated in terms of temperature by which we can measure it, measure
any unknown temperature, right?
At balance, obviously you know that it will be R 2 by R t equal to R 2 by R 1 into R

3. It is very simple. I should not draw it, even though it is a basic Wheatstone bridge,
anyway and for sake of completeness, I am drawing.
15
Suppose I have a circuit like this one, so we have R 1, R 2, this is our R t and this is
our R 3. We make R 3 variable and this is a potential. Suppose this point is a, this is b,
this is c, this is d and suppose this current is I 1 and this current is I 2, right? So, you
see that e ab, e ab will be equal to potential …. across this R 2 into I 2 and e ac will be
I 1 into R 1. At balance, when there is no output at the detector, so I can say I 1 by
into R 1 equal to I 2 into R 2, right? Similarly on this side, because if at balance all the
current will pass through this one, so no current will flow through the detector.
So, on this side, I can write the e db will be equal to I 2 into R t and e dc equal to
same current will flow. So, I 1 will flow through this one; I 1 into R 3, right? So, I can
write I 1 into R 3 equal to I 2 into R t. So, if I divide this equation and this equation, I
will get, suppose this equation number, equation is of some one dash, suppose this is 2
dash, so if I divide 2 dash by 1 dash, I will get, I will get R t, R t by R 2 or multiplied
by R 2 into R 3 by R 1. So, I can put R 1 here and put this in a bracket like this one.
So, you see that if I take now R 2 and R 1, these two resistances R 2 and R 1 equal,
then what will happen?
This will, you will see that this will cancel out. So, R t will be, so in that case if R 2
equal to R 1, so I can write that R t equal to R, R t equal to R 3, right? R t equal to R
3, fine.
16
So, any change in, so R 3 will give you, at balance, I will balance ….. varying R 3, so
obviously R 3 will give you the unknown temperature. So, R 3 should be calibrated in
terms of temperature, fine.
So, this is the balance equation; I have taken R t equal to R 2 by R 1 into R 3. So, R 3
is an adjustable potentiometer. All resistors and potentiometers are made of
manganin, in order to avoid any effect due to ambient temperature changes. You
know, manganin has a lowest temperature coefficients of resistance, so we always
choose manganin to avoid any ambient temperature change, right?
17
Potentiometer R 3 might be calibrated in terms of temperature. Three improvements

must be made in the simple Wheatstone bridge. There is some, simple Wheatstone
bridge you cannot use, some modification you have to need. Why you need these
modifications? We will explain one by one. You see, the contact resistance of the
potentiometer R 3, if you remember the potentiometer, the bridge, basic bridge, the
contact resistance of the …….. is directly in the bridge circuit and this contact
resistance may vary. They vary from fraction of an ohm, but it is quite variable. Since
it is in the bridge circuit, so depending on the contact, your bridge balance equation
will change. That is very undesirable properties, right? This is number one problem.
18
Now, another problem is RTD. You, you see that you, you cannot put the RTD very
close to the point of measurements, right. I need certain amount of distance from the
point of, from the, where I will install the RTD and there should be certain amount of
distance, so where I will make the measurements, right? Even though nowadays all
these intelligent sensors and all those things are coming, where the transmitter itself
are, because you have to, if you want to make the unbalanced voltage as a, as a, if you
want to calibrate this unbalanced voltage as it happen in most of the control systems
that unbalanced voltage in terms of temperature, so we have to convert that in the
current domain, but certain length of wire is there. Even though 1 feet of length or
suppose 1 meter length of wire is necessary, also. These are basically lead wires,
because you cannot install the bridge itself on the sensor.
So bridge will be, that Wheatstone bridge will be starting from, might be 1 meter
away from the actual sensor or actual bulb, I should say ….. bulb, right? So, there is a
variation of temperature along the length of this lead wire, so that will create the
problem, right?
19
Now, current through RTD causes a heating effect, which is variable depending on
the magnitude of the current. The heat thus generated raises the temperature of the
RTD sensor.
Self heating effect - there is an I square R heat loss in the RTD, because it is a
resistance. There is an I square R heat loss in the RTD, obviously any resistance. So
and this self heat is called a self heating effect. You see, it cannot avoid, it, this is, this
20
will be, this remains. So, only thing you have to make a compromise between the
sensitivity and because, I can reduce this supply voltage as I reduce the excitation
voltage to the bridge, obviously what will happen? You will find this self heating
effect will be reduced, because you are reducing the current I in the bridge circuit. But
if you reduce I, current I, what will happen?
You know, you will lose the sensitivity also, but that is not a big problem. Nowadays,
we have, we can easily make the signal conditioning, we can amplify the signals. So,
but we can avoid by, this by, I can avoid the, you see everything it is, it is very simple.
If I say that I am using a, I, I will make the self heating error very less, I will reduce
the current and then no problem, sensitivity will be reduced. But, please remember
whenever we are using amplifier it has inherent noise, so that if we use the amplifier,
so that inherent noise will come in the picture also, right and moreover, if you reduce
the current ,the unbalanced voltage will be not, will be not that small, not that large
rather to operate a V to I convertor.
V to I convertor has some input voltage range. You have to, you have to raise the
voltage up to that range. Only then the V to I convertor, because at the beginning of
this I mean course, you have seen that in all your industrial applications, we do not
transmit the voltage, we always transmit the current. So, even the V to I convertor is
necessary. If suppose zero to 5 voltage, if I say zero to 10 volt, whatever it may be of
your V excitation voltage, so maximum output is zero to 10 volts.
In that type of situations, you know that the current will flow to 20 milliampere. If the
V excitation voltage itself is very, very small, if I make purposefully, to avoid the self
heating error, so it may not work. So, we have to look at that what should be the
minimum V excitation voltage, right? The heating, heating of the resistance due to the
flow of current in it is called the self heating, right? The effect of the self heating is to
make the reading erroneous, quite obvious.
21
Therefore, the current through RTD must be low to avoid the self heating error. Now,
usually RTD manufacturer provides the dissipation constant of the sensor. What is the
dissipation constant? The dissipation constant is defined as the power required to raise
the RTD temperature by 1 degree centigrade, right?
Now, the self heating temperature rise can be found from the following expressions;
very simple, delta T equal to P by P d. What is this P? Delta T is the temperature rise,
22
because of the self heating in degree centigrade. P is the power dissipated in the RTD,
in the RTD in watt and P d is the dissipation constant of the RTD in watt per degree
centigrade. This is a constant for particular RTD. Usually the manufacturer will
supply this value of the P d. Now, all thing will control the what is the minimum
resolutions we can take the measurements? Because you see, if the P is fixed, if the P
d is fixed, then delta is, T is fixed, right? So, in that sense, I can control, I can reduce
it, but I, I can, it will control, the self heating temperature error will control the
resolutions of your instrument or resolutions of your thermometer.
Now, let us look at, as I, even though I discussed the self heating effect, let us discuss
the first problem in the conventional Wheatstone bridge, when I am using the bridge
for measurements of temperature using RTD, because we have seen in the
conventional Wheatstone bridge, we have put the RTD in one arm of the bridge and
put one arm of the bridge and we will see that the, we are varying the other resistance
to get the balance and the problem I told at the beginning that the, the contact
resistance comes directly in the bridge circuit.
So, depending on the, how much contacts you are getting, so it may vary. Suppose if
it is, even if it changes to 1 to 2 or 3 ohm, see it will change your bridge balance
equation, right because I have to change the R 3, again to make the bridge balance.
23
So, I might, this is quite erroneous. It is not undesirable, because contact resistance we
are not, when while we are calculating the bridge circuit, Wheatstone, when we are
writing the Wheatstone equations, bridge equations or bridge balance equations, we
are not taking care of the contact resistance, right? So, that will make the circuits total
erroneous, make our measurement total erroneous. Since it is a very precision
instrument, so I have to take care of all these factors.
So, this is the circuit. You see, this RTD is connected to the bridge by three wires.
You see, three wires means you see here, there is one wire, there is another wire, there
is another wire. Now you, interestingly you will see this R 3 now is known in the
bridge. R 3 is in the bridge, but the contact resistance is not in the bridge circuit, it is
in the galvanometer circuit. It is not in the detector circuit. How do you know it is in
the galvanometers or you can see it is just a multimeter or a voltmeter, is not it, right?
So, if there is little increase in the contact resistance or little decrease in the contact
resistance, nothing will happen. What it will do? It will simply make your bridge
more sensitive or less; your detector more sensitive or less sensitive, you usually use
it, is it not? Suppose in the case, some cases I want to make a, use a milliammeter in a
bridge circuit, which is carrying a current of suppose 500 milliampere, even though
our detector circuits can only, can read 10 milliampere, usually we use a shunt, is not
it, right?
You use a shunt and once I am very close to the measurements, so I remove the shunt,
so that, because I know that bridge is almost balanced. There is a little chance of, I
mean passing a large current through the bridge, through the detector. So, that type of
situation may arise. So, any contact resistance will not, this contact resistance will be
no more in the bridge circuit, so no calculation, nothing, whatever the bridge balance
equation we will write that will be absolutely correct, but it will be in the detector
circuit. So, it will make the detector more sensitive or less sensitive, it does not
matter, it is no way, I mean coming in the picture.
24
So, I am denoting some current, because we have to calculate something. This I 2,
which is the resistance, which is the current through the resistance R 2, I 4 is the
current through the resistance R 4 and at balance, this R4 is, I 4 is also flowing
through R T and since it is, no current is flowing through the detector in balance and
also this I 2 is, full is flowing through the, this R 3 as well through R 1 back to the
battery, right? This is called the RTD is connected to the bridge by three wire method,
right?
Now, the potentiometer R 3 is placed in the detector circuit when it may lie in both
arms of the bridge. The bridge is balanced by adjusting R 3. So, I am writing the
equation I 4 R 4, if I look at the top one, equal to I 2 R 2 plus 1 minus f into R 3 is the
first equation on the bridge balanced on the right hand side or and also I can write I 4
R t I 2 R 1 plus f R 3.
25
So, when f is the fraction of the potentiometer R 3 lying in the arm, R 1 arm of the
bridge and I 1 and I 2 are the current through the corresponding arms of the bridge.
This, actually it should be, sorry this will be I 2 and I 4, it will be I 2 and I 4, also the
bridge. So, it will be I 4, I am sorry, this will be, if I take a, so this will be I 4. I 4 and
I 2 are the current through the corresponding arms in the bridge circuit. That means
the currents which is flowing through the R 4 it is I 4 and current which is flowing
through the R 2 is I 2.
Please note that I 4 is also flowing through the, our, that resistance R t that means
…… resistance of the thermometer and I 2 is also flowing through the resistance R 1
as well as R 3. It is flowing through R 3, then through the, and because at balance, so
detector is not drawing any current. So, at mid balance equations we can write, I, R t
equal to R 4 R 1 by R 3 plus f. If I take ratio R 2 by R 3 plus 1 minus f, right? f is a
fraction of the potentiometer, R 3 lying in R 1 arms of the bridge.
So, obviously in the R 2 arm, the fraction of the potentiometer will be R 3 multiplied
by 1 minus f, is not it? Because, on R 1 arm it is f into R 1 that means I should say
like this one, if I write that means that the f R 3, f R 3 will lie in the R 1, in R 1 arms
of the bridge and 1 minus f into R 3 will lie, 1 minus f into R 3 will lie in the R 2 arms
of the bridge, right? No problem.
26
So, balance is obtained by adjusting the potentiometer R 3, which is calibrated in

terms of temperature, no problem. So, the effect of the contact resistance is avoided,
because the wiper of R 3 or jockey of R 3 is now not in the bridge, but in the detector
circuit, which I told several times before, right? The only effect of the contact
resistance is to cause a negligible change in the sensitivity of the detector, while the
accuracy of the bridge circuit remains very high. This is most important thing.
27
Three wire method is used to compensate for variable lead wire resistance. Now, the
detector circuit connection is now made at the RTD and forms the lead wires, right?
But you know, the lead wires, even though lead wire resistance will remain, in the
case of three wire methods even though contact resistance problem will be overcome,
self heating error will be inherent that will remain as it is, as well as lead wire errors
will remain in the three wire method of RTD. Now, specifications - each lead wire is
made of copper wire of the same diameter and length, so that the each has equal
resistance. This is the specifications of lead wires which will connect from the bulb of
the thermometer to the bridge circuit, right, because it looks like this. I have a
thermometer bulb, R t. This will be connected by bridge, so this is our all lead wires,
so this is our lead wires, right?
28
The three lead wires are run through the same cable, so that they have assumed or
they have the same temperature conditions. This is also necessary.
If the resistance of each lead wire is R L, then the lead wire is in the R t arm of the
bridge and one in the R 4 arm of the bridge. Quite obviously, it is in the R t arm of the
bridge and the R 4 arm of the bridge. The resistance of the detector lead wire has
negligible effect, because it is not coming in the picture. So, because it will make,
29
again it will make more sensitive or less sensitive the detector circuit, because
whatever the resistance in the detector, if it is high, so we can ignore that once we
write the bridge balance equation, clear?
Now, this equation, we rewrite the bridge balance equations for the three wire
method, when I am taking the length resistance of the lead wires, right, in effect.
Previously we have not considered the resistance of the lead wire, so we assume that
each lead wires as the resistance of R L. So, it will be I 4 R 4 plus R L equal to I 2 R 2
plus R 3 1 minus f. So, it will be I 4 R t plus R L equal to I 2 R 1 plus f R 3. If I take
the ratios, R t plus R L R 4 plus R L equal to R 1. If I divide on the right hand side by
R 3, I will get R 1 by R 3 R 2 by R 3 1 minus f, sorry.
30
If the right side of the above equation is made unity, obviously then what you will
get? I will get, if I, R t equal to R 4, right? If the right side of the equation is unity R t
equal to R 4. Now, interestingly when this right side of the equations will be equal or
right side of the equations will be unity, only the cases when my potentiometer at the
mean position that means f is .5, so only at that position of our bridge balance we will
find that the lead wire resistance can be totally eliminated.
So, whatever the temperature variation of the lead wire, so we should not bother of, it
is not coming in the bridge. But any other position of the, our jockey or of the, our
slide wire or our potentiometer wiper, there will be the or any other position, then
50% there will be error, right? Therefore, the balance of the bridge is independent of
the lead wire resistance only at one particular setting of the potentiometer R 3. At one
temperature of the scale that is 50% of the, the compensation is exact and the
resistances must be equal in magnitudes.
31
At other temperature there is a slight and almost negligible lead wire error, when there
are changes of ambient temperature along the lead wires. Ultimately these are, if there
is a change in ambient temperatures along the lead wires, I want to compensate it.
This is not possible for all position of the bridge balance. So, only at 50%, but that is
very quite ….. I mean, it is quite I mean funny. I mean you cannot expect that the
bridge will be always balanced at 50% of its, I mean position, right? f should, cannot
be always .5; can be any position.
32
Now, Mueller bridge or four wire method is a method, where using four lead wires
and the lead wire effect can be totally eliminated in this type of bridge. Even though it
is cumbersome, but for precision thermometer this is the only alternative method and
we have to use it for making the, our circuit. Both you are independent of the contact
resistance, independent of the lead wire resistance. Self heating error will remain as it
is. The lead wire effect cannot be totally eliminated in the three wire method, because
the compensation is not exact at all positions of the potentiometer. However, the
effect of lead wire resistance can be totally eliminated in four wire resistance bulb
method.
Let us look at; the method involves switching the resistance thermometer bulb lead
from one arm of the bridge to the other arm of the bridge, right? We will see how it
can be done.
33
You see, this is our Mueller bridge. Mueller bridge is shown in figure 4. This is the
Figure 4. You can see this is our four lead wires and you see that potentiometer is R
p2 and R p1. R p1 is the, which, I mean, wire used to balance the bridge and R p 2 is
actually, perfectly it is not used once it is set. It is used to make the resistance of the
right hand side of the bridge. That means I want to make total R 3 that means R 3, R,
total R 3 like this one. That means if I take, ….. that means I want to make this R 3
and this R 4 equal that means bridge, resistance up to this, it should be equal to
resistance up to 3. Only then I can write the bridge balance equation. Only in that
cases, only the lead wire resistance will be totally eliminated, right?
The excitation, we will find that what is the different connections and what is the
bridge balance equation? So, in this case, we can totally eliminate our lead wire
resistance problem. You see it looks like this.
34
R p2 is used to make R 3 equal to R 4 in position ‘a’ of the switches. There are two
different positions. Point 1 is connected to 2 dash, 2 is connected to 1 dash and point 4
is connected to 5, right? So, these are the different contact points, which will be
modified and in the case of, right, so in that case, you see the 4 is connected to 5,
right? So it is in the and my bridge balance equations will be like this one: R p1a that
means at the switch position a, the value of R p1 this I am giving the name R p1a.
What is that?
Again I am repeating, it is the, for the switch position a that means all this connection
that means what are those connections? Connections are that I said that 1 is connected
to 2 dash, 2 is connected to 1 dash and 4 is connected to 5. In that situations that I
will, we will use, we will balance the bridge by varying the R p1 and suppose this
new position is R p1a. So, R p1a equal to R 2 plus R 2 equal to R t plus R 1. This is
equation number 6.
35
And in position ‘b of the switches, we have changed the connections. We have

changed the connection. Point 1 is connected to 1 dash, then 2 is connected to 2 dash
and 3 is connected to 5, right? So, the bridge balance equation is R p1b plus R 1 equal
to R t plus R 2, right. This is bridge balance equation.
By adding equation 6 and 7, we get R t equal to R p1a plus R p1b divided by 2; R t

equal to R p1a plus R p1b divided by 2, right? If I add, so obviously this will cancel
36
and we can see that the, if I measure the bridge balance, I mean if at the two, I mean
positions of the switches, if I can measure the value of R p1 that is R p1a and R p1b, I
can totally and add and average it. So, I can totally eliminate the lead wire resistance.
So, this is the beauty of the Mueller bridge, but there are many precautions. You see
that it is suggested that the, all the resistance should be in a thermostat except the RT,
RTD, so that it will equal resistance and all the lead wires should be of equal length.
If we can satisfy all these things, this is excellent thermometry mankind have so far
developed. Therefore, the measurement of resistance RTD is independent of the lead
wire resistance.
Now, platinum, as you know is universally used to make RTD for the following
reasons: there are some advantages of the platinum, I mean which is now de facto
standard, I should say of the platinum. Being a noble metal it is used for precision
resistance thermometry. That means it does not, platinum is stable that is another
important thing. It resists corrosion, it has chemical inertness and it is not readily
oxidized. It is easily workable or it can be drawn into fine wires, because it is
ultimately to make the wires, some of the RTD I have to draw it in wires, so it is
possible. So, this is another advantage of the RTD, why we use the RTD, right? It has
a high melting point; melting point of the platinum is quite high.
37
It is obtained in high degree of purity. This is another advantage of the platinum, the
impure platinum, hardly we will get. So, we will always get pure platinum, because if
the impurity comes, obviously it is, all these inertness, stability, everything will go
away. So, if it is available in the pure form, so it will retain all this original property,
which is quite obvious. Now, these desirable and necessary features however do not
come without effort.
The resistance temperature relationship of platinum alters with little impurity present
in the platinum. This is another disadvantage that means if there is, little impurity is
there everything will be altered, right?
38
Now, construction details of the platinum RTD bulb - you see here the construction
details. It consists of a coil of fine wires wound on a notched mica cross frame. We
will show that. The wire is arranged, so that the good thermal conductivity is provided
and a high rate of heat transfer is obtained. This is, another thing is necessary and the
windings should be made in such a manner that the physical strain is negligible as the
wire expands and contracts with the temperature changes.
This is a typical problem, because you see RTD is made of, it is not a semiconductor
device, so that type of thing. So, what will happen? It is the metal. So it will have, if
the temperature rises, the wire will be stretched. If the temperature falls it will be,
wire will be contracted in length. So, if the temperature rises, already, so if already it
is very tight manner, then what will happen? The strain will be developed. If the strain
develops in the wire, then what will happen? It is as you know that, as we have in the
strain gauges, this property is utilized to measure the strength, because if the
resistance, if the, if the, if the wire is in tension, so its resistance changes, right, which
is undesirable because we are not counting, we are, what we want?
What we want that the, our resistance change will be totally due to the change of
temperature in the thermometer bulb only. That is the reason we have even did so
much of the work - three wire, four wire method for the, the, to avoid the resistance
39
temperature change due to the change of the lead wire. This is the number one thing.
At the same time, we do not want that the change of resistance will be for any other
factor. We want the change of the resistance should be totally due to the change of
temperature, not for any strain or anything, right?
So, for that it is a different, a typical construction will be available. So, we have to use
that type of construction for making RTD.
We can show here; the construction details of platinum RTD bulb is shown in figure.
You see, this is our typical construction details of the RTD. You can see here, this is
our mica cross, you see this is our mica cross. This is the mica cross, right? This is
mica cross here on which the RTD wire, it is wound on like this one in the coil form
so that it will have a least amount of, I mean strain developed while the temperature
rises. We have evacuated space, the mica crossing. This is another problem with this
mica crossing. The mica crossing also limits the temperature of the, high temperature
…...
Even though it is hardly used above 650 degree centigrade and this is the platinum
coil. This is our platinum coil, this is evacuated space and the whole thing is put in a
stainless steel, so that it will be, can prevent the, prevent the any outside environment
40
to react with the, even though RTD is quite inert, but in the case, suppose we are
using other type of nickel, copper or tungsten type of things, so it should be protected
from the environment; it should not react with the environment.
Now, all this thing whenever using, you see one thing is very obvious, RTD is not for
dynamic temperature measurement. That can be only utilized for, you can use it either
thermocouple or thermistor is the best, because thermistor time constant is, I mean
very slow. We have seen that if we increase the time constant of the systems, once
you put a RTD inside a sheath material like this one, so what was the, what is the
effect? It will increase the time constant of the system that it also very undesirable,
right? But for steady state temperature measurement it does not matter. Suppose it
will take 2 minutes to come to the steady state condition, it hardly matters.
One thing is quite obvious, whether you are using thermistor or RTD, it does not
matter, once you use a, I mean you balance a bridge. So, in that type of situation, you
cannot expect a large, I mean small time constant. There will be a time, sometimes it
is necessary to balance the bridge, but if I use it for any dynamic measurement,
suppose in the case of thermistor I will take, instead of making a, the bridge balance I
will take the unbalanced voltage, so that unbalanced voltage I will calibrate.
I will convert in current domain and I will calibrate in terms of temperature. That is
quite obvious. But in this case what will happen that you see, in the case of RTD, it
does not matter, even we use, even if we use an unbalanced voltage to make the
measurements, so what will happen? You see that this, if you put inside a sheath, then
what will happen that your temperature, if your, if your temperature changes, so rapid
temperature changes, I cannot measure it by this type of equipment. In that case, what
I will do? I will simply use some other. So, it is not very used for dynamic
measurement.
It is used for, I mean steady state measurements or very slowly varying temperature
measurements, right, ….. rapid measurements in some situations. You see, any
process you please note that the temperature measurement is not very, I mean very
rapidly measured. In the case of bioreactor suppose if the temperature is very slowly
41
measured, in that type of situations you will find that it, you need only few minutes. I
mean you sample the ……, you measure the signal suppose every 6 degree centigrade
or 10 degree centigrade, not more than that, not more frequent than that that is most of
the cases suffice. In some situation, say rapid temperature changes also necessary to
shut off. Suppose, I mean, I mean very frequent temperature changes also
measurement is necessary, especially where the safety problem is there, where you
have to shut off the power supply or fuel supply to the boiler, so that type of situations
I need some rapid temperature measurement.
So, it is, I mean in the and that type of situation, obviously people will not, will not
use RTD, right? Shut off means it is basically the, any, any protection measurement
means on, off type of measurements, I have to shut off everything. So, that is on off
type of measurements. In that type of on, off control so I do not need that precision
measurement, so it will limit ….. Suppose some, suppose 500 degree centigrade, I
have to shut off the power supply or fuel supply to the systems, I have to supply the,
stop the supply to the heating coil and all those things that can be easily achieved with
some other temperatures and not for RTD. But RTD needs very precision
measurements and only problem is the self heating error.
Other things can be easily resolved like four wire and all these things and it can be,
that is nowadays very standard. Please note that ….. RTD is, is used in the industry;
…. RTD is to be used with the transmitter itself. For all, whatever the unbalanced
voltage you are getting, so for the standard RTD, I mean transmitter is there. That is
utilized to make your measurement further. That means it will convert that voltage,
unbalanced voltage in the current domain and that can be utilized to control, because
that is to be again converted to the digital domain, because the signal which we will
get in analog and that is to be fed to the computer through the …. chord and our
signal will come out of the ….., convert in the current domain, transmit. Again at the
receiver side you convert in the voltage domain, then you use it to some actuator,
right? Some action is to be taken and on the final control elements valve is to be
operated. So, this is all about your RTD, one of the most accurate temperature
sensors.
42
Preview of next lecture.
Good morning! Welcome to the lesson 10 of Industrial Instrumentation. In this lesson,

we will study LVDT. The full form of LVDT is the linear variable differential
transformer. It is basically an inductive based sensor or inductance based sensor and
primarily it is used for the measurement of displacement. However, I can use it for
measurement of pressure and other process parameter also. In this particular lesson,
we will consider the, lesson I mean the basic concept of the LVDT, its circuitry, its
signal conditioning circuits and its basic constructions and how will you design an
LVDT? These are the basic things which we will discuss.
Let us go to the instrumentation lab of IIT - Kharagpur and have a look on the LVDT.
You see here, this is an LVDT - linear variable differential transformer. It has one
primary and two secondary. The secondaries are usually connected in opposition to
get a null position and …. after doing the phase sensitive demodulations, I can sense
on which side of the, which side of the null position my, the core lies? You see here
the core is there, soft iron core which will move like this, which will move like this,
43
so and I am moving, I mean it, it will move like this and since, there are two
secondaries, what will happen if you put in opposition? Obviously, you will get some
output voltage and at null point it is at the, exactly at the null positions the two
secondary coils will be, output voltage will be nullified and I will get zero voltage. At
any position other than that will give you nonzero output voltage. So, LVDT already
we have discussed in details, so this is actually pictorial view of the LVDT.
Welcome back to our classroom again. So, the lag network, as I told you, we need
some lead and lag networks. So lag network look like this one.
You see, this is the phase, this is the circuit if there is a, there is a lagging phase angle.
If there is a, phase angle is lagging, in that situation we will connect … This is our lag
network. So, it will kill the lag and it will have, the output voltage will be in phase
with the input voltage. It is not very important, in most of the cases we will find,
right? So, this is the lag network for compensation of leading phase angle.
44
Prof. A. Barua
Lecture - 10
LVDT
Good morning! Welcome to the lesson 10 of Industrial Instrumentation. In this lesson, we

will study LVDT. The full form of LVDT is the linear variable differential transformer. It
is basically an inductive based sensor or inductance based sensor and primarily it is used
for the measurement of displacement. However, I can use it for measurement of pressure
and other process parameter also. In this particular lesson we will consider, the lesson, I
mean the basic concept of the LVDT, its circuitry, its signal conditioning circuits and its
basic constructions and how will you design an LVDT? These are the basic things which
we will discuss.
Even though I have not written, the contents of this lesson will be the LVDT, its
construction, its signal conditioning circuit, lead network, lag network and a phase
sensitive demodulation circuit, as well as some design equations of LVDT. So, at the end
of this lesson, the viewers will know the basic construction of the LVDT, what are the
signal conditioning circuits have to be used, what are the precautions you have to make it,
its sensitivity, all those things will be studied in details in this particular lesson.
PAGE \* MERGEFORMAT 1
It is the sensor for displacement measurement. You see, this is our basic LVDT. You can
see here that we have a primary here. There are three windings. Look at very carefully,
we have one primary and two secondaries, two similar exactly identical secondaries,
right? Actually here, should be also, if I take, should be also, so the excitation voltage
will be here. It will come here, an E ex excitation voltage. So, it will be, any voltage
between 3 to 15 volts and 50 to 20 kilohertz and two different secondaries.
Here you can see the basic construction. You see here the, on a …… the primary are
wound, primary is wound and the two secondaries are wound in this fashion, right? Later
on we will see that we will connect these two secondaries in opposition that will come
later on and you see there is a core, iron core which will move up and down, this iron
core which will move up and down. Interestingly you see that if this core moves up, so
there will be no linkage between this and this, so the mutual inductance between this coil
and this coil will increase and the mutual inductance between this coil and this coil will
decrease. So, at geometrical null position, mutual inductance between this and this, this
primary and this secondary, and this primary and this secondary will remain the same,
right?
Now, let us go to the instrumentation lab of IIT Kharagpur and have a look on the LVDT.
You see here, this is an LVDT - linear variable differential transformer. It has one
primary and two secondary. The secondaries are usually connected in opposition to get a
null position and through, after doing the phase sensitive demodulations, I can sense on
which side of the, which side of the null position my, the core lies?
You see here, the core is there, soft iron core which will move like this, which will move
like this, so and I am moving, I mean it is, it will move like this and since there are two
secondaries what will happen if you put in opposition? Obviously, you will get some
output voltage and at null point it is at the, exactly at the null positions, the two secondary
coils will be, output voltage will be nullified and I will get zero voltage. At any position
other than that will give you nonzero output voltage. So, LVDT already we have
discussed in details, so this is actually pictorial view of the LVDT.
Welcome back to our classroom again. So, we have seen the construction of an LVDT
and how it works.
Actually this core will, if you look at this core, this core will move up and down; this core
will move up and down and I will get the different mutual inductance. Also, the voltage
also will be changed and there will be the phase difference between the input and output.
So, that is actually we, we have written in a, in a nice way. You see, the flux formed by
the primary is linked to the two secondary coils inducing an, a, it will be an, an AC
voltage in each. An iron core which moves through the bobbin without contact provides a
path for magnetic flux linkage between the coils. The position of the iron core controls
the mutual inductance between the primary and two secondary coils; this is most
important thing, right and usually we will find the entire …… assembly we put, in some
cases we will put in another casing, so that it will not be influenced by the other external
magnetic field; obviously, it depends on the magnetic fields.
So, in some cases we will find that the, we put the entire assembly, so that the external
magnetic field will not affect the output voltage, because the out, we want that the output
voltage will be, will be a function of the position of the core of the LVDT.
Now see this is the, we have given the, if you look at the screen, you see we have given
the excitation voltage, a sinusoidal voltage in the primary and the two secondaries are
identical. So, obviously I will get two secondary voltages e 1 and e 2. We are assuming in
these cases that the core is in null position that is geometric null position, so I am getting
excitations of equal phase difference between, in the, between input and output. This is
our input or excitations. I am not giving, telling it input, because input in the case of
LVDT is the displacement which we will give on the core and output is the output
voltage and in this case it is e 1 and e 2, right?
Now, you see that there is an equal phase difference between input. This is, the two
secondary coils are identical in nature, so there will be equal phase difference in the two.
So, you have seen that if you not consider, might be there will be some attenuation.
Obviously there will be no question of ……, so there will be some attenuation we will
get, but we have ignored that part. So, we have assumed that the, I will get the equal
magnitude in both the secondaries.
Now a secondary is, now you see a LVDT, we never use two secondaries separately; we
always put the two secondaries in opposition. What is that thing? If you look at the board,
our digital board, you will see here that excitations we are giving here. This is the core
positions. We have 3 to 15 volts, 50 hertz to 20 kilohertz we are giving and output
voltage. You see the two secondaries here, these two secondaries in opposition, there is
no connection at this point. Please note there is no connection at this point. So, this is one
secondary and it is connected to the bottom and it is coming like this one.
What is the advantage of this one? You see, in these cases advantage is, when this core is
at the geometric null position you will find that the output voltage e naught will be zero.
So, if I put this core slightly up and slightly down, I will get a non-zero output voltage,
right? So, if I put this core slightly up, I will get a non zero output voltage. Because, what
will happen? In this case, the two secondary output voltage are exactly in oppposition and
this two will cancel out. I will not get any voltage at the output, right?
Now, see this is the output voltage we have drawn with respect to time, when for the
different positions of the cores. Since these two cores, since these two secondaries are in
opposition, we will find, if you look at the board, are in the oppositions, we will find that
core, this is the core at the null positions, theoretically this should be exactly zero. So, see
here, even though we are giving some excitation voltage which is sinusoidal voltage, I am
not getting the output voltage since the two secondary voltage will cancel each other.
Whereas, you see, when this core is above null, I will get a voltage like this one with a
phase shift. If the core is below null, I will get a phase shift, a 180 degree phase shift,
separate or opposite to the core above null, exactly, got it?
This is necessary, because in some situations, I must know in the, which position of the,
core lies in which position, whether it is above null or below null. You see here, what will
happen? It is slightly, we are getting like this one, right? So, you see, these two phases
are exactly out. That means if this is 180 degree out of phase and there is a phase
difference between this input voltage and this or excitation voltage and this and excitation
voltage and this, but these two voltages, when the core is above null and when core is
below null, the two phase difference is exactly 180 degree out of phase.
Now you see this is the, now we have, I mean taken the magnitude of the output voltage,
right? We have taken that is the reason we have written here e naught mod, absolute
value of the output voltage we have taken. Now, I change the position of the core, I
measure the output voltage right? For, suppose for a x centimeter, 1 centimeter, what is
the output voltage I have plotted, then for x equal to 2 centimeter what are the output
voltage I have taken, for x equal to 3 centimeter what are the output voltage I have taken,
right and these voltages I have plotted like this, clear?
So, you see, we are getting a linear curve that means the relation between the
displacement and the output voltage when the core is in, when the secondaries are in
oppositions, exactly we are getting a relationship of linear, I mean linear relationship we
are getting. So, at the geometric null position, so you see, we are not getting any output
voltage, right? Now, this is the, what is plus x, what is minus x? Plus x, if I move the core
in one side of the, suppose I mean, if I take the, looks like this one.
Suppose I have a core here, right, this is my output voltage, is not it? So, this is the
position x is equal to suppose zero, if I put it up suppose in this position, so if I put it, I, I
am taking it a positive displacement, if I put it down I get a negative displacement. So, in
that respect, I am calling it as plus x and minus x. Now, if I put on the other side of the
null position, I will also get the voltages, obviously with phase shift, but if I take a
magnitude, so I won’t, the phase shift, I cannot identify that phase shifts. Just if I take the
absolute value, so you see that I will get a curve like this one and this, I mean you see it is
falling down, because at that position what will happen?
I am taking the core totally out of the bobbin. In that type of situation, it will fall like this
in both the cases. It is, when it is going up it is going out and it is when it is going down
like this one, you see I am getting a straight line, straight line relationship like this one.
So, this is our, I mean the absolute value, the, that means the output voltage of the LVDT,
when in series opposition versus the opposition of the core, right?
Now applying KVL if I take the equivalent circuits, if I, if you look at, I mean, I mean
applying the KVL on the primary side, I can write you see i p R p plus L p di p by dt
minus ex equal to zero and i p equal to ex R p plus L p D. Second, if I go back, if I look
at, I mean that will be better, if I go back here, this is actually should be, yes, I am sorry,
you see this is our equivalent circuit of LVDT.
We have taken here, this inductance is L p. Primary inductance, this is primary resistance
R p, this is excitation voltage ex and this is the resistance of the, there are two
secondaries, we are taking the, if we assume that two exactly identical, so I can take this
inductance is L s by 2, this resistance is R s by 2. So, we will get a simplified equation.
This is the, M 1 is the mutual inductance between the primary and the one of the
secondary.
This M 2 is the mutual inductance between primary and one of the secondary. This is the
output voltage e s1 and this is the output voltage e s2. That means voltage source here
and this R s2 and this, for the time being we have not connected, so we are connecting an
R m, a meter impedance. When there is a, current is flowing through the, this circuit,
right? See, if I now draw the, I mean write the KVL around this loop, I want to write the
KVL around this that means in the primary side I will get, if I go back, see if I, now see e
s1 and e s2 will try induce voltage in the primary.
However, this will nullify, since it is in series opposition, right? What is e s1 and e s2?
So, already we have seen, so if the e s1 and e s2 will try to induce the voltage in the
primary and this will nullify, since it will be, is in series opposition. Now, if I go back,
you see, can see here, so you see, this is our basic equation, I am sorry.
Now, see here, if I apply a KVL around primary sides I will get the equation like this and
already we have seen the equivalent circuits of an LVDT. So, if I apply the KVL, I will
get equation like this one - i p R p plus L p di p by dt. Now, you see and so, i p equal to
ex, where D is that operator. So, as before we know what is D? We know that D equal to,
D equal to d by dt, right, d by dt, so that is the operator. So, I can come back, you see
here and ex equal to R p plus L p D. An induced secondary voltage e s1 is M 1 di p by dt
and e s2, e s2 and e s1, already we have seen what was this?
If I go back, you see, e s1 is this one and e s2 is this thing, right? So, this is the induced
secondary voltage, right? So, this is our actual equation. So, you see the induced
secondary voltage. This is applying on the primary sides. But in the secondary side this
will induce a voltage like this one, e s1 equal to e s2; e s1 equal to M 1 di p by dt and e s2
equal to M 2 di p by dt. So, all, with respect to the, the primary current, right?
Now, you see, this already we have discussed, now let us look at, so e s1 and e s2 will try
to induce voltages in the primary. However this will nullify, since it is in series
opposition. So, find the problem. So, the net secondary voltage, if I look at e s or e
naught, whatever you say, equal to e s1 minus e s2. This will be equal to the, if you look
at the expressions which you have seen, say M 1 minus M 2 di p by dt, because e s1 equal
to M 1 into di p by dt and e s2 equal to M 2 into di p by dt, so e s1 minus e s2 M 1 into
minus M 2 di p by dt. The net mutual inductance M 1 minus M 2 varies linearly with the
core position.
For a particular core position, e naught equal to e s equal to M 1 minus M 2 D operator L

p D plus R p ex, right, excitation voltage. Now, you see, this e naught by ex I can write,
excitation output voltage by, if I normalize the output, it will be M 1 minus M 2 by R p D
operator tau p D plus 1, where tau p equal to L p by R p, right? Now, you see here that
this is the time constant. So, this is the first order system, so it is a time constant of the
system. Now, e naught by ex if I look at, now obviously what will happen? Where is the
x, where is the position?
Position is here; in the inside there is, it is M 1 and M 2, because this is directly
proportional to the position of the core, is not it? This M 1 and M 2, it, it directly depends
on the position of the core. So, you see this output voltage also depends on this value of
the M 1, so all other things will remain constant. M 1 will vary, M 2 also will vary, while
the, if I move the core up, then what will happen? If they move the core up what will
happen?
The M 1 will increase, M 2 will decrease. If I decrease the, if I make the core down, M 2
will increase and M 1 will decrease, is not it? So, there it is, this x is hidden within this M
1 and M 2. So, obviously, this output voltage will be proportional to the position of the
core R x, right? So, if I write in the frequency domain, so it will be like this one, M 1
minus M 2 R p square under square root omega tau p square plus 1 with an angle phi. It is
not come, so the angle will look like this, right? So, this is the magnitude, this is the
magnitude part and this is the, your phase part. So, what is the phase, let us look at.
So you see, the phase is equal to pi by 2 minus tan inverse omega tau p, right? So, the
phase also depends on the frequency, tau p is a constant. So, in fact, I can make omega
tau p constant, because tau p does not depend on the position of the core, is not it? Tau p
depends on L p and R p. You see, here this is constant for particular L p dt. Omega is the
excitation frequency, if I keep constant, it will also remain constant. So, phi equal to pi by
2 minus tan inverse omega tau p, this will remain constant for a particular excitation
frequency.
If ….. I can make it zero also. If I can cancel, if I make tan inverse omega tau p equal to,
exactly equal to pi by 2, so this will be cancelled. There will be no more, but sometimes
this is not be possible, so you will use some lag network or lead network to kill the phase,
phase, I mean lag or lead, right? So, I can make the phase compensation with simple RC
circuit. Let us look at that.
If the input impedance of the voltmeter, now you see the, all the analysis, so far whatever
we have made, we are assuming that the, there is, no current is flowing through the
secondary. Current is not flowing through the secondary, because we are assuming that
the, we are measuring these impedances. We are connecting that secondaries open circuit,
but most of the cases it is not true. Secondary we must, we will connect some, but you
can say to some, I mean if I do some I mean approximation, this is also true, because we
will connect a very high input impedance.
Suppose if I connect a CRO, cathode ray oscilloscope to measure the voltage output e
naught, so obviously it has a very high input impedance. So, in that type of situation, so R
i will be very high. So, I can assume almost that the, your secondary is open circuit. But
if I assume, so i s will b e zero, but obviously if I take the finite case, if you take, assume
that the, there is some current is flowing through the secondary. So, in that case we
assume that the finite resistance R m, meter resistance is R m and a current i s is flowing
through the secondary, because if you take a finite resistance R m, so obviously the
current also will flow.
Previous analysis we have seen that there is, no current is flowing through the secondary.
Only two voltages e s1 and e s2 was induced due the, due to the mutual inductance
between the primary and the two secondaries, right? So, in that type of the situations my
equation is, if you look at the digital board it will look like i p R p plus L p Di p minus M
1 minus M 2 Di s equal to minus e to the power e ex, subscript ex equal to zero. So, if I
can write this equation, M 1 minus M 2 this will be equal to Di p plus R s plus R m i s
plus L s Di s equal to zero. So, if I, I mean take all these, so I can write the simplified
equations in order by e naught by, I mean e ex D M 1 minus M 2 whole square L p L s D
square plus L p R s plus R m L s R p D plus R s plus R m R p, right? This is equation
number 3.
So, lag network, as I told you, we need some lead and lag networks. So, lag network
looks like this one. You see, this is a phase, this is a circuit. If there is a, there is a lagging
phase angle, if there is a, phase angle is lagging, in that situation we will connect …..
This is our lag network. So, it will kill the lag and it will have, the output voltage will be
in phase with the input voltage. It is not very important, in most of the case you will find,
right? So, this is the lag network for compensation of the leading phase angle. If the
phase angle is leading, you can use this lag network. Lead network, similarly if it is
lagging phase angle I can use a lead network.
So, lead network just interchange with capacitors and the, and capacitors and the
resistance, I will get a lead network. So, it is a leading phase angle, it will look like this
one.
The frequency of excitation voltage applied to the primary winding can range from 50
hertz to 20 kilohertz.
The power requirement is usually less than 1 watt and the sensitivities of different
LVDT’s vary from 0.02 to .2 volt per millimeter of displacement per volt of excitation
applied to the primary coil, right? This is typical sensitivity, it is a very sensitive device.
Now you see, most important thing of the LVDT is, you see it is a displacement sensor,
as I told you. But, directly you can, displacement in the, in other case you can use it for,
you can, so many you can use it for a load sensor, you can use it for pressure.
If you use in conjunction with a diaphragm you will find it is used as an, I mean as a
pressure sensor. You can use it as a flow meter if you connect a, you will see that if you
connect to the bob of a rotameter, you will find that this will be used as an, as a flow
meter, as the indication of the flow. So, all these different, I mean applications are there,
but directly it is actually measuring the displacement.
Now, other type of, simplest form of displacement measurement, it is a just

potentiometer. A simple potentiometer if I look at, you will see if I connect like this one,
suppose I have a like this one, so let me take another page that will be better or I will take
a different colour. You see, what will happen, so I have, sorry let me write in a new page.
A simple resistance, a rheostat or a potentiometer, pot what we will call, suppose I have a
wiper or jockey, in the case of potentiometer we will call it wiper, in the case of I mean
rheostat we will call it jockey. So, if the jockey moves in this direction, then what will
happen? I will give some excitation voltage, suppose and I am measuring the voltage on
this one e naught suppose this excitation, so what will happen? This output voltage will
be a function of this displacement, is not it? Suppose this is x, so this simple, a
potentiometer can also be used as a, I mean as a, I mean displacement sensor.
However, the difference between the potentiometer and first of all the potentiometer and
the LVDT, number 1 is the, its resolution. Resolution of LVDT is extremely good. Please
note that resolution good means its numerical value will be less and another thing is you
will find that sensitivity in the, both the cases I can change. I can change the excitation
voltage, ex the excitation voltage of the, if I look at the digital class, I mean if I change
the excitation voltage, my output voltage will also be increased or I can modify this thing.
I am not ….. I am talking of the resolution. Resolution ….. you see that whenever a
potentiometer if I look at the digital class you will find that the potentiometer here what
will happen?
You see it will, even though I am showing here that it is getting contact only one, one,
one turns, but you will see that practically potentiometer looks like this one, is not it? So,
it will, always this wiper will contact more than one or two windings. So, resolution will
basically depend how closely you make the windings. You are, it is, it is possible for you
to make the windings as well as whether the, whether the, this wiper or jockey is getting
connected to one winding or two windings, but that is not the case in the case of, I mean
that is not the case for LVDT.
So, LVDT, what will happen? You see there is no contact between the, between the
wiper which is basically a core, movable core and either primary or secondary. So, the
resolution I could not, so it is, I mean, I mean it is first of all life will also increase. You
will see here, in the case of, if you look at the digital class you see that what will happen
that in this case that this core will, this wiper will move and after few use you will find
there is wear and tear it is called, it is, this contact will be a, will become noisy because
the contact also depends on how much pressures you are giving. That type of case does
not arise in the case of, case of LVDT, because LVDT I have core, this bobbin and inside
the bobbin this core is moving, is not it? So, there is no physical contact. So, life is almost
infinity. You can say the mean time …. failure is very, very high.
Moreover, resolution is also high because resolution here is very, very small, I mean
small means, I mean, I mean good, extremely high because in this case you see there is
no, there is no question of even though whatever the windings we have on the outside
that does not depend, that, that does not matter how fine and how close those windings, is
not it, because the resolution here depends on the, the magnetic linkage between the two.
It does not depend on the physical position or the physical number of coils which is
getting touch with the wiper. So, that is the great advantage of the LVDT.
So, that is we have seen, we have written, you see there is no contact between the core
and the coils. Therefore, the friction is eliminated thereby giving infinite resolution and
no hysteresis. This is another important, you see in the case of potentiometer there will be
certain order of hysteresis you will find, but there is no hysteresis in this case.
No contact also ensures that life will be very long with no significant deterioration of
performance over this period. The small core mass and the freedom from the friction give
the sensor, gives the sensor the some capability for dynamic measurements.
Now, phase sensitive demodulation is very important in the case of LVDT, because
LVDT we have seen that if the core moves up and down and I mean, if I look at, you see,
if I, we have seen that I have a primary and two secondaries like this one. You see, this
moves, so it moves up and down. What will happen? I will get a output now, is not it? So,
if I move up, I will get some sinusoidal voltage, if I move down I will also get the same
sinusoidal voltage. But how will I know on which side of this one, whether it is lying,
sorry, whether it is lying up the meet position, null position or below the null positions?
For that reason, I have to make the phase sensitive demodulation of the circuit, I mean of
the output. If I can make phase sensitive demodulation that means in that cases we will
separately fully rectify the output to a secondary voltage, then make the algebraic sum of
the two. So, one side I will get positive voltage, other side I will get a negative voltage,
right?
You see, this is the phase sensitive demodulation circuit. You see the, each half is
actually rectified with a full wave rectifier. In all the cases, whether the, this is positive,
this is negative we will find the current is flowing through this direction. If this side is
more positive and this side is more negative, current is also flowing through the same
direction through the resistance R. Similar is the case in the lower half that means in the
other secondary. So, these two voltages, these two voltages we are making algebraic sum.
If I can make this algebraic sum, then what will happen? You will find that the output
voltage, polarity of the output voltage will depend on the position of the core that means
whether the core is above the null or below null, right? If it is above null, it will be
positive if we assume positive; if it is, this core is below null it will be negative, right?
So, this is the phase sensitive demodulation circuit.
Now you see the dynamic displacement measurements. If the LVDT is to be used to
measure dynamic displacement, the carrier frequency should be 10 times greater than the
highest frequency component in the dynamic signal. We will show the phase sensitive
demodulation output after few slides. The excitation voltage ranges from 3 to 15 volts,
right, because in some situations, I may need to make the dynamic measurements even
though frequency should not be that high.
So, for dynamic measurement, the frequency of excitation voltage should be much
greater than the frequency of core movement. If a frequency ratio of 10 is to 1 or more is
feasible, a simple RC filter may be adequate.
Now this is plus e, obviously this is … since it is fully rectified, so it is dc. Even though
our excitation voltage is AC, now our output is DC because the, our two secondaries we
have fully rectified and we are adding it algebraically, so the output voltage will look like
this one. I am sorry, let me take another page. So, right, so this is our output voltage. This
is positive, this is minus e naught, this is x, plus x, zero, this is minus x. As I told you,
plus x, minus x is nothing but the above null or the one side of the null is positive. If one
side of the core movement is positive, one side of the null, of the core movement is
positive, other side of the, movement on the other side of the null is we are taking
negative. So, you see we are, distinctly we are getting a linear relationship, but along with
the change in the magnitude. This is ……..
Now, I can tell it is above null or below null looking at the voltages; this is very
important. Look at the digital class. If I look at, you will find that if the core below null,
so obviously what will happen? So the, I will get a negative voltage.
Now, this is, you see the, again we are coming back to our RC circuits which is used for,
I mean our simple RC circuits to make our, for dynamic measurements. In that case the
carrier frequency should be much higher that means excitation frequency should be at
least 10 times higher than the maximum expecting dynamic signal in the LVDT, right?
That means what is dynamic signal here? Dynamic signal is the, movement of the core is
the input, right?
Input is the movement of the, so that input component, frequency of that dynamic input
component should be 10 times lower than the frequency of the excitations of the LVDT,
clear? So, if I have that type of, I can have a simple RC filters. So, this is first order RC
filter, so that will suffice that I can extract the original signal. So, the response will look
like, you see, e 1 by e naught. This is the output voltage e 1 by e naught, so it is minus 20
dB per decade. So, it is a ….. frequency omega equal to 1 by tau, right? So, low pass
filter and its frequency response.
Now, another thing, even though I say it that in the case of LVDT you see that the two,
two secondaries are exactly identical, but you know it is very difficult, absolutely
difficult to make two, I mean secondaries exactly identical and the harmonics also will be
present in the secondaries, right? So, this will create problem. That means this will make
our, our circuit, this will make our circuit in such a way that I won’t get zero voltage at
geometric null position, right? That is the thing we have seen.
The harmonics in the excitation voltage if I look at and stray capacitance between the
primary and secondary will lead to non zero voltage at geometric null position of the
core. The mismatch of the two secondaries plays, also plays a role in generating this null
voltage as shown in the figure 12, sorry.
You see, this is a null voltage. It is supposed to be zero, but there is an, so the figure 2 is
the null voltage. So, this you see here, the null voltage it is the voltage non zero, I mean
voltage I am getting at the geometric null position, right, because making two identical
secondary is almost impossible, right? So, what will happen that will be, some non zero
voltage will remain. Actually what will happen if I look at the, I mean vector diagram it
looks like this.
So, the two secondaries should be like this one. You see the, this is, if the secondary
current I s1, this is I s2. If these two cancels it will look like, but actually what will I
have? Due to mismatch, you will find one will be like this one, other will be like, this is I
s1, I s2. So, the resultant will go like this one. In this case the resultant will go like this,
right? If I take a different colour, so the resultant will go like this, whereas in this case the
resultant will go like this. So, clearly I am getting non zero voltages, right?
Now, this null voltage is a nuisance, we must kill it. So, there are various circuits
available to kill this null voltage. One of the voltages, you see the null voltage reduction
circuit when the centre tapped primary excitation voltage source is available. If you have
the centre tapped primary excitation voltage source, I can use this type of circuits, where
you have seen that this centre tap is grounded, sorry this is not grounded. Actually this is
the core motion, so it will come like this one, it will come out like this one, right and you
see here, so I have a, centre tapped primary source is available. So, this is, the centre
tapped is grounded with the ground of the one of the output terminals and I am getting
outputs. In this case, case we will find the output voltage will be, the null voltage will be
zero, right?
Now, figure 13, in next figure we will show that the, if the centre tapped excitation
voltage is not available, how it we can kill this null voltage? It will look like this one.
You see the figure 14, the null voltage reduction circuit, when the centre tapped primary
voltage source is not available. You see here this will vary this position. You see centre
tapped is not available, two potentiometers it is dropped. This is grounded, this again in
the previous case it is grounded, so this will vary. We will vary the position on the, this is
potentiometer. It is all the, I mean voltages has come across the potentiometer. We will
change the position on the core, position of the wiper, so that the output voltage will
become zero.
So, figure 14 shows actually method for null voltage reduction when centre tapped
excitation voltage is not available. Here, the potentiometer is adjusted until the minimum
null voltage reading is obtained. If it cannot be zero, we can make, we cannot make it
zero, make it minimum, so that that will be that means at zero null position, that some
minimum voltage is, we will get. That will suffice in most of the cases.
Now, how will you design this LVDT? So, this thing must, you must see.
This is our, see schematic of our LVDT. Even though you see that these are the two
primary windings, this is a primary winding, because you see this is basically a cylinder
sought of thing, right? It is cylinder, I mean here we have the, this core, you see this iron
core, so this is our primary. This is the coil, like this one; it is a primary and two
secondary, identical secondaries. Now you see, this length of the secondary we have
given S, this is length of the primary we have given P.
Secondary, two identical secondary of equal length and r i and r zero is the, this r i is the
inner radii and inner, r i and r zero are the inner and outer radii of the LVDT or assembly
I should say and l a is the length of the armature or length of the core, right? This is our
movable core, please note.
Now, net induced emf of the secondary coil is, will be given by this. You see here,
so this is our net induced emf in the output: e j omega j omega I p 4 pi N p N s mu naught
px upon 3 s natural log of r o by r i 1 minus x square by 2 p square - equation number 4,
where omega is the frequency of excitation voltage in radians per second and I p is the
primary current, N p, N s are the number of turns of primary and secondary windings.
r naught and r i is the outer and inner radii of the LVDT assembly, I should say and
displacement of the core from the, x is the displacement of the core from null position,
from the, right from the beginning we have used that one and mu naught is the
permeability of free space - 4 pi 10 to the power of minus 7.
Now, if you look at the nonlinearity term, this is our non linearity term. You look at, this
is our nonlinearity term, is not it? x square minus upon 2 p square this is the nonlinearity
term in the LVDT, right?
So, the non linearity term x square minus by 2 p square in the equation 4 is dependent on
the length of primary winding P for a desired range of X max and the error due to
nonlinearity, I mean epsilon, the length of the primary winding is given by this.
P equal to X max upon root 2 into epsilon. So, this is actually the length of the primary.
So, we will get from this expressions ……. The length of the secondary winding is also
given by this expression: s equal to P plus X max. The length of the core and the length
of the secondary are kept little more to accommodate the small spacing between the
primary and each secondary windings, right?
This is necessary and you see, if you look at, so this ratio, in fact this will be ratio, the
ratio of r i by L a is about 0.05 and the ratio of r o by r i varies between 2 and 8. The
number of secondary turns should be as large as possible to produce larger sensitivity and
since it is likely to be connected to high input impedance amplifier, quite obviously and
the meter, the secondary windings can be of finer wire, right?
Now, you see that in practical cases we will find it is very difficult to actually
commercially if you are not available, I mean LVDTs are not available, you can make
your own LVDT. Even though, we have made it in our laboratory also according to your
need, but it is an excellent and it is, it is an application, even though I said that the, it has
applications in the measurement of, I mean process parameters measurement, that means
it is usually used for the measurement of displacement and along with the, sorry along
with the displacement, with the, it is used for the measurements of the pressure if I use a
diaphragm gauge and it can be used for the measurement of, in the measurement of I
mean flow. So, in the case of diaphragm you see how it is used. If I look at, you will see,
here you see, fine?
What they do? If I look at the digital class, so I have a stretch diaphragm, right? So, if I
increase the pressure on this side, you see this diaphragm will be stretched like this. We
will study this diaphragm ……. Now you see, this interesting thing is that that the
linearity of the diaphragm depends on the how much is the displacement. Less is the
displacement we will have the better linearity that in pressure versus … If I now connect
a core of LVDT like this one primary to secondary, right, output voltage I will be getting
excitation ex. Now you see, what will happen? As the pressure increases I will get the,
this core will move from the null position. I will get a non zero output voltage e naught. I
won’t need any phase changing demodulations of that type of thing.
This is not necessary because in the case, in these cases diaphragm will move only in one
direction; it will never move in this direction, clear? So, this will give you the output
voltage, the non zero output voltage, right? Now, this is the one applications of LVDT.
Other applications of LVDT if I look at, you said that we have seen that the rotameter in
many places, so we have a core and a channel like this one. So, the float is here, so the
flow is in this direction, it is coming out in this direction. So, what they do? With a ……
ceiling they connect a core here, same similar two secondaries in opposition. So, if I draw
it nicely, it looks like this.
You see, so I have a, so I have a bob here and it is flowing like this one, the ceiling is
there, so output voltage now e naught, sorry e x. So, what will happen if the bob moves?
This float moves, I will get a non zero output voltage. So, the voltage will increase and
the output …. quiet obvious. In the case of rotameter also as the flow increases, this float
will go up and up. It is another example of an LVDT. I can put some more examples.
One of the examples of the, another I mean, sorry pressure sensor we will find that, you
know that our …… gauges basically depends on a this type of pressure measurements, a
C tube, okay let me draw it nicely.
We have a C tube like this one. So, this tube movement I can, this tip I can connect an, a,
if I connect a core of the LVDT here, again this will come here and this will go here, e
naught this is the excitation. What will happen? You see, if the core moves then what will
happen? If the, if the pressure increases, as you know this tip will try to move out. If it
moves out, so within the, what will happen? So, this position of the core will move out,
move from the, change from the null position. So, I will get a output voltage. So, even
though measuring pressures, I am getting electrical output. Even though, I told on
previously in the case of …. gauges, it is very difficult to get the electrical output. But
here you can see, I can get the electrical output here by using an LVDT. So, this is the
several application of the LVDT, even though it is basically used for the measurement of
displacement.
Moreover it is, as a biomedical applications even when the child is, when it, it is, I mean
it is in mother womb, so the, its respiratory signals also can be collected by the, using
LVDT, right? So, these are the different applications of an LVDT. Now LVDT, as I told
you, basic I mean the advantage of this type of devices is, its sensitivity is also there. I
can increase the sensitivity increasing the excitation voltages and I can increase the
resolution obviously, which is much higher than the conventional, our other displacement
sensor. So, whenever I need a better resolution, I need a better sensitivity, we have to use
LVDT instead of, I mean instead of simple potentiometers or linear potentiometers,
which is used for measurement of displacements.
With this, I come to the end of this lecture of LVDT.
Preview of next lecture
This is lesson 11 of Industrial Instrumentation. In this lesson we will study the capacitive
transducers.
You see, the capacitance transducers, the contents of this lesson - capacitance transducers
as a whole, displacement transducers. Capacitance transducers in the sense that the basic
principles of the sensor we will discuss and capacitance transducers are basically used as
displacement transducers, level gauge or level sensor both liquid and solid, differential
pressure transducers, because this differential pressure transducers are utilized in the, we
will see later on in the flow measurements, the DPT or differential pressure transmitter is
basically a differential capacitive transducer. Then, we have pressure pick up. We will
also discuss the pressure pick up as well as we will solve some problems also, right?
And at the end of the lesson the viewer will know, viewer will know the basic principle
of capacitance sensors, linearization of the sensor, differential measurements, right? This
basically we will discuss in this.
Now, see the capacitive transducers, a capacitor consists of two conducting metal plates
separated by an insulator, right? These metal plates can be either a, just a rectangular
plates, very thin rectangular plates or it can be a cylinder also. It looks like that you see it
can be a two, just two parallel plates like this one. It could be two parallel plates, in
between we have a dielectric, right and so it happens that a capacitor consists of two
conducting metals plates separated by an insulator or dielectric medium. When a voltage
is applied to the metal plates equal and opposite electric charges appear on the plates,
right and interestingly this can be a, these plates can be a cylinder also. That we will see
in, I mean later on. It is not necessarily, it can be a, it should be rectangle, it can be
circular also, the plates might be cylinder also. That means it can have a sensor like this.
One plate is like a one cylinder like this one and inside there is another plate which looks
like this. So, this is another. So, we can take out the, from here and here two terminals
and if you measure the capacitance, this is also a capacitance sensor.
So, let us now go to the instrumentation lab and see some capacitance based sensors.
This is another application of the capacitance sensor which can be utilized to measure the
displacement. Usually capacitance sensors are based on the three different criteria. One
you can vary the area of the plates, you can vary the displacement between the plates that
means separation of the plates and also you can vary the position of the dielectric
medium inside the capacitor. The particular gauge which you are, you are looking at now
basically depends on the displacement.
You see, if you, if you make the displacement, here the separation between the capacitor
plates changes. So, obviously if the separation changes, the capacitance value also will
change. So, that, so if you put in a bridge, so obviously bridge unbalanced output also
will change. So, that bridge unbalanced output can be calibrated in terms of the
displacement, because ultimately we are not measuring the capacitance, we want to
measure the displacement, which is our input that is converted to the change of, change
of variation of the capacitance and ultimately which will give you some unbalanced
voltage. This is another application of the displacement sensor based on the capacitance.
Welcome back to the class room. Basically, in instrumentation we will find this
capacitance sensors have basically three uses. One is as a displacement sensor as you
have seen - either permittivity variations or the variations of the area or variation between
the, separation between the plates - number 1. Number 2 is the level gauge or level
sensors. That means level will increase that means the permittivity will change. So,
obviously the permittivity will change and the capacitance will change and third is the
differential pressure pick up, which is extensively used in all process industry for
measurements of flow, okay DP transmitter. This is, basically it is called DP transmitter.
We have seen that if I change the capacitance of the differential pick up, I will get a
change of voltage that can convert voltage to the current and transmit it, right? So, these
are the basic three uses of the capacitive sensors instrumentation, but we have other
applications, as I told you earlier also. We have discussed also, that in this case that we
are using the sensor as a, as a pressure pick up. That means some pressure measurements
also we can utilize this type of capacitance sensors. Capacitance sensors are very, I mean
very nice to use, because it is independent of temperatures and all those things. That is a
great advantage of this one.
It can be used in the corrosive environments, all those things are very much true, but
please note another two most important things of capacitor sensors that those cables
which is connected to this should be, we have taken, so the parasitic insensitive
measurements you have to make, because whenever we are measuring very small
capacitance value, whether you are using measuring in a bridge or LCR meter, it does not
matter, so the, the parasitic, I mean capacitance will influence your measurements and it
will pick up the signals like 50 Hertz signals and all those things. So, you should be very
careful about those parts. So, that non-shielding should be very much good, so that if at
least one part of the shielding should be, should be grounded, so that the, you can make
the measurement faithfully.
With this I come to the end of this capacitive sensor. Thank you!
Prof. A. Barua
Lecture - 11
Capacitance Transducers
This is lesson 11 of Industrial Instrumentation. In this lesson we will study the

capacitive transducers.
You see, the capacitance transducers, the contents of this lesson - capacitance
transducers as a whole, displacement transducers. Capacitance transducers in the
sense that the basic principles of the sensor we will discuss and capacitance
transducers are basically used as displacement transducers, level gauge or level sensor
both liquid and solid, differential pressure transducers, because this differential
pressure transducers are utilized in the, we will see later on in the flow measurements,
the DPT or differential pressure transmitter is basically a differential capacitive
transducer. Then, we have pressure pick up. We will also discuss the pressure pick up
as well as we will solve some problems also, right?
1
And at the end of the lesson the viewer will know, viewer will know the basic
principle of capacitance sensors, linearization of the sensor, differential
measurements, right? This basically we will discuss in this.
Now, see the capacitive transducers, a capacitor consists of two conducting metal
plates separated by an insulator, right? These metal plates can be either a, just a
rectangular plates, very thin rectangular plates or it can be a cylinder also. It looks like
that you see it can be a two, just two parallel plates like this one. It could be two
2
parallel plates, in between we have a dielectric, right and so it happens that a capacitor
consists of two conducting metals plates separated by an insulator or dielectric
medium. When a voltage is applied to the metal plates equal and opposite electric
charges appear on the plates, right and interestingly this can be a, these plates can be a
cylinder also. That we will see in, I mean later on. It is not necessarily, it can be a, it
should be rectangle, it can be circular also, the plates might be cylinder also. That
means it can have a sensor like this. One plate is like a one cylinder like this one and
inside there is another plate which looks like this. So, this is another. So, we can take
out the, from here and here two terminals and if you measure the capacitance, this is
also a capacitance sensor.
The ratio of that charge to the voltage is the capacitance.
3
Now, see the capacitance of a parallel plate capacitor is proportional to the area A of
the plates. Area can be anything, can be the, can be circle, can be, can be a cylinder
and inversely proportional to their separation d. Separation between the plates are d.
In the case of cylinder it will be separation which is the difference of the outer and the
inner radius of the two plates and neglecting fringing, you know the fringing effect is
there always in the capacitance. If we express, that means it lies that if you have two
plates always it will be like this one.
Suppose I have plates, two plates, so the lines will be like this one and the end, since
it will repel each other, so it will be straight, but this will be like this one. So, if we,
this is called the fringing effect and actually the, for the fringing effect exactly you
cannot get, I mean mathematical expression for this fringing effect. But, if you neglect
the fringing, then the mathematical expression of the capacitance looks like this.
4
C equal to epsilon naught into epsilon r A by d Farad. F is farad. This F is basically

Farad and where epsilon naught is the permittivity of free space; it is a vacuum and
with a value of 8.854 10 to the power minus 12 Farad per meter. Epsilon r is the
dielectric constant of the material of the gap. For air, epsilon r is equal to 1 and A is
the area. So, it should be in meter square and d is in meters, right?
The capacitance is a function of a shape and so, we can see that capacitance is a
function of a shape and size and permittivity and alterations of A, d or epsilon r
5
causes a change in capacitance as shown in figure 1. We will show the figure 1. So,
that means you can see that if either of this one, either A, d or epsilon r changes, we
will get the change of capacitance. See, if you measure this capacitance value either
with a bridge or some LCR meters, so that change can be utilized to make it a
displacement sensors, pressure pick up and level gauge. So, these are the basic, three
basic applications of the capacitive transducer. See, if we can measure the change in
capacitance value it can be utilized as displacement, level and differential pressure
sensor. This actually I want to tell.
Now, here you see there are various types of capacitance we have shown. These are
basically as a displacement sensor. You see, the two area between the x and y are
cylinders. You see this x and y are cylinders basically, x and y and you see the
dielectric medium might be air also. There is no problem. So, if I push it this way, you
see if I push this y, x is the fixed plate, x is the fixed plate and y is the movable plate.
These two can be cylinders. If I push it like this one it can be simple parallel plates
also.
Suppose I have two parallel plates, right? You see here, suppose I have a parallel
plates, two parallel plates and if I shift this parallel plate, suppose previously it was in
this position, now we have shifted to this position. So, area between this one has
changed, is not it? Previously the area of cross section was A, now it will be less.
6
Only this much of area is there. So, we can see that the, excuse me, that capacitance
has been changed due to the change of area.
Instead of taking plates I can take two cylinders also; so, where if I vary y that means
this way, so the capacitance value will change. This capacitance change can be
measured by some LCR meter or by some bridge, so that capacitance change can be
calibrated in terms of the displacements; as simple as that. Now, you see that d can be
varied also. That means the separation between the plate, this is the separation. This is
actually if you look at, this is d, right? This is our d, so that d can be changed also. If I
change d, obviously what will happen? The capacitance will also change. The more
the value of d, so the less will be the value of the capacitance. So, quite obviously this
can be also utilized to make a displacement sensor, right? So, here the area between
the two plates changes. Plates can be either plates or can be, it can be two cylinders.
Here the distance between the plates vary, so both way the capacitance change will be
there. So, that can be utilized to make our displacement sensor.
Now geometrical variation, you see, here actually what will happen?
You see instead of, instead of the displacement between the plates or the area between
the plates, now the permittivity between the, or dielectric medium between the two
plates are varied. You see here, actually these are the two plates. Actually this is the,
7
those who are familiar with the heterodyne receivers, you know that in the heterodyne
receivers we are, our radio receivers, we change the capacitance value to tune
different radio stations. So, basically what will happen? There, actually I am changing
the permittivity between the two plates, because this plates if you look at very
carefully on the, on this one, see these plates, you see here these plates, these moves.
This is a dielectric medium. It might be plastic also, insulator. So, obviously dielectric
medium must be an insulator. So, this will vary in this position. So, the plates are
fixed. If the plates are fixed neither area between the plates changes; so, only the
dielectric medium changes.
So, what will happen? If I take it out you see that the, there will be more air between
the two, two plates and there be less, less the plastic, I mean that between the plates
and if I move it this side, so because it is a semi circular plate, entire separation
between the two plates will be filled up by this plastic. So, the epsilon r, if you look at
C equal to C naught epsilon naught epsilon r A by d what we said in Farad, so this
epsilon r will change. So, this epsilon r it changes. So, obviously I will get a different
value of the capacitance. Here actually what we are doing that we are changing the
permittivity between the medium, right? So, by changing the permittivity, either
permittivity or area or displacements, I can get the capacitance sensor.
Here you see, in the, this is another example of a liquid level gauge. You see the, it is,
cylindrical in shape, right and we have a one inner rod also, right? So, this inner rod
will consist one plate, this is one plate and this, this cylinder, out of cylinder will be
another plate. Now, if you pour liquid here, you see the dielectric medium between
this plate and this plate consists of two different materials. One is air, another is
liquid, right? So, if the air, if it, if it, if it is more, if you fill up this, I mean this tank,
what will happen? You see that there will be more liquid within, in the tank and less
of air in between, in between the two plates. So, the dielectric medium changes. So,
obviously what will happen? The capacitance value will, also will change.
This principle, this method, this can be utilized to make a level sensor or level gauge
and this is very interesting that this can be utilized to make a level of solid as well as
for the liquid, right?
8
Now this all, in all the previous cases we have seen that this output is single ended
output, right? But, we always prefer differential output. There is some advantage of
the differential output, because non-linearity will be less. So, the output, unbalanced
output, because ultimately if I want to make a very accurate measurement, please note
that I have to put that capacitance in a bridge; see, if I put in a bridge, so if I, due to
displacement or due the permittivity change that means due to area change,
permittivity change or the, or the, or the displacement between the, separation
between the plate changes, what will happen? Your output will change, right? But, in
the case that if you, if you, if this output is more that will be always better for us,
because our, that will simplify our measurements.
Exactly that thing we are doing in that case of differential output. You see the
capacitance change can be C plus minus delta C or C plus minus plus delta C, all the
same. See, here the X and Y and Z are cylinders, area between the plates varies. Now
you see, their …. capacitance between this plates and this cylinder and this cylinder
again between this cylinder and this cylinder. Now, if I shift this cylinder on this side
what will happen? Area between this plate and this plate will increase and area
between this plate and this plate will increase, decrease, right?
Now, you see, if it is exactly in the geometric mean position, then what will happen?
The capacitance between X and Z and Z and the capacitance between Z and Y will be
9
identically should be equal, right? That means capacitance between X and Z and
capacitance between Y and Z should be exactly identical. Now, if we push it on this
side, then what will happen? You see here, if I push this one this side, then what will
happen? That capacitance between X and Z will increase, because it is getting more
area between the plates and capacitance between Y and Z will decrease. Similarly, if I
put on this direction, capacitance between X and Z will decrease, because the area
will be less and the capacitance between Y and Z will increase, right?
See, if I put this in the two arms of the Wheatstone bridge, obviously my bridge
unbalanced output will be doubled. So, that is the advantage of using the differential
measurement. It is a differential output for a, from a capacitance sensor.
Here also you see that I have two plates, fixed plates P and Q and one movable plate
is M, right? So, I have a displacement. Separation between the two plates are d and if
the area is, these are all parallel plates, if the area is actually, exactly area is d here,
here, here it is d, here it is d, then what will happen? The capacitance between this M
and Q and capacitance between P and M will be exactly same. If this is exactly at the
meet position, so the capacitance between this and this and capacitance between that
means capacitance between P and M, capacitance between Q and M will be identical.
10
Now, suppose I push this plate M slightly upward by displacement of, by
displacement of X meter, then what will happen you see? If you displace, the
capacitance between M and Q will increase and capacitance between M and P will
decrease. Again the same advantage I am getting that if I put in opposite arm of the
Wheatstone bridge, I will get the more unbalanced output, right? Here, the distance
between the plates vary, geometrical variation. Again the same thing that means I am
getting differential output here, right?
Now, X, W, X, Y and Z are cylinders. Now, in this case you see the permittivity
variations, variable capacitor with differential output. You see the, what will happen
here? This M, this is, this is Z basically, right, this outer bigger cylinders. We have
capacitance between X and Z. X is a solid, solid rod, Y is also a solid rod and that Z is
a cylinder. So, what will happen normally? Between, you have capacitance between X
and Z and capacitance between Y and Z. So, if the dielectric medium, … Now, here
the dielectric medium or the W, I am varying that means its position is getting varied.
If the position is getting varied, then what will happen? You see, here if I move it in
this side, so there will be more dielectric medium between X and Z and there will be
less dielectric medium between Y and Z. So, the capacitance between this X and Z
will increase and capacitance between Y and Z will decrease. Similarly, if again this
dielectric W is, we shift on the right hand side, suppose I am shifting it on the right
11
hand side, then what will happen? That means if I show, show it like this, if I shift this
on right hand side, then what will happen? The capacitance between W and Y,
because I am getting more dielectric medium, so W, sorry capacitance between Z and
Y will increase, because the, I am getting more dielectric medium and capacitance
between X and Z will decrease, because I will get the less dielectric medium. So, this
again I am getting a differential output. So, the capacitance can be, you can utilize as
a, as a displacement sensors and you can have a single ended output. Also I can have
a differential output, right?
One or more of the plates are not connected to ground and therefore, electrostatic
screening is required to avoid pick up, because capacitance measurements, because
these are basically the …. of the capacitance are very small and there is always stray
capacitance. So, all these things you have to take care while you are making a signal
processing circuitry. Usually this 50 cycles being supply frequency, all these will be
picked up. So, we have to use proper shielding to get the capacitance value. Thus,
screened cable connection to the capacitive sensors can be source of error, because it
might change its capacitance when there is a movement between the cable conductors
and cable dielectrics.
So, let us now go to the instrumentation lab and see some capacitance based sensors.
12
This is another application of the capacitance sensor which can be utilized to measure
the displacement. Usually capacitance sensors are based on the three different criteria.
One you can vary the area of the plates, you can vary the displacement between the
plates that means separation of the plates and also you can vary the position of the
dielectric medium inside the capacitor. The particular gauge which you are, you are
looking at now basically depends on the displacement.
13
You see, if you, if you make the displacement, here the separation between the
capacitor plates changes. So, obviously if the separation changes, the capacitance
value also will change. So, that, so if you put in a bridge, so obviously bridge
unbalanced output also will change. So, that bridge unbalanced output can be
calibrated in terms of the displacement, because ultimately we are not measuring the
capacitance, we want to measure the displacement, which is our input that is
converted to the change of, change of variation of the capacitance and ultimately
which will give you some unbalanced voltage. This is another application of the
displacement sensor based on the capacitance.
Welcome back to the class room. Now you see that we have seen some sensors
already in the laboratory.
So, we are, we have to observe, we must have observed that the variation in area A or
separation d requires physical connection to the moving part, while permittivity,
epsilon r variation does not because it is, it is not very simple. I mean, even though I
am showing in the diagrams, I mean two plates are there and we are making the
separation, physical implementation, it is not that easy. It is quite difficult to happen
that type of making connection; that are two plates will be exactly parallel to each
other, while you are, I mean making the separation. That mean plates, I mean two
14
plates are like this one; see, two plates are there. So, if I make a separation like this
one, I will get different value of the capacitance.
But actual implementation, it is not very easy, it is quite a task. I mean plate might be
this way or that way. So, entire mathematical calibrations, I mean entire calculations
will be wrong, so that the calibrations also will be, will be different. Similarly area
also, if I …. I mean while we are changing the …. two plates will, two parallel plates
should, should be equal exactly parallel, but in one fixed plate that is separate thing.
But once it is movable, it is very difficult to keep them parallel all the way long, all
the way in the entire, entire range of the movement. So, all these things you have to
think of, right? But in the case of permittivity, there is no such problem arises,
because it is simple.
There is no change or mechanical change of the systems, most of the cases. Especially
in the level gauge we will find this is very much true, but in our case, while I am
showing one differential that we are changing the another cylinder by changing the
permittivity that is also the same problem we will face, right? With the variation in A
or epsilon r have a linear operating range, it is usually 1 to 10 millimeter, although the
capacitance transducer is mostly used for small displacement, where d varies. So, it is
usually, it is used for the very small displacements; it works for very well. For larger
displacements we have many other sensors like potentiometers, presets and all these
things. Usually we can use this capacitance sensor for a smaller displacement
measurement.
15
The most common form of variable capacitor used in the displacement transducer is a
parallel plate capacitor with variable air gap. So, just simple air gap in between and I
am varying the separation between the plates; that we have seen in the laboratory. The
problem of non linearity between the distance and the plates d or capacitance c is
shown in the following figure. You know that, if you can look at very carefully that
the, the relation between the capacitance and the d is the change of capacitance and d
is not linear, right?
16
You see here if I take a, see if I, because you have seen that C, C equal to epsilon
naught epsilon r A by d, right? You can see this relation, see obviously it is not, the
relation between C and d is C is directly, inversely proportional to 1 by d. So, it is a
non-linear relation, because if you take a differentiation, then you find it is a non-
linear relation, right? But, we can see that with the use of simple op-amp we can
linearize it. This is very much necessary, because in many applications we need this
type of measurements.
See, this is a op-amp based circuit. You see here that you are using one op-amp and
this is a capacitance which will be variable. That means whether you are using as a
displacement sensor, level sensor, it does not matter, we put on the feedback path of
the op-amp and C f is the fixed capacitor. I am giving an excitation voltage of e ex
and taking the output from e naught, if you assume, e naught and ground. If you
assume that the, this, this ideal op-amp it is offering a large input impedance, so that it
is not drawing any current, what about the current coming in this frame through this
path? So, quite obviously, immediately I can write the output voltage like this; I can
write the output voltage like this.
17
This is a, I mean this is an ideal op-amp if you take, assuming the op-amp as an ideal
devices, I can immediately write e naught by e ex is equal to C f by C x, right? So, it
is e naught by e ex equal to minus C f by C x, right and quite obviously, where C is
the fixed capacitor, C f is the fixed capacitor, C x is the variable capacitor or
capacitance sensors actually and immediately what we can write e naught equal to C f
by d epsilon naught into epsilon r into A. So, you see that we have linearized it. The
output voltage is linearly proportional to d.
So our, I mean measurement is becoming more simple if we use this type of

arrangements. But obviously if I want to, in this case if I want to vary A that will not
be a very, I mean good proposition, because in this case you will find it is again
becoming inversely proportional to A, output voltage is inversely proportional to A.
But …… situations we will use only d or the separation between the plates that can be
utilized to make our sensor. So, obviously output voltage will be proportional to d.
18
It is clearly seen that the output voltage is now directly proportional to the plate
separation d. Thus linearity is achieved for both large and small motions. In
commercial instruments this e ex excitation is 50 kilo Hertz sine wave of fixed
amplitude, amplitude is fixed, right? The output e naught is also a 50 kilo Hertz sine
wave. This will be wave, I am sorry, this will be wave, this will be wave.
So, in a differential capacitance pressure transducer, the plates are circular. Plate M is
thin diaphragm across which the pressure differential will go. I think I have skipped
19
something, let me go back; okay, no, no problem. If I go, yes, yes, yes, so in a
differential capacitance pressure transducer, the plates are circular and plate M is, the
figure will come in the next slide, plate M is a thin diaphragm across which the
pressure differential to be measured is applied, right? With equal pressure applied to
the both pressure ports that is diaphragm is in the neutral position and the bridge is
balanced and the output voltage e naught from the bridge is zero.
How does it look? You see this is our example of our differential pressure pick up.
You see here this is our movable plates. This plate is movable plate. It is moving in
this direction or this direction. If this pressure is higher than this pressure, okay,
this pressure is higher than this pressure, so the diaphragm will be stretched in this
direction. It will look like this one and if this pressure is higher, then the diaphragm
will be stretched in this direction, right and X and Y are the fixed plates.
Actually this is, if you look at, it is a section of a, section of a sphere. Actually if you
look at a section of the sphere and this inner side is coated with metals, so that inner
that coating will work as a plate of the capacitor. Now, this type of pressure
transducers has a, differential pressure transducers has tremendous applications. In
modern day instrumentations you will find everywhere I am using flow sensors and
flow sensors as you know, in the most process industry these are huge in number and
in all flow measurements specially our, in the case of orifice meter, Venturimeters and
20
pitot tube, we use some differential pressure pick up, because these sensors, these
output is to be transmitted to the control room and this signal is to be transmitted to
the control room. So, I need this differential tapping.
It looks like this one in the flow measurements. Why this is important? You see, we
have a pipe here and this orifice. We will discuss orifice later on, so we have a
tapping here, we have a tapping here, right? So, liquid is flowing like this one. So,
there is a high pressure, this is a high pressure area port and this is a low pressure port.
This high pressure port is connected here and this low pressure port is connected here.
Then, what will happen? This diaphragm will be bent like this one and then what will
happen? If it bends like this one, since I told you this inner side is coated with metal,
what will happen?
This is a section of a, section of a sphere, so inner side is coated with a metal. Then
what will happen? You see, this plate will move this side and the capacitance between
this plate and this plate will decrease and capacitance between this plate and this plate
will increase or capacitance between X and M will increase, capacitance between M
and Y will, sorry capacitance between X and M will decrease and capacitance
between M and Y will increase. So, if this differential, if we put on a bridge we will
see in the next slide we will put on a bridge, I will get a large bridge balanced output
that can be converted to the current of 4 to 20 milliampere and transmitted. That 4 to
20 milliampere of current will correspond to the 3 to I mean 15 psi of pressure, right?
So, that is the reason I am saying stretched stainless steel diaphragm movable plates
of capacitors. Actually in that process you will find that we are using pressure
capsule, actually capsule, so that it is not necessarily directly you have to give. The
pressure will come down here, because this liquid will directly come down here. So,
in between there is another liquid, so the, it will in turn make a pressure on this side,
so that it will be stretched like this one, right?
Now, you see, these are our signal conditioning circuit. This is bridge circuit of the
differential pressure pick up.
21
We have X and Y and M movable plate and two capacitors. What will happen? If this
is increased, this will decrease. That means if this increases, I am sorry, if this
increase, let me go to the previous slide.
Yes, so if one pressure is greater than the other, so obviously what will happen? The
diaphragm deflects in proportion and giving an unbalanced output at e naught and in
proportion to the differential pressure, quite obviously. For the opposite pressure
difference, e naught shows a 180 degree phase difference. What is that?
22
If you look at very carefully, you see here what will happen? You see here that if I put
on a bridge, so it is very difficult to know, if you look at, because if this is, if, if the, if
the diaphragm moves in this direction, then what will happen? The capacitance
between X and M will decrease and capacitance between M and Y will increase,
right? Now, if the capacitors, if the, suppose the pressure on this side increases, so the
diaphragm is stretched like this one. Diaphragm in that case will be stretched like this
one. Then, what will happen? You will find that the capacitance between this and this
will increase and capacitance between this and this will decrease, right, is not it?
23
So, therefore a phase sensitive demodulation is necessary, right? Even though I am
saying, you see that I am saying that like this one, let me see, so I said that I have a
capacitance here, I have a capacitance here, right? So, if I put a signal here and
measure the output voltage e naught here, so if this is increased, this will decrease;
that is no doubt about that, so I will get an unbalanced voltage, right? Now, if this
decreases and this increases also I will get unbalanced, but how will I know that
which side it is?
So, for that reason, we have seen as, it as, it as we did in the case of LVDT, I need
some phase sensitive demodulation. With a phase sensitive demodulation this type of
problem can be easily, I mean solved.
Now, see this method is basically, I, I should tell that this method is basically for,
therefore this method allows measurements of static deflections, because it, if it is a, a
static deflection, so I can easily measure by this type of method, right? Such
differential capacitor arrangement also shows the considerable greater linearity than
do the single capacitor types. You will find that the linearity will be more in this type
of capacitor arrangement right?
24
Now, capacitance pickup or capacitor microphone is a very important devices, is, is a,

I mean you will find that this is basically capacitance sensor and the separation
between the plates changes in that type of situations. You see here, this is a
capacitance sensor. This separation between the d will change and I am measuring the
output e naught. Now, only problem with this type of capacitors that is pickup is that
it is only for the dynamic displacement measurements.
For static displacement measurements, I won’t get any output. I am using an

excitation of e ex, e subscript ex. So, what will happen that if there is a, if this is fixed,
so there no current will flow through the capacitors. So, I will get the, I won’t get any
output. Whenever this moves, as it happens, I want to say, talk, this microphone
works. You see, this microphone is also capacitive transducers. You can see here this
microphone is also capacitive transducers. So, when I, I am talking, so this
capacitance between the plates ((in)) changes. So, I will get a output here, right?
25
We know that, you see that we have already seen, because you see what will happens
that C equal to, I said equal to epsilon naught epsilon r A by d. So, from that
equations I can write delta c by delta d, c by d, right? Similarly, delta c by c, delta d
by d, right? So, that means, sorry this will be delta c by c, this will be, no delta c will
be there, so delta c by c, delta d by d.
So, when the capacitor plates are stationary with separation d naught, no current flows
and the output voltage will be zero, as I told you earlier, right? If there is, if there is
26
then a relative displacement d i from the position d o, a voltage e naught produced and
is related by this expression.
e naught by d K D tau d plus 1. As you know that, sorry, I am sorry, they should go
away, I think I should escape from here. So, what will happen?
You see that this is our magnitude and phase plot of the capacitance sensors. You can
see here that magnitude is e naught by d i, because e naught is the input, d i is the, e
27
naught is the output and d i is the input, right? It is from the nominal positions, so I
am making and ex is the excitation that is not a, I mean that is not the input to the
systems and the phase plot will be like this and the magnitude plot will be like this
one and the phase plot of this one will be, look like this one, right?
So, we can see here that the frequency range of this one will be, after some certain
point there is a large phase shift at the beginning, when the omega is low. Otherwise,
there is a, I mean as the frequency increases, so I, we can see here this is also, please
note this is also omega. As the frequency increases, the phase shift is also becoming
zero and your representation or of the original signal also will be better and better.
Now, as I told you, as I told you earlier also, this arrangement does not allow
measurements of static displacement, since e naught is zero in steady state for any
value of d i. For sufficiently rapid variations in d i, however the signal e naught will
faithfully measure the motion.
28
You can see that e naught by d i j omega k j omega tau upon j omega tau plus 1,
where tau is the time constant of the systems, for omega tau much, much greater than
1, right? Instead of omega, I am writing omega tau, omega tau much, much greater
than 1 e naught by j omega equal to K, right?
So, microphone usually need not measure sound pressure smaller than 20 Hertz,
because usually we talk between 20 Hertz to 20 Kilo Hertz. That is the reason we
have to tune all musical instruments also within this range 20 Hertz to 20 kilo Hertz.
29
So, above such arrangement is perfectly satisfactory, because as you have seen, the
plot, phase plot and the gain plot, so low frequency production is not very good, but
the high frequency is quite good. So, if you go above 20 Hertz, so there is a faithful
reproduction of the original signal. Our goal is to reproduce the original signal that
means the sound pressure on the microphone.
To make omega tau much, much greater than 1, for low frequency it requires a large
tau. So, to make, how can I make omega tau much, much greater than 1? Only if the,
at low frequencies only possibility is, tau will be much, much greater than 1. Only
then in that case it will be more than 1. If it is more than 1, so I will get the input
output. My relation is, the input output relations will be K, that e naught by d will be
equal to K, approximately will be equal to K, is not it? So, that is possible, right?
Now, that we can see it again; here, you see that we can go back. You see here, so no
problem, we can see here that if the, this e naught by d i is better and better faithful
reproduction of the original signals, so omega tau for low one.
30
Yes, for a given capacitance and d naught, the value of tau can be increased only by
increasing R. Typically, typically R will be 1 megohm or more. So, if you can make 1
megohm or more, obviously omega tau will be quite large. So, even for low
frequencies, so I can faithfully reproduce my original signal, right?
Thus to prevent loading of the capacitance transducer circuit, high input impedance
amplifier is required, a buffer amplifier is required. So, there are various types of
31
buffer amplifiers. It is not very big deal, nowadays to have, to have a good buffer
amplifier. We can have a good buffer amplifier and can make the reproduction.
Capacitive sensors are now widely used for the, as I told you earlier also, the, for the
levels of liquids and the solids in powdered or granular form.
So, sensor is suitable for use in extreme conditions - liquid metals at high
temperatures we can measure it which is not possible in other, liquid gases - low
32
temperatures also we can measure. Corrosive liquids like acids and all these things,
where I mean or caustics, where it is very difficult to measure with, with any other
sensors, but capacitive sensors ….. You can have two stainless steel, stainless steel
containers and within that I can pour some corrosive liquid and make the level
measurements. High pressure process, where the other sensors will fail in that type of
situations also capacitor sensors works, work very good, because it is totally
independent, measurement is totally independent of the pressure of the liquid, right?
Now, we will solve some problem, so that for the digital class, let us look at, so the
problem is like this one.
You see we have, let me take a new, we will solve some problem later on also.
Problem 1 - problem is like this that I have a sensor, right. Rather, I should first draw
this.
33
It looks like that I have a capacitance, a differential capacitor, sorry differential

capacitor like this one, right? So, separation is d, this separation is also d. So, there is
a movement of this one, which is we are assuming delta d, right? This is problem
number 1, right? So, I have plates. So, this is also I am taking connection, there is
another connection and there is another connection here. So, this will be C 1, will be
equal to C naught. C naught is the, when you, when the capacitance plate is exactly at
the meet position, capacitance between that two are same. So, that is C naught plus
minus delta C and this will be C 2 capacitance between this and this equal to C naught
plus minus delta C, right?
If I draw the equivalent circuit of this one, it looks like … Our goal is that is a
variable area, this is a displacement type sensor, displacement type sensor with
differential output, differential output, right, where area of the plate, of the plates
equal to A meter square, d is the nominal separation between the plates, between the
plates and epsilon is the permittivity, permittivity. Actually in this case we have a
combined epsilon naught and epsilon r. So far in the text whatever we have discussed
epsilon naught and epsilon r, we have combined in epsilon here, right? No problem
and delta d is the displacement. It is a displacement sensor, so delta d is the
displacement, right?
Now, what you have to do?
34
Calculate, number 1, calculate delta C by C naught, where C naught is the nominal

capacitance and delta C is the change in capacitance corresponds to, to delta d, right?
Also we have to find, number 2 show that, excuse me, the output voltage e naught is
equal to V by C f C naught 2 delta d by d upon 1 minus delta d by d square, right?
Now, the push pull configurations we will show in the next slide. The circuit looks
like this.
35
I have an arrangement here. This is grounded. C f, this is C 1, this is C 2, this is plus
V, this is minus V, this is our, this is our output e naught, right? This is our circuit. So,
we assume two current. Suppose this current is i 1, this current is i 2 and total current
is i, right and op-amp, we are assume is an ideal op-amp, so it is not taking any
current inside, right?
See, if this is a situation, now we can write, let me take a new page, see for the part a
or part, sorry it is not part a, I will take a new page; part 1, C equal to epsilon A by d
or delta C equal to, I can write delta C by delta d into delta d, is not it, which we can
write quite obviously, minus epsilon A, here the epsilon, please note, it is epsilon
equal to epsilon naught into epsilon r; epsilon A by d square into delta d with first
order, I mean without, with approximation first order approximation. So, this will
become again minus C naught into delta d by d.
Now, you see, so I can write delta C by C naught equal to minus delta d by d, right?
Now, current i, actually you can see here i 1 minus i 2, right? So, which we can write
equal to V, voltage omega C 1 minus omega C 2 or which we can write V omega,
omega take out, it will be C 1 minus C 2, right?
36
See, if I take a new page, so the output voltage e naught equal to i minus i upon
omega C f equal to minus V C f C 1 minus C 2 equal to minus V C f epsilon A upon d
plus delta d minus epsilon A by d minus delta d. So, this will become V by C f into
epsilon A upon d square minus delta d square into 2 delta d. So, this will become V by
C f into C naught 2 delta d by d upon 1 minus delta d by d whole square, right? So,
this is how we have seen that, so this is our proof actually. So, we have seen that, in
these cases that how we can use a, a differential pressure pick up and I can use a
sensor which has a differential output and how can I connect in the bridge all those
things are we have discussed.
Basically, in instrumentation we will find this capacitance sensors have basically three
uses. One is as a displacement sensor as you have seen - either permittivity variations
or the variations of the area or variation between the, separation between the plates -
number 1. Number 2 is the level gauge or level sensors. That means level will
increase that means the permittivity will change. So, obviously the permittivity will
change and the capacitance will change and third is the differential pressure pick up,
which is extensively used in all process industry for measurements of flow, okay DP
transmitter. This is, basically it is called DP transmitter.
We have seen that if I change the capacitance of the differential pick up, I will get a
change of voltage that can convert voltage to the current and transmit it, right? So,
37
these are the basic three uses of the capacitive sensors instrumentation, but we have
other applications, as I told you earlier also. We have discussed also, that in this case
that we are using the sensor as a, as a pressure pick up. That means some pressure
measurements also we can utilize this type of capacitance sensors. Capacitance
sensors are very, I mean very nice to use, because it is independent of temperatures
and all those things. That is a great advantage of this one.
It can be used in the corrosive environments, all those things are very much true, but
please note another two most important things of capacitor sensors that those cables
which is connected to this should be, we have taken, so the parasitic insensitive
measurements you have to make, because whenever we are measuring very small
capacitance value, whether you are using measuring in a bridge or LCR meter, it does
not matter, so the, the parasitic, I mean capacitance will influence your measurements
and it will pick up the signals like 50 Hertz signals and all those things. So, you
should be very careful about those parts. So, that non-shielding should be very much
good, so that if at least one part of the shielding should be, should be grounded, so
that the, you can make the measurement faithfully.
With this, I come to the end of this capacitive sensors.

Thank you!
Preview of next lecture:

basically study flowmeters. As you know, the basic three process parameters are
pressure, temperature and flow. However, if you look at the number of sensors
available or the varieties of the, varieties of the particular process parameters like
pressure, temperature is much less compared to the flow. The reason is that the flow
you will find, I mean not only the, the type of liquid flowing through the pipe or it
also basically depends on the what type of liquid is flowing. So, depending on that
there are various types of flowmeters available and used in the process industries.
There are, sometimes there are volumetric flow, there are sometimes mass flow, there
is sometimes …. displacement flow, there is open channel measurements, there is a
38
closed channel measurements. So, depending on this, all this circumstances, the
number of flowmeters are the highest among the, all the process parameters if you
look at. Now, in this lesson, basically we will discuss some, the contents of this
lesson, rather.
The flowmeters, we just give the name of the different flowmeters, then we will
discuss in details the differential pressure flowmeter, flow transmitter also we will
discuss in some brief and also we will show the, I mean separate, the separate video
on the flow transmitter.
Now, let us go to the instrumentation lab and look at one of the transmitter, one of the
differential pressure transmitter used extensively in the industry.
39
This is differential pressure transmitter. As you know that in the flow measurements,
actually the flow is calibrated in terms of pressure. So, in all flowmeters, I mean, I am
talking on the variable area like orifice meter and Venturimeter, so there is an
upstream pressure tap and downstream pressure tap. We have seen in the class that,
we have shown one nanometer sort of thing, by which I am measuring the pressure.
But unfortunately in the process, actual process, I need some instrument or some
equipment, which will give some signal that will be transmitted to the control room.
Now, for that reasons we need some electrical output. So, what they do? In this case
of this differential pressure transmitter, because there are two pressure tap, one is
upstream pressure tap, another is down downstream pressure tap, now you can see
here, this is basically a capacitive type, DP transmitter. Here, here we have a upstream
pressure tap and here you can see the downstream pressure tap. See, you can open it
and connect to the upstream pressure tap here. So, inside what we have? We have a
capacitor with movable plates, so and the inside of the two capacitors are plated with
some metal. So, you will get, once the capacitors, suppose this side is high and this
side is low, the capacitor movable plates will move on this side. So, I will get a push
pull sort of variations of the capacitance, which can be calibrated in terms of pressure
and ultimately that will be calibrated in terms of flow.
40
Now, this is, basically this capacitive sensors inside and this measurement circuits and
signal conditioning circuits are here on this side.
You can see here if I open, there is a connection for the signals. You see here that we
have positive signals, because the, in the process the output is 4 to 20 milliampere and
the supply voltage maximum is 45 volt DC supply, because in all the process, as you
know it is, you cannot give this …. voltage much higher. That is also not necessary
and it will give the, according to the variation first of all this, this circuit will first
measure the variation of that capacitance and that variations of the capacitance will be
converted in terms of current - 4 to 20 milliampere of current. So, ultimately that 4 to
20 milliampere of current will be calibrated in terms of flow. But please remember,
we have also differential flowmeters like, pitot tube is also differential flow, pressure
flowmeter, elbow meter is also differential pressure flowmeter, in next session which
will be discussed. So, this ends the lesson 12 of Industrial Instrumentation.
41
Prof. A. Barua
Lecture - 12
Flowmeter – I
Welcome to the lesson 12 of Industrial Instrumentation. In this lesson, we will basically

study flowmeters. As you know, the three basic process parameters are pressure,
temperature and flow. However, if you look at the number of sensors available or the
varieties of the, varieties of the particular process parameters like pressure, temperature is
much less compared to the flow. The reason is that the flow you will find, I mean not
only the, the type of liquid flowing through the pipe or it also basically depends on the
what type of liquid is flowing. So, depending on that there are various types of
flowmeters available and used in the process industries.
There are, sometimes there are volumetric flow, there are sometimes mass flow, there is
sometimes positive displacement flow, there is open channel measurements, there is a
closed channel measurements. So, depending on this, all this circumstances, the number
of flowmeters are the highest among the, all the process parameters if you look at.
1
Now, in this lesson, basically we will discuss some, the contents of this lesson, rather is
the flowmeters; we will just give the name of the different flowmeters. Then, we will
discuss in details the differential pressure flowmeter, flow transmitter also we will
discuss in some brief and also we will show the, I mean separate, the separate video on
the flow transmitter.
2
And now, commonly used flowmeters are like this. We have differential pressure
flowmeters, then we have variable area meters. Now, in the differential pressure
flowmeters you will find, we have basically most commonly used orifice meter and
Venturimeter, then we have elbow meter, then we have pitot tube. These are all basically
depends on the differential pressure. We have variable area meters; one of the variable
area meter is the rotameter, which is very commonly used in the process to get rough
estimate of the flow measurements. Then you have positive displacement meters and
positive displacement meter is a meter by which it is a, you can get the, not exactly the
flow velocity, but the quantity, quantity of liquid flowing in a particular situation.
One of the common examples of the positive displacement meters is the gas stations or
petrol pump. If you go to the petrol pump and take some amount of a petrol or diesel,
whatever it may be, you will find that actually I am not interested in the flow rate; I am
interested in the total quantity of the liquid. So, that type of situations we need positive
displacement meter. There may be turbine flowmeters which is, directly electrical output
you will get from the turbine flowmeters. We have electromagnetic flowmeters. It has, all
have some relative advantages and disadvantages. We will find the electromagnetic
flowmeters is very good, because there is no obstructions and all those things, we will
discuss later on.
We have vortex shedding flowmeters, flowmeters. Then, we have ultrasonic flowmeters;

well, using the ultrasonic sounds we are measuring the flow and it is basically non
invasive and it is like a ….. flowmeter. It has also, it does not create any pressure loss.
But, if you look at the other, like differential pressure meter, even though it is used to
extensively in the process, we will find it there is a permanent pressure loss in the, in the
pipe. We have cross correlation flowmeters also. These are all the, I mean glimpses of the
flowmeters available in common or commonly used in the anywhere, I mean like ….,
water supply or the chemical plant or steel plant, everywhere you will find these types of
flowmeters. We have also Laser Doppler flowmeters.
3
Now, if I look at the closed channel meters you will find that, I will just give you the
glimpse, I have not, I have not come exactly to the, I mean details of the, exactly the text
of this, our particular this lesson. Now closed channel meters, that means the closed pipe
I am interested to measure the flow. In that type of situations we have, obviously we will
mostly use the differential flowmeters and the differential pressure flowmeters, we have
two - Venturimeters and orifice meters. This we include, discuss in very much details,
right? We have variable area flowmeters, as I discussed. This is also closed channel
meters and it is a rotameter, … is one of the good examples of the closed channel
meters, I mean variable area flowmeters, which is used in the closed channel.
Then, in open channel meters in some situations, we need, we need to measure the open
channel, like one of the good example is a, is a irrigation purposes. If I want to, I want to
know the flow of the fleet, velocity of the fleet by which we can calculate the total
quantity of the liquid which is used for the irrigation, in that type of situations it is open
channels. So, in that open channel meters we have basically two types. One, it is the
weirs, we have a v notch weirs and rectangular weirs and we have also pitot tube. Now
pitot tube is, basically it is, can be used basically for the gases and gases, clean gases
4
rather and it can be used for both open channel and closed channel. That is the advantage
of the pitot tube.
But however, in this particular lesson we will discuss basically this differential pressure
flowmeters. Others we will try; variable flowmeters, weirs, pitot tube, we will discuss in
the subsequent lessons. The total quantity of liquid is obviously what is that? That is the
time integral of flow rate. If I measure the time and multiply by the flow rate, if the flow
rate is uniform or average, so we will multiply by that, I will get the total quantity of
liquid flowing in a, through a particular cross section.
Now, at the end of this lesson the viewer will know the principle of operation of a
differential pressure flowmeter, orifice meter, Venturimeter, discharge co-efficient of
Venturimeter or orifice meter, then pressure tappings. There are various pressure tapping,
available that will be discussed, pipe bending near the differential pressure flowmeter,
because whenever you are using any differential pressure flowmeters, there is restrictions
of the pipe bendings, both upstream and downstream that you must look at, right? Then
pressure recovery, because you see, the differential pressure flowmeters, there is loss of,
there is permanent pressure loss and permanent pressure loss means pumping, extra
5
pumping cost, so that you have to take care and this is very crucial in the case of any
water supply and all those things, because pumping needs a lot of money involves, so that
the, whatever you are supplying, the water or other things, the cost will be increased.
Now, what is the need of flowmeters in a process? The, first of all you now, we know the
flowmeters is necessary for the, I mean for suppose in the water supplies flowmeters is
necessary, because I know how much quantity of liquid or water I am supplying to the
particular town or city. That is one thing, but what is the need actually in the process?
Measurement of the volumetric flow rate and mass flow rate are necessary for the
purpose of determining the proportions of materials introduce to a process and the
amount of materials produced in the process, right?
6
Measurement of volume flow rate if you look at, now differential pressure flowmeters
comes under these. They are most widely used flowmeters for liquids and gases. A
restriction, a restriction or constriction is placed in the pipe and the differential pressure
developed across the restriction is measured; as simple as that. It is very simple, I mean
principle is just applying Bernoulli equations you can find the flow rate and the
differential pressure output is calibrated in terms of volume flow rate. So, the differential
pressure is calibrated in terms of volume flow rate. Then, we will convert into the current
and all those things that will be taken care later on.
7
Now you see, this is a typical differential pressure flowmeter. The schematic of the
differential pressure flowmeters we can see here. You see, if I look at, I do not know
which, fine, these I can take. You see here, this is a pipe. Liquid is flowing through these
directions. Liquid is flowing through these directions, it is coming out this direction, clear
and you see, there is a restriction; this is our restriction in the form of orifice. It is
basically a circular plate. Please note, there is hole in the, a consetting hole in the middle,
a circular plate. So, liquid is flowing through this plate. So, liquid is coming and it is
flowing through like this one, clear?
8
So, it will not be exactly like this one. So, let me go back again. So, so it will look like
this one. So, the liquid will flow and there is minimum pressure area after that. So, it will
go and after that the liquid will flow like this way, clear? Now, the area of cross section
in the upstreams we are taking A 1, area of cross sections of the restrictions, of the
constrictions is A 2, P 1 is we are measuring the upstream pressure tap, we are calling it
upstream pressure tap, P 2 is the downstream pressure tap. P 2 is the downstream
pressure tap, P 1 is the upstream pressure tap and Z 1 and Z 2 is a, is a, is a height of this
central point of this area of cross section of the pipe as well as of the other point of
measurements of upstream pressures as well as the orifice, orifice plate, right? So, we are
making Z 1 and Z 2. Let us go back.
9
Now, if I apply the Bernoulli equations across these two, I mea sections will upstream
and downstream easily I can write P 1 by row 1 plus V 1 square by 2 plus Z 1 into g
equal to P 2 by row 2 plus V 2 square by 2 plus Z 2 g. See, unit you can see all the
dimensions are equal. It is, all are, it is a meter per second square, a units of the
individual. So, it is, we have taken it per unit meter. So, we can write like, Bernoulli
equation like this, right? Total height, I mean total height of the liquid at any point is
same that is the basic principle. P is the pressure in Newton per meters square, v is
average velocity in meter per second. v 1 is a upstream velocity and v 2 is the velocity at
the restrictions or the minimum area and row is the fluid density, which is the kg on
meter cube, everything is in SI, no problem; A is the cross sectional area, it is in meter
square and Z is the elevation above datum level. So, you can take any datum levels,
obviously this is not very important if the pipe is horizontal and that is in meter.
10
So, the following assumptions have been made in calculating the volume. So, you have to
make some simple assumption to calculate the volume flow rate. These are: the flow is
frictionless. It means there is no loss of energy due to friction either in the fluid itself or
between the fluid and the pipe walls. There is no heat losses or gains due to heat transfer
between the fluids and surroundings.
11
There is conservation of total energy – pressure, kinetic and potential. We have written
the Bernoulli’s equations from that only, right? So, it is nothing new. So, there is
conservation of total energy pressure plus kinetic plus potential at any point of the liquid.
The fluid is incompressible, incompressible. So, it is row 1 is equal to row 2 is equal to
row, right, so that the density is same everywhere and the pipe is horizontal. That means
Z 1 is equal Z 2. So, it means that the equation 1 reduces to v 2 square minus v 1 square
upon 2 is equal to P 1 minus P 2 by row.
Now, conservation of the volume flow rate if it is there, so we can write Q 1 is equal to Q
2 equal to Q. That means Q 1 is the upstream flow volumetric flow rate, Q 2 is the
downstream, downstream volumetric flow rate. Also we know Q 1 is equal to A 1 into V
1. A 1 is area of cross sections of the pipe and V 1 is the pipe, the liquid flow diameter,
flow velocity, Q 2 is the downstream flow velocity. Obviously, Q 1 will be equal to Q 2.
This will be equal to the area of cross section of the orifice multiplied by the velocity of
the liquid at the orifice.
Obviously, if the P 1 is greater than P 2, that will come in the next slide. Let us go to the
next slide.
12
So, since A 2 is less than A 1 it follows that V 2 will be greater than V 1 and P 2 will be
less than P 1, right? Since A 2 less than A 1 it follows that the V 2 will be, that is obvious
from the last equation V 2 will be greater than V 1 and that means the flow velocity at the
restriction or at the, at the, at the minimum area of cross section is high compared to the
actual flow of velocity, right? We are not interested in that. We are exactly interested in
the volumetric flow rate that is same both in upstream and downstream.
Therefore, the theoretical value of the volume flow rate in a differential pressure
flowmeter, both in the case of Venturi and orifice is that Q theoretical we are writing is A
2 upon under the square route 1 minus A 2 upon A 1 square under the square root 2 P 1
minus P 2 by row, this is equation number 2. The theoretical value of volume flow rate
always differs from the actual flow rate due to two main reasons.
There is some restriction. Why I am calling it theoretical? We will find that we have to
multiply by the discharge coefficients with the actual flow rate, actual volumetric flow
rate. The frictionless flow is never occurred in a pipe. It is true for turbulent flows in
smooth pipe where friction losses are small. The laminar and turbulence flows are
13
characterized by the Reynold number and a Reynold number is given as R e equal to
VDP by eta.
This will give you the Reynold number, where D is the pipe diameter, eta is the viscosity
of the fluid flowing in the pipe, V is the velocity of the fluid in the pipe and P is the
differential pressure across the section of the pipe, right?
14
Number 2 is A 1 and A 2 are the cross sectional areas of the pipe and the restriction,
respectively. The cross sectional area of the pipe is pie D square by 4, obviously and the
cross sectional area of the meter, orifice meter or Venturimeter, it is pie D square by 4,
where D and capital D and small d are the respective diameters. That means pipe
diameters, internal pipe diameters and this is diameters of orifice or the throat diameters
of the Venturi. If the fluid fills the pipe then A 1 equal to pie D square by 4.
However the area of the, minimum cross sectional area will be not exactly at the, at the
restrictions. So, it will be .99 pie d square by 4 for a Venturimeter, because Venturi is a,
is not a sudden, I mean restrictions. However, the orifice plate is a sudden restriction
which causes the fluid cross sectional area to have minimum value of .6 pie d square by 4
at the vena contracta. That means it is little far off from the, little off from the orifice
plate, where you will get the minimum area of the cross section of the fluid, right?
15
Therefore, the theoretical expression of volume flow rate is corrected as C. We are

introducing a new factor C under the square root 1 minus beta to be the power 4
multiplied by A 2 under the square root 2 P 1 minus P 2 by row, where C is discharge
coefficient and beta is the flowmeter pipe diameter ratio, small d by capital D and A 2 is
the flowmeter cross sectional area. This is equal to pie d square by 4.
16
Values of the discharge coefficient depend on type of flowmeter that is orifice or Venturi,
Reynold number R e, diameter ratio beta. So, therefore for a given flow rate, flowmeter, I
can write C equal to, is a function, C is a function of R e, Reynold number and beta.
Values of C are found experimentally for several types of flowmeters over a wide range
of fluid velocities, right, because flowmeters velocity should be, C should be known to
you.
For a given fluid and known volumetric rate of fluid flow, C can be found from the
equation 3. If the fluid, volumetric rate of the fluid is known, so obviously I can and if the
density of the fluid is known, so obviously I can found the, find the value of C. Now,
general features of differential pressure flowmeter, what are the general features let us
look at. These are most important that is the reason it is used widely in industry for last
several decades.
Number 1 it has no moving part, so you look at there is no moving parts. So, it is robust,
reliable and easy to maintain and widely established. It is established over the years. It is
documented, data are available. So, all these things, all the supports we are getting from
the past users, so that is reason it is used widely in industry. Number 2, there is always a
17
permanent pressure loss and the extra pumping energy. This is not a good feature; this is
a bad feature, obviously. So, there is always a permanent pressure loss and the extra
pumping energy is necessary to overcome it. So, whenever there is a permanent,
permanent pressure loss, I have to meet some extra pumping energy. So, obviously this
pumping energy needed in the case of orifice meter is more than that of the, that of the
Venturi, because in Venturi the permanent pressure loss is much less.
Now, let us look at, you see it very carefully.
So, you see here that if I look at the, from this one, you see and this is our pipe. This is
the distance from the orifice we are taking, orifice or Venturi throat. Here you see at the
differential pressure measured, here there is a sudden fall of the differential pressure near
the orifice. After that, here this pressure loss is, it is pressure, so this is a pressure
recovery. There is a sudden pressure loss. This is good for measurement. So, why ….. the
pressure loss if it is temporary that is good for measurement, because we are getting, we
can, our sensitivity of our instrument will be higher. Because, if you see here, the
pressure loss should be always, would be higher and higher. Here, pressure ….
18
differential pressure should be higher and higher, because in that case what is the, what is
our input?
Input is the flow and output is the differential pressure. So, differential pressure by flow
if it is higher and higher, obviously that is better for us, because static sensitivity will be
increased in this case, right? But you see, this should not be permanent. That is the
problem with the orifice meter. So, you see, there is a permanent pressure loss that will
never recover. So, I need extra pumping energy to overcome this pressure loss, right? So,
this is most important, right? You see, this is the permanent pressure loss, say this is the
temporary pressure loss we are measuring at this point, preferably. Then, there is a
permanent pressure loss, right?
Now, both Venturi or orifice meter is non linear that is most important. If you see,
volumetric flow rate is proportional to the square root of the pressure differential, right?
So, this limits the useful range of a meter in between 25 and 100% full scale output
reading. At lower flows, the differential pressure measurements is below 6% of full scale
output reading and is not accurate enough for the measurement.
19
Now, you see there, so useful range. Here you see it is a nonlinear curve. So, because you
can see that the flow velocity or the quantitative flow, volumetric flow rate is
proportional to the square root of the differential pressure, so obviously it is nonlinear.
You can see here that if the flow velocity is less than 25% of the full scale output rating,
if the full scale output rating is there, 25% of the full scale output rating, so the, whatever
the measurements we are doing that is much erroneous, right? So, it is not very good for
the low flow measurements. It is good for the higher flow measurements. That means the,
the error will be less in this case, right?
20
These error you can see here, excuse me, say error you see here, error here is quite low at
the low flow velocity, quite low at the, at the high flow velocity and is very high when
the flow velocity is, volumetric flow rate is slow.
Number 4, it can be used for turbulent flows, because turbulent flow is characterized by
when R e is much, is greater than 10 to the power 4, 10,000. A typical flowmeter system
21
consists of the differential pressure sensor, differential pressure transmitter, data
acquisition system and a PC. The transmitter gives a current output signal, current output
signal of 4 to 20 million ampere and the DAS consists of an amplifier, I to V converter
and ADC, right?
This is our system; you see, differential pressure sensing element and transmitter, data
acquisition system and PC. Please note that the, this differential pressure transmitter
should be an assembly or is to be installed in the, on the pipe itself just outside the pipe
and it should give you 4 to 20 milliampere of current which will go to the computers
where the data acquisition card is there, so, through which will convert to the voltage
domain and will be used for further processing after conversion to the, I mean digital
domain. ……. milliampere of current is obviously is an analog current, right?
Now, let us go to the instrumentation lab and look at one of the transmitters, one of the
differential pressure transmitters used extensively in the industry.
22
This is differential pressure transmitter. As you know that in the flow measurements,
actually the flow is calibrated in terms of pressure. So, in all flowmeters, I mean, I am
talking on the variable area like orifice meter and Venturimeter, so there is an upstream
pressure tap and downstream pressure tap. We have seen in the class that, we have shown
one nanometer sort of thing, by which I am measuring the pressure. But unfortunately in
the process, actual process, I need some instrument or some equipment, which will give
some signal that will be transmitted to the control room.
Now, for that reasons we need some electrical output. So, what they do? In this case of
this differential pressure transmitter, because there are two pressure tap, one is upstream
pressure tap, another is downstream pressure tap. Now, you can see here, this is basically
a capacitive type DP transmitter. Here we have an upstream pressure tap and here you
can see the downstream pressure tap. So, you can open it and connect to the upstream
pressure tap here. So, inside what we have? We have a capacitor with movable plates and
the inside of the two capacitors are plated with some metal. So, you will get, once the
capacitors, suppose this side is high and this side is low, the capacitor movable plates will
move on this side. So, I will get a push pull sort of variations of the capacitance, which
can be calibrated in terms of pressure and ultimately that will be calibrated in terms of
23
flow. Now, this is basically this capacitive sensor inside and this measurement circuits
and signal conditioning circuits are here on this side.
You can see here if I open, there is a connection for the signals. You see here that we
have positive signals, because the, in the process the output is 4 to 20 milliampere and the
supply voltage maximum is 45 volt DC supply, because in all the process, as you know it
is, you cannot give this live voltage much higher. That is also not necessary and it will
give the, according to the variation first of all this, this circuit will first measure the
variation of that capacitance and that variations of the capacitance will be converted in
terms of current - 4 to 20 milliampere of current. So, ultimately that 4 to 20 milliampere
of current will be calibrated in terms of flow and the output signal we can take from this
side.
24
We can open this side and see here the circuitry and we can connect here and there is a
….. things here and this side you can make all the connections sort of things for taking
the, both the, giving a signal and taking the 4 to 20 milliampere of current. This circuitry
which we, this, this actually the …. which you can take it out for making repairment and
all that type of job.
So, now we will discuss a particular, we already, we discussed the orifice and Venturi,
but there is some difference between the orifice and Venturimeters and also we discussed
Dall tube and flow nozzle and all those things. So, let me first discus orifice meter which
is, I think if we look at the flowmeters, it is most widely used flowmeters in the process
industry. I should say 80% of the flowmeters are the orifice. If you look at the, if you
compare the number of meters like so many meters we have discussed at the beginning,
sophisticated meters, all these things, but orifice because it is established over the years,
this is one of the very widely used, cheap, easily replaceable meters and quite …. of that
is most important thing. Maintenance is also very easy, just replace it. It does not need
calibration. Once we have made several orifice of the, from the same lathe machines, so it
won’t take time. So, just you go and …. if the diameters ratio are correct, if small d by
capital D is same as before, if it is square edge or if it is, so there is no question of
25
making any further calibration. That is great advantage of this. So, during routine
maintenance just replace the orifice. That will simplify. If you look at the other meters it
needs lot of, it is expensive also. So, among the differential pressure flow meters orifice
is the cheapest. Let us go back to the orifice.
Now, it is a thin plate square edge orifice, it is the most widely used differential pressure
flowmeters. Thin plate number all square edge orifice is the most widely used differential
pressure flowmeter in process industry, mainly because of its simplicity, low cost and
moreover it is well established over the years and the data are, data are available for its
behaviour that is most important thing. Just you cannot have a new, I mean something,
somewhere in, in some laboratory in the world I develop some flowmeters, so
immediately that cannot be used. It is used over the years and people …
Ultimately our goal is to make our product good, is not it? I told you several times before
also, quality of steel, quality of fertilizer, quality of petrol, diesel or quality of a life, you
…… anywhere, ….. quality of our drugs which you are producing in the bioreactors, so it
does not matter I mean what type of flowmeters you are using. If the product is good,
26
product is pure, so there is no question. So, orifice meter, it is working for several years
for all these reasons and we are using it extensively in the industry.
Now, typically three types of orifice plates are available. It is basically a plate. If you
look at it looks like this, you see. So, it is a plate here and there is a hole, right? If you
look at, so it will be like this. Square edge orifice, so it should be, if I look at, so the
liquid is flowing …. through that like this, clear, right?
27
Let us go back again. So, these are the three different types of orifice, if you look at. So,
there is a concentric orifice, eccentric orifice and segmental orifice. All has different
application that is the reason we are plotted, we are given like this.
So, the concentric orifice is the most widely used plate. Eccentric and segmental orifices
are employed to measure the flow of the fluids containing solids. In both cases, the
28
bottom of the hole is located in a way, when the bottom of the hole is at the same level
inside the bottom of pipe installation. These two orifice plates need separate calibration,
because the discharge coefficients differ from that of the concentric orifice, clear? I need
separate calibration for this one. Suppose I have suspended particles, all those thing, for
that reasons if you, suppose you have some slurries, if you use a concentric orifice, so all
these things will be deposited at the bottom of the pipe, but if you use eccentric orifice
that will be flushed away. So, that is the reason some application of the liquid flow
measurements we use concentric orifice, some I mean we will use eccentric and in some
will use segmental orifice.
The concentric orifice plate is installed in the pipe with its hole concentric to the pipe
itself. It is a flat metal circular plate made of steel, stainless steel, phosphor bronze, like
that. Its thickness is only sufficient to withstand the buckling forces caused by the
differential pressure exists across the plate. So, so far that it can I mean withstand that
differential pressures you can use a thin plate, but it should not buckle, then your entire
calibrations will go wrong. For that reasons if you have …… differential pressure and all
those things, we have different types of flow measurements, although that also basically
depends on differential pressure flow, I mean differential pressure measurement.
29
Circular hole is made with 90 degree square and sharp edge upstream. It is a 90 degree
that I have shown you and sharp edge upstream. The change of sharp edge will modify
the discharge coefficient of the orifice meter and it is advisable to replace the orifice
during the routine maintenance of the plant for better accuracy and of the measurement.
30
Now, orifice pressure taps, there are various pressure taps available for the orifice. There
are, here in this particular lesson, we will discuss three pressure taps. Flange taps, you see
there a flange tap, this orifice plate with the flange taps. Even though I have drawn like
this one, you can see here, but it is actually as I told you earlier also, so it is I mean drawn
like this one, so this is actually, so square edge it is like this one, right? So, but actually
we have plotted like this, anyway.
Now, it is constructed so that the taps for measuring the differential pressure are integral
part of the orifice plate assembly. It is true, the one thing it is in the, in the orifice itself
you have the pressure tap installed; its entire assembly is available, right? The pressure
taps are usually located 2.5 centimeter either side of the orifice plate. The advantage of
the flange tap is that the entire orifice assembly is easily replaceable and the pressure taps
are accurately, are accurately located, right?
31
Now, we have an orifice plate with D and D by 2 taps. Here you can see that it is,
upstream is D and downstream D by 2 tap. All these have some advantage and
disadvantages.
Then we have orifice with Vena contracta taps. I am sorry, this will be vena contracta,
contracta, so this will be contracta, vena contracta; sorry this is will be a, vena contracta
32
taps. So, orifice because you know the area of the minimum cross sections does not
appear here. It is somewhere away from this one and we always try to measure there at
the area of the vena contracta, because the differential pressure will be, will be the, if the
area of cross section is low, the differential pressure at that point will be higher, highest.
So the, there is ease of measurement. Our static sensitivity of the sensor will be increased.
So, that is the reason what the people do. Because, as I told you, the area of the minimum
cross section is not here, it is little far away. So, this is the tap where I am measuring that
the pressure, downstream pressure at the area of the, at the place of the vena contracta
which is the area of the minimum cross section of the fluid flow.
It is arranged in a way so that the volume, so that the downstream pressure tap is located
at a variable distance from the orifice, right, depending on the pipe and the orifice size.
The upstream tap is at one pipe diameter and the downstream tap is at the vena contracta.
In a vena contracta tap, the pressure differential is a maximum for a given flow rate; this
is important. At the vena contracta tap pressure differential is maximum for a given flow
rate. So, that pressure differential, as I told you repeatedly, what is advantage of that
having a large pressure differential, right?
33
Now, restrictions of the pipe fittings adjacent to orifice meter - this is another most
important thing practicing engineers must follow, so as that is the restrictions of pipe
fittings adjacent to orifice meter. There cannot be pipe fittings, pipe bendings, tees, bags,
etc., very near to the orifice, both upstream and downstream. What are those? Let us look
at details. The discharge coefficients are experimentally determined on straight pipes,
right?
Now, flow disturbances in the pipe line adjacent to the orifice alter the value of the
discharge coefficient. Therefore, elbow, pipe bend, tee, valve are not allowed near the
orifice meter. How much near that we will first know or that is what do you mean by
near? Nearing that is, I mean this some you must quantify, I mean you cannot say just
qualitatively very near or something or very far.
34
There should be no fittings closer to the five pipe diameter from the orifice on the
downstream. Same - no bending, no elbow, no valve, no tee, there should not be any
fittings closer than twenty pipe diameter upstream. Upstream is more restricted, that we
can see. If the minimum distance is not feasible, in some situation you will find it is very
difficult to achieve that, because in the process you will find that the, it is not possible, I
mean to have these pipe bendings everywhere, right, I mean you cannot avoid, because
there will be some pipe bending. So, in that type of situation when the pipe bending is
unavoidable, you cannot avoid the pipe bendings, I have to use some flow straightners
both upstream and downstream.
What is those, what are those flow straightners, flow straightners that is clear from this.
You see, if the minimum distance is not feasible especially in the upstream, flow
straightners can be installed and the flow straightners are bundle of smaller tubes welded
inside the pipe. It has the bundle of smaller tubes welded inside the pipe. So, it will make
the flow straight, so that the, the problem which occurred that means due to the pipe
bending, elbow, tee can be, I mean can be minimized by using the flow straightners.
Now, next type of flowmeters already we have discussed orifice meter.
35
Next type of flowmeters are the flow nozzle, dall tube, Venturi all are same and orifice
meter, Venturi meter, flow nozzle, dall tube, Venturi, these are basically same, but all are
depends on differential pressure measurements. We will discuss one by one - flow
nozzle, dall tube, then we will discuss in more detail the Venturimeters, because that is
widely used for the large flow measurements, large quantity, I mean flow or the
volumetric flow rate. This is a flow nozzle, you can see. This is a, is a shape like this one.
It is a, this is called flow nozzle. We have an upstream pressure tap here, a downstream
pressure just at the end of the nozzle outside the pipe. Just here we have upstream,
downstream pressure tap, this is upstream pressure tap.
36
Here you see, this is a, this is our Venturi tube. Venturimeter is something like, this is
upstream pressure tap. These are piezometer rings. We want to average the pressures.
This is the throat tap. Instead of downstream we are calling it throat tap. This is the area
of the minimum cross section, because there is difference of area will be A 1. This is A 1,
area of cross section. This is the capital D, this is the small d. So, that ratio will give you
the beta. This is the Venturimeter and these piezometer rings is to average the upstream
pressure and this is to average the downstream pressure.
37
Now, this is a dall tube. So, we have a pressure, you have pressure tap here P 1 and
pressure tap here P 2. So, these are all, basically there are some relative advantages and
disadvantages in all these types of things. Flow nozzle and dall tube, let us first discuss
the flow nozzle and dall tube, compare, then we will discuss the Venturi in details.
38
The flow nozzle, Venturi tube and dall flow tube have the same principle as the orifice.
Flow nozzle, Venturi tube, dall flow tube have the same principle as the orifice; no
difference, same. There is a differential pressure and we will measure the differential
pressures. We will calibrate the differential pressure in terms of the volumetric flow rate,
so no difference. So, that is the same, exactly same like orifice meter and it is non linear.
Any differential pressure, I mean flowmeters are non linear. That means the volumetric
flow rate is a, is the square root, proportional to the square root of the differential
pressure or if you look at the differential pressure, it is proportional to the square of the
volume flow rate.
Dall tube is a modified Venturi tube and it has a low permanent pressure loss; not low
like, I mean Venturi tube, but it has a low permanent pressure loss, right? The flow
nozzle are more expensive than the orifice meter, but cheaper than the Venturi meter.
Venturi is the most expensive, as you know. So, the flow nozzle are more expensive than
the orifice, but cheaper than the Venturi.
It is also a variation of the Venturi in which the exit section is omitted. We have, do not
have any exit section if you look at very carefully, so that it is similar to the, similar to an
39
orifice with well rounded upstream edge. The upstream tap is about one pipe diameter
from the entrance to the nozzle and the downstream tap is made on the pipe opposite to
the straight portion of the nozzle - that already we have discussed, right? We have shown
that thing.
Flow nozzles are used for, because you know that in some situations or there is high
pressures and high velocities, sometimes the orifice is not very stable. There is the
question of buckling and all these things; that type of situation we have to use the flow
nozzle and we cannot spend that money everywhere, I mean to install a Venturi meter
instead of the orifice meter. So, in that type of situations you can use flow nozzle,
because it is cheaper than the Venturi meter.
Flow nozzles are used for high velocity stream flows and it is dimensionally more stable
at high temperature, velocity than an orifice. This is the most salient feature of the flow
nozzle. That is the reason for high velocity, high temperatures we can use this flow
nozzle instead of using the orifice meter and the pressure tapping is also not that difficult.
So, you can have upstream pressure tap at the one pipe diameter, downstream pressure
tap just at the end of the straight portions, just at the end, opposite to the end of the
40
straight portion of the pipe, straight portion of the nozzle that will be on the pipe, right?
The permanent pressure loss in the flow nozzle is same as orifice. So, there is a
permanent pressure loss. It is quite high in the case of orifice meters. So, it is same also in
the flow meters. It is the least or minimum in the case of Venturi meter, right? So,
whenever there is a savings of the extra pumping cost you have to consider, you must go
for Venturi meter; there is no other alternative.
Now if, let us, let me discuss Venturi meters in detail. Venturi meter is an expensive
instrument, but offers very good accuracy. It is I should say one of the most accurate
flowmeters developed by the mankind. So, it is plus minus 1% accuracy, so you can see
that it is quite accurate. It has the lowest permanent pressure loss that I told several times.
So, whenever there is question of permanent pressure loss, so you can go for the, this
type of sensor.
Now, construction of the Venturi if you look, look at the construction how the Venturi
actually look like, it is made of cast iron or steel. Sometimes if you have larger, then I
need a, it can be made of concrete also; large …… it can be made of concrete, made of
cast iron or steel. Sometimes the throat portions of the Venturi, because if you look at the
41
Venturi it looks like this, isn’t it? The Venturi looks like this. This we are calling it throat
of the Venturi. We have a tap here; we have a tap here you know, so this is called throat.
So, in some application, throats are made of separate metal, I mean it is usually made of
bronze, because it is easily replaceable, all those things. The main systems, this, both this
and this will be made up cast iron or steel, right and this all about …
The large Venturi tube is usually made of concrete, because if it is very large, so it is very
difficult to make, made of cast iron or steel. In that case we go for the concrete.
Sometimes the throat is made of bronze, as I told you earlier, for easy replaceability.
Upstream section has an angel of 20 degree, included angle. That means if you look at, so
it looks like this.
42
If I take a white page, so it will look like this that, I have to take different colours, so I am
talking of this angle. That means if you look at this angle, this angle is 20 degree in the
case of Venturi meters and this angle should be 7 degrees. In fact, you will find that
basically there are two types of Venturi. One is for the larger downstream portion,
another is the smaller downstream portion. You will find that if you, if you make this
angle even lower, your pressure recovery will be also higher. So this angle, but it takes
large space and all those things; that is the restriction of the Venturi. But, if you make this
one large, you will find your pressure recovery also will be large. In fact, pressure
recovery in the shorter Venturi is less.
That means the permanent pressure loss will be more in the case of short Venturi than the
longer, standard Venturi. So, if this is large, suppose that I mean if you have this, this
type of Venturi, then we have that, you will find that the total length might be increased,
right; total length has increased, but the thing is that pressure recovery will be better and
the permanent pressure drop also will be less. So, this is very important.
43
Upstream section has an angle of 20 degree and downstream section has an angle of 7
degree. These are all included angle, right? The pressure taps are, this should be, I am
sorry this will be pressure taps, this will be pressure taps; pressure taps are made of
piezometer rings, so as to average the measurement around the periphery.
44
Diameter ratio, small d by capital D for Venturi typically lies between .252 and .50. It has
almost no maintenance requirements and its working life is very, very long. So, almost
absolutely no maintenance, which is necessary in the case of, in the case of orifice meter,
special orifice meter. Whenever there is a routine maintenance you have to check that the
square edge section is steel square edge, whether the diameter is exactly it remains as
small d, which we have used for calibration purpose. If it is not, then you have to replace
it. Usually, I mean during routine maintenance you must exchange this, replace this old
orifice with a newer one, which is not necessary in the case of Venturi, because this is
permanent, this throat is made out of steel, there is no chance of increasing the throat
diameter which is there, because if you, if you after long use that means if I, you look at,
that if the orifice is, I do not know; so, you see that we have a, if I take a, if you take like
this one, so if this edge has got, I mean instead of like this one it is getting like this one,
right and this is getting like this one, right so, the entire calibrations will go wrong. So,
that, this, this type of situation does not arise, arise in the case of, arise in the case of
Venturi meter. That is the great advantage of the Venturi meters.
It is widely used in high flow situations such as municipal water system where large
savings of pumping costs are possible due to low permanent pressure loss across it.
45
Usually that type of situations, I mean say, if the, if you look at the range of this
flowmeters, orifice meter has a larger range than the, Venturi meters are larger than the
orifice meters. So, this is the two great advantage of the, except the cost, the main
advantages of the, this Venturi meter is that it is reliable, its range is very high, very, very
high compared to orifice, any other permanent pressure drop is also very low. So,
keeping all this in mind and whenever you are, I want to use, because you have to,
whenever you want to this type of meters where the accuracy is very important, so I have
to use this Venturi meter, there is no other alternative, right?
Now, the smooth internal shape of the Venturi tube means it is unaffected by the solid
particles or gaseous bubbles in flowing fluid and in fact, it can measure the flow of liquid
containing slurries. You see, this is very important. Even though in the case of, you will
find that in future also you will, some flowmeters, electromagnetic flowmeters and
electromagnetic flowmeters are there, so where you can use the, even if the liquid has
some suspended particles, because since there also, there is no abstraction type of things.
In the Venturi also there is no abstraction unlike your orifice meter, though both depends
on the, both the, they depend on the differential pressure meter, you will find, differential
pressure principle you will find that the, if the liquid contains the gaseous bubbles, we
46
have tried, you see that we have seen already, discussed in the orifice meters we have
eccentric orifice, we have concentric orifice, we have segmental orifice, but this does not
suffice. In many situations, you will find it does not work. If you have, as … as the liquid
is clean it is very difficult to measure with the orifice meter, right? That type of situation
does not arise in the case of … Even though I am saying that the, you, you can do it with
electromagnetic flowmeter, but electromagnetic flowmeter has problems. That means the
liquid should have some conductivity levels. So, without that suppose in the case of
hydrocarbon industries you cannot use electromagnetic flowmeters, because the
conductivity is very poor there, right?
So, that type of situation does not arise in the, in the case of Venturi meter. It does not
matter whether it is conducting liquid or not. Not only Venturi, orifice also, whether it is
conducting liquid or non-conducting liquid does not matter, right? So far there is a
differential pressure, I can measure it, right and since we are not measuring with
manometers, we are using u tube manometer, we are using some differential pressure
transmitter, so even though the liquid has some suspended particles, some slurries, it does
not affect our pressure measurement, right, because you are, basically we are using a
separate type of, I mean capsule type differential pressure transmitter, so which will sense
some signals and give 4 to 20 milliampere of output.
47
Now, its range is very high. You can see its range is extremely high. It is possible to
measure the water flow rate as high as 1.5 into 10 to the power 6 meter cube per hour. No
other flowmeter will give you this huge range. The dimensions will be quite big that I
agree, but the thing is this type of high fluid flow rate or the volumetric flow rate, we will
never achieve in any other flowmeter. This is very important, right and this is also, please
note that this is also needed in some type of situation, because in that type situations you
will find that I need flowmeters where I need differential pressure output as well as I need
flowmeters, where I need large quantity of the volume flow rate. In that type of situations
I have to use the Venturi flowmeters. So, this is very important measurements, right?
Now, another thing, please note in the case of Venturi flow, I mean orifice meters, once
you install, the advantage of the Venturi meter you cannot even find that if you go to any
process industry you may find one or two, I mean Venturi meters, but you will find
hundreds of orifice meters, because of its easy replaceability. It is very easy to replace it.
Just take out the plate and replace the plate of the same diameters and same thickness, so
you do not have to recalibrate your sensor. That is if it is square edge orifice, if the small
d is same, so there is no question of recalibration and in the case of Venturi meters there
is no question of that type of calibration or anything because its life, I say, it is just, if
48
there is no weir, or the surface, the internal surface or if there is no chances of increasing
the internal diameters of the pipes, so there is no question of changing of beta. So, these,
all these advantages, I mean make these two particular flowmeters, even though we are,
you will find that there are many different types of flowmeters - electromagnetic
flowmeters, we have …… flowmeters, ultrasonic flowmeters, …… flowmeters, but this
does not used much in the any process industry.
What we are using actually you will find that this type of signal in, in large I mean this
type of sensors that means Venturi and flow in large number. Only disadvantage of this
type of meter if you look at that is orifice and Venturi, everywhere you need a
transmitter. I need a transmitter, both in the orifice and Venturi I mean a system, not only
a system, because here the output is not electrical, unlike your turbine flowmeter, unlike
your electromagnetic flowmeter, where the output is direct electrical. Basically it is
pneumatic output, which is coming in the case of, in the case of, or hydraulic outputs
which is converted by capacitance change and we are getting 4 to 20 milliampere of
current from the transmitter which is transmitted, but if the electrical output is there either
in voltage or current, directly that can be transmitted. If it is the case of voltage, we will
convert to the 4 to 20 milliampere of current and transmit.
These are the some advantages of the, of the non differential type of flowmeters, but
…… differential flowmeters, see we are using it for a quite long, many years in the
industry. So, it does not matter that the, what is the, I mean how much it is cumbersome.
It looks very cumbersome, we have seen already in the, when we went to the lab and see
this differential pressure transmitter, so every orifice meter and Venturi meter must have
that type of things just above the pipe. So, that should be there.
So, if you, there are suppose 500 orifice meters, so we must have 500 transmitter also. It
does not matter you know exactly when you, you are getting the correct measurements. It
is easy, ease of maintenance is I mean leaving all these things if you have this type of
advantages that I mean pushes us, I mean to a corner, where we have no other alternative
than using a orifice meter and all the data are available for the years, for the all different
49
process industries, all different manufacturing industries. So, this helps us to make the
use of this particular type of flowmeters.
We have discussed only this orifice and Venturi. There are many other flowmeters, like I
told you that this is the only two differential pressure flowmeters we have discussed in
this particular lesson. But, please remember we have also differential flowmeters like,
pitot tube is also differential pressure flowmeters, Elbow meter is also differential
pressure flowmeter. In the next session this will be discussed. So, this ends the lesson 12
of Industrial Instrumentation.
50
Prof. A. Barua
Lecture - 13
Flowmeter - II
Welcome to the lesson 13 of Industrial Instrumentation. We will continue with the

flowmeters. So, in the lesson 12 also we have seen the flowmeters, because as I told you
earlier, that the flow, the, the variety of the flow, I mean the variety of the instruments for
measuring the process parameters like flow is, are the huge in number. So, what we have
to do? We have to cover lot of sensors; unlike the temperatures and pressure level, very
few number of sensors are available. But, due to the varieties and the environment
conditions, the number of flowmeters are quite large in number. So, this lesson 13 also is
for the flowmeter. So, we have given the name Flowmeter II.
You see here, the contents of this lesson is Pitot tube, which is extensively used for, not
only for the flow measurement, also for the velocity measurements. If I can install and it
is basically mostly used for the, not for the liquid, but for the gas. Even for high speed
vehicles also I can use this for measurement of the speed of the vehicles by using the
1
Pitot tube. One of the common, I mean applications of the Pitot tube is the aircraft, the
speed measurements, where in high altitude, so the, there is a differential pressure and
another important thing this Pitot tube as well as, as well as the elbow meter which is
next, you will find basically depends on the differential pressure measurements.
We have seen that differential pressure, differential pressure measurement technique is

utilized in the case of orifice meter and Venturi meter. So, in Pitot tube and elbow meter
also, you will see that we will use the same principle. That means there is a differential
pressure and that pressure is calibrated in terms of flow, flow velocity or the volume flow
rate. So, the contents are Pitot tube, elbow meter, then rotameter is another example, but
rotameter as you know, it is a, it is not the differential, it is not the flowmeter. It depends
on the, based on the differential pressure measurement, it is basically variable area meter.
So, that we will discuss in detail how it works actually and also we will solve one
problem on the Venturi meter. So, the differential pressure meters, so either Venturi or
orifice, so in this particular lesson we will consider one problem on the Venturi meter,
right?
2
So, at the end of this lesson, the viewer will know the principle of working of Pitot tube,
advantages and disadvantages of elbow meter, also its principles also we learn in this
process. So, principle is common. Principle of working of Pitot tube, I mean elbow meter,
your rotameters, everything will be, all will be discussed here. But also we will discuss
advantage and disadvantage of the elbow meter. Also we, we will see that rotameter can
be used as a linear sensor that is a great achievement, because in most of the differential
pressure, in all differential pressure measurement based flowmeter, you will find its
relation between the flow and the differential pressure is non linear, so which creates a
problem. So, we need some other circuitry to linearize, which is not necessary in the case
of rotameter. Though it is usually used for very short range of flow measurement, range
is not much, right?
So, Pitot tube, you see, it is like this. It measures the velocity at a point in a fluid, right?
That means, it will measure particularly velocity at a point. Unlike the, unlike the orifice
meter and Venturi meter which measures a, over a area, I mean, if you look at the, along
the pipe, if you go along the pipe, so but the Pitot tube basically measures the velocity at
a point, particular point in a fluid, right?
3
So, this is very important that means what I am saying, here you see if I have, if I take a
white board, that means what I am saying that if I have a pipe, you have seen that in the
case of orifice meter or if I have a, in the case of Venturi meter we have seen, so it is
average the velocity at the whole point. So, it is like this only, it average the velocity. If I
am interested by the Pitot tube, if I am interested to measure the velocity, suppose at this
point of the fluid or at this point of the fluid or this point of the fluid or this of the fluid
that is not possible by Venturi or orifice meter, right?
But, it is possible by the Pitot tube and another important thing is it is an open channel
meter, it is not closed. So, you can use it as a closed channel meter, but you can use it as
well as open channel meter. That means I have a tube what we actually do we will
discuss in details. So, this tube is installed inside the pipe. Suppose this is a pipe, so
depending on installation, I can move it little up here in this here or I can move it down.
So, I can, at a particular point I can measure the velocity of the fluid, right, like where
this is also an open channel meter, but we can use it as a closed channel meter also.
4
It measures the velocity at a point in a fluid. It is an open channel meter that I told you. It
is suitable for investigation around an aerofoil in a wind tunnel or the measurement of
velocity profile in a pipe, prior to the installation of the permanent flowmeter.
Sometimes, some estimation is necessary. That means how much is the flow, I mean
before I mean installing your flowmeters like Venturi meters or orifice meter or … tube
and flow nozzles, so we need some estimation. So, in that type of situation, Pitot tubes
can be easily inserted and take the measurement, because you know that in the case of
orifice meter, Venturi meter, it is a very cumbersome device, it needs a lot of time for
estimation. This is not, you cannot spend that much amount of time, neither what will
allow to disturb the, the installation.
But, by using the Pitot tube you can overcome that type of difficulties; that I can estimate.
As well as I can measure quite accurately the flow, flow velocity or flow measurement.
You can see, there in the case of Venturi meter and orifice meter also we are actually
measuring the velocity, then we are multiplying the area of cross section, finding the,
finding the total volume flow rate. Here, we will find the velocity, then with the area, if
you, of the cross session of the pipe if I multiply, obviously you will also get the volume
flow rate.
5
The principle of operation of a Pitot tube is like this. If a solid body is held stationary in a
pipe in which the fluid is flowing with a velocity, as the fluid in the tube approaches the
body, the fluid particles are decelerated and until at a point directly in front of the solid
body, the velocity of the fluid is zero, right? What does it mean? Let us take a blank page,
then it will be more clear.
6
See here, so what I am saying, suppose a fluid, I have a pipe and liquid is flowing
through this, right? Now, I put an object, suppose like this one, a solid object, then what
will happen? See, here the liquid will flow; flow in this direction, flow in this direction,
all this direction? What will happen to the flow which is coming to this, hitting this
portion? It is coming to rest at that position. Supposed to be the, if it is raised the pressure
will be very high compared to the pressure of the fluid here, right? So, that means the
liquid which is coming to the rest in contact, so liquid, so what will happen? The fluid
will be decelerated.
When it is approaching this solid body, suppose if I take, this is the solid body, fluid, the
fluid will be decelerated. So, at this point, it will be total stop, totally stopped. So, the
velocity of the liquid at this point will be zero exactly, identically zero and the velocity of
the fluid, suppose this is V, so this liquid is also flowing with the velocity V, then it is
flowing out. Again it is flowing out, right? But, whenever the fluid is flowing here, so
obviously the, just in front of this one, if we look at here, at this position the fluid will
come to rest. Using this principle actually people developed the Pitot tube, right?
Just take the, so again I will repeat.
7
If a solid body is held stationary in a pipe in which the fluid is flowing with a velocity, as
the velocity in the tube approaches body, the fluid, that solid body, obviously this is the
pipe and the solid body, the fluid particles are decelerated until at a point directly in front
of the solid body, the velocity of the fluid is zero, velocity of the fluid is zero, right? So,
accompanying the deceleration is an increase in pressure, right? That means whenever I
told you earlier also that means what will happen?
That means I have a solid body here. Again, I am telling, so liquid is coming and hitting
here, so velocity at this point is zero, right, velocity V, it is flowing, right? This is a pipe,
pipe is flowing, so what will happen here? I will take a new page.
8
So, like this one, it is flowing like this, right? So, what will happen, you see here. So, I
have a solid body, so which is coming and coming to …. So, the pressure at that point
will be high compared to the or highest I should say, compared to the fluid, I mean static
pressure here, right? So, accompanying the deceleration is an increase in pressure.
9
So, it is the process of converting the velocity head to an additional static head, right? We
are calling it head, because it is a, basically if we look at, it is basically we are, we are
writing the equations in terms of the liquid head, right? That is the reason we have to,
because I mean if the pressure is, as you know, hdg if you divide by density and the
acceleration due to gravity, so obviously it can be expressed in terms of only the height.
That means assume the liquid, the pressure is usually defined by this one, is not it?
P equal to height into d into g. So if you express it, if you multiply it, I mean divide by d
and g, obviously I will get a, if I write the Bernoulli’s equation only in the, sorry, only in
the case of, in the terms of height, so it will be all in, I mean head. So, that is the reason
sometimes we call it head.
10
It is a process of converting the velocity head to an additional static head, right, because
additional static head is coming, velocity is becoming zero, the pressure is increasing, so
that head is also increasing. The velocity of the fluid can be found by measuring the
differential pressure. Again, the same principle, because this also depends on the,
basically the principle of work depends on the measurement of differential pressure. At
the impact hole or stagnation point, the fluid is brought to rest and this point therefore has
no kinetic energy, right? If the velocity is zero, obviously that what is the kinetic energy?
Half MB square; if the velocity is zero, the kinetic energy also will be zero, right?
11
So, here you see the, a Pitot tube. Look at very carefully. Right? So you see, this is the
Pitot tube, our liquid is flowing. It is installed in a pipe; I am not showing the pipe, so
pipe will be there, right, pipe will be there, so and inside, the pipe it is going. If I take a
different colour pen, so it is installed in a pipe. Please note, always installed in a pipe,
right? So, this impact hole should be always in a direction opposite to the flow. Flow is,
flow in this direction, so directly it is impacting here. So, it is to be installed in a direction
where this that means I am saying like this one. If I can see the camera here like this one,
you see the, this is Pitot tube suppose, right? So, this is our static hole, right, impact hole
or stagnation point I should say, so what will happen here? See, the liquid is flowing here
and hitting this position, right and liquid is flowing over this one. So, there is a hole on
this, on this side. This side there is a hole. The fluid is flowing in this direction and it is
going out. So, liquid is, I mean this, it is impacting here, so there is a point where the
velocity of the fluid will be zero. If we consider this as solid body, so velocity of the fluid
at this point will be zero, kinetic energy will be also zero. But, there is a pressure …..
inside the pipe also. That pressure we are calling it static pressure, right?
12
And this pressure, the pressure at this one will be the stagnation pressure so obviously the
stagnation pressure see the velocity is zero will be much higher than the static pressure
right
so this static holes usually can be collected as it happens in the case of Venturi meters
this static holes can be this static pressure can be sense by physio meter rinse that means
it is averaging the ah pressures and this impact hole stagnation pressure we can be
measured by if i just take out one hole here and take it out
so these two will give you differential pressures right this minus this this pressure P one
minus P two will give you the differential pressure as a flow velocity increases P one will
increase obviously delta P will increase what you have to see the delta P differential
pressure [Noise]
so this is equal to P one minus P two right [Noise] so this P one minus P two so this is
{stag} (00:16:17) this pressure differential pressure so this is our stagnation pressure so
the {vel} (00:16:21) of velocity increases P one will be also increase
so delta P also will increase right so this way so this by measuring this differential
pressure so i can calibrate this in terms of flow velocity right and if i know the area of
cross sections of the tube where it is installed that Pitot tube i can measure the volume
flow rate
13
or if i installed on a some moving body this entire Pitot tube is install and a moving body
is like an aircraft obviously what will happen you know that um having measure the
aircraft velocity
(( )) (00:16:57) velocity of any vehicles also only thing the problems we need a (( ))
(00:17:01) flits because if the there is a if there is a this hole is the this is small hole you
can see here if this hole is clot by some particles or dust
so that will give us error in reading right that is happened actually in the case of aircraft
as you know when it goes to {lar} (00:17:17) i mean huge height
so as you know the goings and all this things they are going very height and around thirty
three thousand feet three thirty five thousand feet at the situation there is a ah ice
formations and the stagnation point then what they do they put a heater coil around this
one right
so continuously is melting that ice so there is no question of entiting erroneous reading so
the flow can be calibrated in terms of actually measuring the differential pressures i can
measure the flow velocity or the velocity of the {vehi} (00:17:48) vehicles and aircraft on
which it is installed lets go back theory again [Noise] right
(Refer Slide Time: 00:18:00 min)
now this is the you see the this relation between the velocity of a air passes the
differential pressure right x axis we have plotted the velocity of air and y axis we have
plotted the differential pressure delta P
14
which is in Pascal and velocity air is in meter per second and density of one point two kg
per meter cube and differential pressure versus velocity of air measured by Pitot tube
right
so this is the figure as you can see we will give some calculation is non linear in nature
you can see okay
so this creates a problem both the kinetic energy and the pressure energy we will present
at the static holes because fluid is moving at those positions right [Noise]
assuming energy conservation and no frictional loss and no heat loss obviously we can
write P one by rho zero gz one P two by rho plus V two square by two into gz two
you can look at very carefully all the units of this one will be in you see here that
reducing if i take meters or seconds square right so g will be ah z will be in meter
so it will be meter per second whole square this one this also will be meter per second
square right P two it can be as a pressure so Newton per meter square right so if you ah if
you simplify this one suppose this Newton’s in kg P is in Newton right
so it means ah it means that ah P is in Newton means signifying that is a kg meter per
second square so if you divide by rho that is kg per meter cube again it will be so this P
by rho will find will be in meter per second square right
so it will be like this right so all the units are same units of which item will be same
because this is the there is no velocity of the ah velocity at the stagnation this is a
15
stagnation part the Bernoulli’s equation is applied at the stagnation point this is the
Bernoulli’s equations apply at any other point of fluid right
here P one is a stagnation pressure P two is the static pressures (( )) (00:20:57) the z one
is the elevation of the static i mean stagnation point and z two is the elevation of the your
static point right if we have this slide we can go to the next slide [Noise]
so where z one z two are the elevations of the holes above the datum line and g is the
acceleration due to gravity which is nine nine point eight one meter per second square
and if z one equal to z two then we can simply write v equal to two P one minus P two
rho so this equation number two
now whether the rho varies let us look at that now from equation two we can write delta P
equal to so this is delta P delta P equal to ah v square into rho divided by two isn’t it
differential pressure equaled half of rho v square
16
where delta is a differential pressure P one minus P two so v obviously i can see here you
can see this most important equation count for all differential pressure from it you can see
this is v equal to v proportional to [Noise] root over delta P right
so delta P if you calibrates in terms of v obviously there is non linear relationship right
[Noise] so for air at twenty degree centigrade and pressure P two of ten to the power of
five Pascal that is static pressures and i am talking about with rho equal to one point two
kg per meter cubed gives delta P equal to point six v square
thus at v equal to [Noise] ten meter per second we have delta P equal to {sixteen}
(00:22:38) sixty Pascal and delta P by P two equal to six into ten to power minus four
17
and at v equal to hundred meter per second delta P equal to six into ten to the power three
Pascal’s and delta P by P two into six into ten to the power minus two
what does it mean it means the when the flow velocity ah the this the low value of delta P
by P two ratio means that for the velocity less than hundred meter per second hundred
meter per second is quite fast and quite high speed right
that mean the difference it’s a extremely high speed hundred meter per second right the
difference in density between the air at the stagnation point and static holes is negligible
right
the error introduced by considering the incompressible fluid is within one percent right
now differential pressure transmitter is a special type of {difren} (00:23:32) differential
pressure transmitters is necessary for ah particularly in this case due to low differential
pressure because if the ah if the fluid {velo} (00:23:45) fluid is actually air gas
so the differential pressure also will be less because ah as you know the differential
pressure also depends on the density of the fluid right so the air as a lesser density so due
to low differential pressure so the your differential pressure also will be low
so we need some more sensitive differential pressure sensing element because this
differential pressure first you will sense and before converting to the electrical domain
that is four to twenty milli ampere of current domain i need some ah i need some high ah
i {nee} (00:24:18) i need some higher pressure higher differential pressures right
18
so i need some more sensitive ah i mean this type of system which will convert this ah
your differential pressure to some other change like capacity change or any other change
displacement change that can be converted in the electrical current domain of four to
twenty milli ampere
so it gives a four to twenty that voltage LVDT unbalanced voltage that means the
secondary voltage which is in positions can converted easily to the current domain of four
to twenty milli ampere output for differential pressure
a scheme of velocity you see everywhere its likes this one you have seen that in the case
of this LVDT having the case of this flow measurement also what is a minimum flow that
must be described right what is maximum flow that also in the process it is fixed
so what will do will accordingly we will fix of four to twenty milli amperes so the
calibration is most important all the calibers so for the minimum flow velocity we must
get four milli ampere of current for maximum flow velocity we must get it twenty milli
ampere of current
accordingly our electronic circuit that converter which will convert because always you
know that ah there are many simple circuits are available for to convert this type of
signals voltage to current signals right
so depending on the whether you suppose have a zero ten volt accordingly for zero volt i
want four milli ampere for ten volt i want twenty milli ampere
19
so that type of things can easily made right so similarly here also instead of suppose zero
volt i have a input voltage (( )) (00:25:50) output sign getting suppose ah ah zero two five
volt no problem
see if it is zero to five volt accordingly i can set my circuit some reference voltage raised
terms by which i can make four to twenty milli ampere of current right now scheme of
velocity measurement is shown in figure three
you see here that you can see here the input is velocity v right i have a Pitot tube i am
getting into differential pressure of delta P so i have DP transmitter so DP transmitter is
basically here not capacitive it is a heavy duty based in four to twenty milli ampere of
current is coming we have data acquisition system
and we have PC data acquisition system is obviously it will convert this current to the
voltage domain then it is digitizing it before we giving feeding to the {p} (00:26:42) PC
where i can record the the flow velocity and if i have to take some action that means i
have to feed back suppose this PC output through that data acquisition card again can go
back to the control one to reduce the flow
ultimately the what is a use of this Pitot tube and all this things i have to set some already
the pre prescribed the this much of oxygen will flow this much of hydrogen will flow in
the pipe so to set that so i have to measure this i mean velocity then if feed it back to the
some control valve
20
so that by which it will either open or close to keep to being flow velocity to the
prescribed level right
now elbow meter ah with this ah elbow meter is a different kinds of system it is actually
usually used in a huge installation of the pipe is not is not suitable for the small bending
of pipe so for the large bending of the pipe
now in all the cases except Pitot tube Pitot tube usually used for the gas you find that ah
you will find that the there is a large differential pressure and and and there is a loss
permanent loss have the permanent loss in the both in the case {or} (00:27:55) orifice
meter or Venturi meter
in the case of orifice meter we have seen that this loss is more in the case of Venturi
meter permanent loss is more in the case of Venturi meter we have seen the permanent
loss is less
now aerometer is an instrument its also based on the differential pressure measurement
but that permanent pressure loss is not there
but in fact it is there in fact not in that way because whenever there is a pipe bending
there is a pressure loss as you now head loss is their so utilizing that principles without
having any addition loss for the installation of the installation of the ah meter we can
have a elbow meter
21
let us look at what is that the orifice meter flow nozzle dall tube and Venturi meter we
have seen cause permanent pressure loss in the system this will create permanent pressure
loss in the system right all this flow nozzle dall tube less or more so obviously it will
make some permanent pressure loss
we have seen that in the case of Venturi meter there is a i mean there are different types
Venturi but long short in the long we are much {be} (00:29:05) better pressure recovery
the short we have less pressure recovery in the case of orifice meter pressure recovery is
watts right aerometer does not introduce any additional losses in the system send it
simply replaces and existing elbow or pipe bending that is being used to change the
direction of flow right in a process we will find the always you need this type of things i
am need some pipe bending somewhere okay
because pipe cannot all pipe cannot be straight in the process so you find a some there is
pipe bending so in that bending will install that meter
so that no additional loss will be there [Noise] so aerometer does not introduce the any
additional losses in the system since its simply replaces and existing elbow or pipe
bending that is being used to change the direction of flow right
you see here the elbow meters we have it looks like you see that i have this is a this a ah
section of pipe bending i have shown here you see this is a section of pipe bending you
22
can see here this is a section of pipe bending right that means there is a pipe here there is
a pipe here the continuous pipe and there is pipe going also here in this direction right
so just that pipe bending at the {a} (00:30:27) at the pipe bending i have installed this
elbow meter fluid is flowing through a velocity v and it is coming out going out through
this velocity and there is a two only two pressure tapings there one will be inside the pipe
another will be {ins} (00:30:41) outside the pipe
you see what will happen the liquid which is flowing through this {pi} (00:30:46) outer
side of the pipe it will must faster velocity than the pipe which is the liquid is flowing
through the inside this one right it is usually increases in this direction
it is maximum here and by {min} (00:30:59) velocity is maximum here in this direction
and minimum in this {dire} (00:31:01) here every where the flow velocity if you take any
point to flow velocity suppose to be same right but here what will happen the liquid
which is flowing very near to the pipe bending it has much higher lower flow velocity
and here we have a higher flow velocity according to the {bondis} (00:31:19) Bernoulli’s
theorem obviously what will happen the pressure to this one will be [Noise] on the liquid
side excuse me liquid i mean [Noise] will be much higher compare to the pressure
because here the flow velocity will be flow velocity will be higher right so we have ah
datum point Zi and Zo from the datum level we have taken the height of the pressure tap
this is the ah i mean inside the {paperpe} (00:31:42) pressure tap this is outside pressure
tap
this is our basic principal in the elbow meter so what the advantage that means you see
that i am not putting any restrictions inside the pipe like orifice or Venturi clear
23
so velocity pressure and elevation above the datum level for pressure taps on the inside
and outside surfaces of a ninety degree elbow can be related by the expression is that
shape is ninety any bending in a industrial process is ninety degree
that is the reason we call it ninety degree elbow can be related by the reexpression Ck
equal to Ck multiplied by v square by two g P naught by rho g plus Z naught minus Pi by
rho g minus Zi equal to this equation number three right
where Ck is the co efficient that depends on the (( )) (00:32:36) of the elbow that means
shape and size of the elbow a normal of Ck range from one point three to three point two
units you see here if i go back
24
so Ck v square by two g Ck v square by two g P naught by rho g Zo Pi rho g so all this
units let us look at P is a Newton per meter square rho is kg per meter cube and g is meter
per second square
so all the units all the heads are in meter you can look at so P by rho g is in meter and Z is
meter so all are in coming in meter right
so volume flow rate will be expressed as Q equal to Av A (( )) (00:33:30) root over Ck

two g P naught by rho g plus Z naught minus Pi by rho g minus Zi C into A we have
combined ah instead of writing A and root over Ck we write C into A two g P naught by
rho g plus Z naught minus P naught up and rho g minus Zi equation number four right
where A is the this should be where please note where A is the area of cross section of
pipe in meter square
25
the value of C ranges from point five six to point eight eight and the primary advantage
of the elbow meter is the savings of {ex} (00:34:18) extra pumping cost for the range
installation that is a ah good amount of saving right
so the primary disadvantage is that of each meter must be calibrated on the site its very
difficult to i mean find the flow coefficients as we know there is value of this C ranges
from here all has given the range
so we have to install it depends on the slight pipe bending is not exactly ninety degree
something else may happen suppose it is {exa} (00:34:43) internal cross section exactly
not in circular in shape
so obviously everything will ah change the installation change the calibration constants or
the your the {da} (00:34:54) discharge co efficient C right a flow co efficient C so that
creates the problem
so each and every installation that is um true for even for {al} (00:35:02) almost all the
installations right though we can suppose ah we have a thermo (( )) (00:35:06)
we can calibrated we can calibrate in the laboratory but it’s better to install {inst}
(00:35:11) calibrate at the site itself
the primary disadvantage is that each meter must be calibrated on the site right so the low
operating cost can usually justify the calibration cost the low operating cost and usually
justify the calibration cost
26
the elbow meter again the that restrictions the upstream flow restrictions and the
downstream flow restrictions that means obstructions that means any pipe bending this
that ah will be there so that is rather restriction is there which we had previously in the
case of orifice and Venturi
the elbow meter requires a minimum of twenty to thirty pipe diameters of unobstructed
upstream flow to reduce the turbulence and swirl for accurate measurement otherwise the
flow straightners ah are to be installed in to stabilize the flow as it is used in the case of
orifice meter
it’s a bundle of tubes all held by insert and put inside the pipe so the liquid will be i mean
so it will eliminate on the turbulence or reduce the turbulence and where swirl right
27
now rotameter is ah slightly different ah um it’s a ah meter it basically depends on the ah
ah is a variable area flow meter first of all so with this we are finishing this differential
pressure meters we are starting now rotameters
rotameter you see the basic principle is widely used for meter for flow rate indications it
is not used for transmitting instruments or rather so i mean {monite} (00:36:47) it’s
basically monitoring instrument
but you will find this meter is extensively used for many ah crucial applications or vital
applications like biomedical applications also and also in the process we will find these
are in plenty right
but basically it is a so far if the ah if the indicating instrument is concerned i think
rotameter as the only choice if we look at the flow meters we have a several flow meters
transmitting facilities and all these things
but if i want to make a simple ah i mean indicating type of instrument or monitoring type
of instrument rotameter is the only choice please note that
the meter consists of a float or bob bob is typically called in the industry people call it
bob i don't know why anyway ah with in a vertical or transparent tube tapered to an
increasing cross sectional area at the outlet it must be tapered okay
if its ah if does not tapered that is not a rotameter so the meter consists i will repeat the
meter consist of a float or bob typically called by the engineer in process within a vertical
transparent tube tapered to an increasing cross sectional area at the outlet right
28
the fluid entering through the bottom passes over the float which is free to move only in
the vertical directions the float can move only in the vertical directions right that is a
problem rotameter always use to be install in the vertical direction right
rotameter is always installed in the upright positions or in vertical directions when the
fluid is flowing through the meter the three forces are acting on the bob these are let us
look at the rotameter now
29
you see this is our rotameters right i have a can i take yes see there is flow in the liquid is
flow in this directions and liquid is flowing out this directions so it is to be always install
in this directions vertical direction right
there is a graduated square rotameter is {com} (00:38:45) accompanied by graduate {sc}
(00:38:47) you see this tapering is so small that it is very difficult to if you look at the
rotameters it’s like a very difficult it’s a usually it size like this pen most of the rotameter
we find
it’s like this pen the size is very small it’s not that very huge in size you will find this size
right and it is slowly increased in taper angle is slowly
so bottom it is a smaller angle included angle so it is slowly increasing right so that
means it is increasing here slowly it is increasing in this direction
but this tapering is so small that is with the naked eye it is very difficult to visualize
because these rotameter since it is made of hot i mean ah glass or i mean strong glass
what they do is you find its puts on a casing also the back side it is getting only some
front side you can see right
that you have seen in the case of the in the our first lesson one we have shown some
rotameter so liquid is coming in this direction that is through pipe it is coming in this
direction it is going up then it is moving in this direction
again i will tell liquid is flowing in this direction it is going inside the pipe inside the
rotameter liquid is flowing through this then it is going out right and the float is moving
inside the pipe float can only move in this {di} (00:39:59) vertical directions right
float can when the flow increases the float will move like this and when the flow
decreases float will come down here and it will totally fit the inside diameter of the pipe
float is designed like that it will fit the inside diameter of the pipe right
there are different shape of the float that we will discuss later on [Noise] so three forces
you see the three forces are acting on the rotameter what are those forces weight of the
system okay mass of the system buoyancy force and drag force
now weight of the system that will act down word and buoyancy and the drag force will
act up ward right so this FW will always balanced by FD and FB then FW equal to FB
plus FD that is always so rotameter you can see this basically in auto balancing system
30
because you see is FW and FB will remain constant because that is the weight and the
buoyancy of a {liqu} (00:41:08) the liquid is same if the liquid is same that means same
type of liquid is flowing through the pipe it does not matter what is the velocity density
will remains same
if the density is remains same density of the fluid remains same here buoyancy force is
also will remains same and since we are not changing the bob that FW also will remains
same so FB plus FW will be equal to FD then FD also should remain same
that is the main [Noise] main interesting point in rotameters rotameter is a basically when
it is auto balancing systems if we look at its very interesting part this is a auto balancing
systems right
let us look at that principle of operations how it works for a given flow rate the float
remains stationary when the weight of the float is balanced by the buoyancy and the drag
force
buoyancy and drag force will act up ward because flow of the liquid is always from the
lower to like flow of the liquid is always like this one that means i have a rotameter here
so flow of the liquid is always in this direction right {buo} (00:42:19) buoyancy and the
drag force so the float will this will act down ward FB that’s we have seen already FB
weight and FD and FB like this one
31
it is a auto balancing system that’s you must note how the why i am calling it auto
balancing system it will be create from this subsequent points the annular area between
the float and the vertical tube varies continuously with the vertical displacements of the
float or bob how this annular area is changing means yes obviously it will change
you see look at let we take a white page you see rotameter is looks like this let me take a
different color pen um let me take another one so we have a rotameter here [Noise] if i
look from the top what will happen it will look like this you see here the liquid is flowing
through this
32
so what is this annular area as you move bob what will happen the area cross section if
you find leaving this one we increase area cross section will remain i mean it is always
increasing as we go up right as we go up the area cross section will increasing
area cross section of the float is also same so the annular area that means if you look at
here annular area which is the shaded area [Noise] will change as the this flow goes up
the means i am telling that the area of annular area at the position of the float
you had think of the {ar} (00:44:13) annular area at the position of the float that will
increase if the float goes up even the at the float {go} (00:44:20) comes in it is annular
area will be no more so at that time what will happen float will just fit on this one
in that case it will there is no no annular area so at the flows this float goes up annular
area increasing and increasing right this principles we are discussing clear
we are so the annular area between the float and the vertical tube varies continuously
with the vertical displacement of the float or bob right
for a particular liquid the weight of the float as i told you the weight of the for a particular
liquid weight of the float is constant buoyancy force is also constant therefore the drag
force is to be maintain as a constant level that for must be constant also for a particular
liquid the weight of the float and the buoyancy force are constant therefore the drag force
is to be maintain at a constant level right
33
since the area of cross section of the float is constant the pressure drop across it should be
constant since the area of cross section of the float is constant the pressure drop across it
should be constant right this is a key point
now when the float is in particular position for a flow rate the differential pressure varies
with the square of the flow rate clear when the float in a is in a particular position for a
flow rate the differential pressure varies
what is the differential pressure across the flow across the float differential pressure
across the float when the float is in particular position a flow rate for flow rate the
differential pressure varies with the square of the flow rate right
so now therefore to keep this differential pressure constant for some other flow rates
when the differential pressure differential pressure must be constant otherwise the drag
force will not be constant that force must be constant because it is equal to FW plus FD
FW plus ah sorry
i mean ah FW equal to FD plus F buoyancy weight must be counter balanced by the
balance in the drag force so drag force must be constant
so to keeps in the differential pressure is constant so obviously what will happen the area
of the annular area must change that means the float must move up and down right
34
therefore to keep the differential pressure constant for some other flow rate the annular
area in between the float and the vertical tube must change must change means if the
differential pressure as the flow increases differential pressure is constant
so to keep that drag force constant what will happen the flow area must change right right
now the variable area is provided by this vertical tube this variable area as i told in the
earlier when see between the float annular area that’s if it if you doing the tap or then
there is type of thing will not be achieved
so the position of the float can be made essentially linear with the flow rate by making
the area of cross section of the tube vary linearly with the vertical height right
considering the incompressible flow the volume flow rate is expressed as it is expressed
as this is the equation Cd equal to At minus Ab two gV Vb rho b rho f Ab rho f At Ab
so let us look at what are those so this is the flow coefficient or the this such coefficient
At is the area of the tube area of cross section of the tube area of cross section of the float
sorry bob or float whatever area of cross section of the tube area of cross section of the
float area of cross section of the tube
this is the volume of the bob okay this is the density of the material of the bob this is the
density of the fluid this is the area of the bob and this is the density of the fluid okay
we have written all this in the in the all the legends are given in the next slide let us look
at where keys is the volume flow rate in meter meter cube per second
35
then Cd is the discharge coefficients At is the area of cross section of the tube in meter
square Ab is the area of the cross section of the float or the bob in meter square Vb is the
volume of the float in meter cube
rho b is the density of the float in of material kg per meter cube rho f is the density of the
flowing fluid in kg per meter cube okay
now if you assume that there is no variation of the discharge coefficient with the float
position we assume that discharge coefficient does not changes with the float position we
changes obviously
36
and if we assume that At minus Ab square by two is much much less than one if i go to
the previous slide is more clear okay
i am assuming that the Cd is constant okay and this portion that’s At minus this portions
that means i am talking of this portion this At minus Ab by At whole square is much
much less than one then [Noise] okay
is At minus Ab less than the equation five can be simplified to Q equal to K At minus Ab
because all that items to be constant is not a in that equation so the volumetric flow rate
equal to K At minus Ab
where K is equal to Cd root over another square root two g volume of the bob Vb rho v
minus rho f upon Ab into rho f right this is our constant K
so now if the cross sectional area of the vertical tube varies linearly with the float position
with the if we vary the cross section of that i mean tube in such weight varies linearly
with the i mean with the float position
then i can write the volume flow rate equal to Q equal to K one plus K two x okay to
constant and x x is the position of the float right position of the bob right
so the rotameter usually has an accurate range of ten is to one that is better than the
square root sensor so that is quite obvious so it is a square root sensor much better than
the square root sensor
37
square root sensor we have seen that at the ah um twenty five percent less than twenty
five percent of the maximum full skill range the error is this reading is very much
erroneous that is not the case in the case of rotameter right
now shape of the {rotam} (00:51:28) so what will see that ah in the case ah previous we
find that [Noise]
that’s we have seen also that and we take a ah a different color so we have seen that its
tapered like this one okay the float is moving like this one so it’s in a casing okay and it is
graduator scale is there right [Noise]
so this is liter or we need all liter per second whatever the way you like actually
represented right so this is the float so this is calibrate in the i mean volumetric flow rate
or i mean velocity
so is the volumetric flow rate it is calibrated so whenever by looking at the position of the
bob i can if i look at here i can tell that the what is the flow velocity right so this is very
important in the case of ah rotameter
so it is basically used for ah the basically used for the monitoring instrument or indicating
instrument but not for the in transmission instrument right but that does not neccesarily
mean that’s accuracy is poor accuracy is quite good in the case of rotameter
as i told you i mean you will find that it is extensively used in the case of anesthesia as
you know that ah when the patient i mean under operation first they put an injection
38
physicians for ah to go the patient under this say that ah the patient will go some
subconscious state right to maintain that state they must have continuously supply the gas
so that the patient will remain so to how much gas they will put to the patient nose so that
will be the measure by the flowmeter that the rotameter and there are the float ah ah let
me go back again i am sorry [Noise] shape of the rotameter you will find
ah that ah the float is actually they are using some light heavy plastic short of thing so by
looking at the position of the float i can tell how much gas is i am giving to the patient
that because this very important
those who who are the anesthesia for the actually looking at and determining the
regulating the valve there controlling the how much liquid is flowing how much fluid is
going to the patient right
now floats with sharp edges are less sensitive to fluid viscosity that changes with
temperature right so the ah you find that’s because viscosity of the liquid as you know the
changes with temperature right so the but if i make the float the sharp edges that will
almost independent to viscosity right
vertical tube of the rotameter is made of glass to make it a monitoring instrument
rotameters are used in applications so the accuracy is not of prime concern but it’s not
that in accurate also otherwise it cannot be used in biometric application
39
such like very crucial when the because if you give more gas ah to the patients we patient
will die and if you i mean ah and if you reduce this supply of gas to the patient
so what will happen patient will come out of the subconscious state i mean physicians
cannot operate so till the operation is complete the patients will be on the that state
now here now we will solve one problems on the Venturi meter you see here now
problem looks like this
a Venturi meter is to be used to measure the flow rate of water in a pipe of diameter D
will be point two meter the maximum flow rate is two one three six meter cube per
minute Venturis with throat diameters of point one zero meter point one two meter and
point one four meters are available
choose the most suitable Venturi meter assuming the differential pressure at maximum
flow is nine one eight kg per meter cube
40
and calculate the accurate value of the differential pressure developed across the chosen
Venturi at maximum flow rate right this is our now we are given some ah chart also
yes it tells for orifice plate this is beta equal to ah we have seen that some chart for
Reynolds numbers how much the ah flow coefficient changes
41
so table for Venturi also we are given this is a pipe and throat diameters for the different
Reynolds numbers how the discharge coefficient changes okay
that is we have given these are tables are necessarily for solving the problems of the
Venturi meter or orifice plate meter right this is another chart we are given
last chart we are giving that is the the Reynolds numbers have the beta changes you know
if orifice plate of point zero five mille {met} (00;56:13) ah meter that is a dimensions of
the orifice ah so we find that the how much the Reynolds number changes right so let us
solve the problem
42
here you see the problem is like this one ah here Cd we can see for Cd is equal to point
one okay it is not right yeah for point ah sorry let me take new page
so rho equal to one thousand [Noise] kg per meter cube so from that if i apply the our
Venturi meter formula will find that Q equal to i am getting [Noise] three forty meter
cube per second
now here i am taking that d ah ah equal to point zero one meter for which Cd is coming
from the chart point nine eight eight okay Venturi meter always you know the flow
coefficients is very high right
43
then we have we can find for d equal to point one two meter we have calculated from the
formula that that Cd will come point nine eight seven [Noise]
so the Q two the this is a Q one if i assume the Q two will be equal to ah five zero seven
point six meter cube [Noise] per second right
so if we take a new page now for D equal to [Noise] point one four meter will find Cd
equal to point nine eight five meter point nine point nine eight five so Q three equal to
equal to seven three seven meter cube per second
so we can see that the point ah the orifice meter with point one zero ah meter diameter is
the {be} (00:58:37) best chose right
because that is the most close one and in this case for this type of situations now in the
case ah we can calculate that the delta P ah max if i apply this again this our main
formula
so now will apply this one point two zero that means it will be ah will if i take a new page
it will like this one
44
so it will be three point five six into ten to the power four point nine eight eight by point
nine six eight you will find so you will put the all other pi twenty five root over two delta
Pg you will find that the delta P max will be one zero zero six point one kg [Noise] per
meter square right
so this is our equation right so this is so best chose is point one zero meter and delta P
max will be one thousand six point one kg per meter square this ends the lesson thirteen
thank you
45
Prof. A. Barua
Lecture - 14
Flowmeter – ΙΙΙ
Welcome to lesson 14 of Industrial Instrumentation. In this lesson, we will continue

with the flowmeters. As I told you repeatedly before that the number of flowmeters
are huge in, in numbers. So, we have to continue on several lessons and we have to
discuss details, in details all the flowmeters. So, in this lesson, basically we will cover
the one open channel meter which is very widely used for irrigation purposes that is
weir and also we will discuss direct electrical output flowmeter that is turbine
flowmeter, right? So, let us look at the contents of this lesson, Flowmeter III as it
happened.
Now, we will have contents, so we will first discuss the weirs, its principles, its
application, what are the different types of weirs that also we will discuss and then we
will discuss the, a variable reluctance tachogenerator, because the, using that principle
we will use, we will make the turbine flowmeter. Turbine flowmeter advantage is the
direct electrical output. So, in many applications it is popular and moreover, it is
unlike differential pressure flowmeters, you do not, we do not need any conversions
1
and square rooting, all those problems are not there in this type of flowmeters. We
will also discuss the turbine flowmeter.
Now, at the end of this lesson, obviously the viewer will know that how to use the
weir in the systems, I mean how the person can be, the person can use the weir for
measurements of water flows, especially mostly it is used for water flow in an open
channel like canal or duct, right? That means principles of operations of open channel
meter that is weir. Open channel meter, we have already discussed the, not only weir,
we have, actually pitot tube is also used. But pitot tube as you know it is mostly used
for the, used for the gas and it is not suitable for the dirty liquids, because when we
are talking of the measurement of the flow of water in a canal or river, you cannot
except that it will be very clean. So, it will be extremely dirty in that sense, so we
need some other rugged flowmeter.
So, in that sense, whereas no other, there is no other alternative than weir to use the
measurement of flow in open duct, I mean open channel, right, like canal, river, like
that. So, we will also look at the magnetic pick up, because this magnetic pick up, we
will find that we will use in many applications for measurements of velocity and all
those things. So, the principle is same, whether I, suppose if I want to know that an
RPM of a motor, so that can be utilized and the magnetic pick up is also used there
2
and also if I want to know the, the liquid flowing in a pipe, so what is the liquid
flowing in a pipe, if I use some magnetic pick up, so that is also, I mean that is the
part of the turbine flowmeter that is the reason we will discuss magnetic pick up also;
also, the turbine flowmeter with direct electrical output. So, these three things we will
discuss in, I mean people, I mean the viewer will know the details of these three
components.
Now, let us look at the weir.
Weir, you see, it is an open channel meter, first of all. That is I told you repeatedly it
is an open channel meter, it is not a closed channel meter. Now, if the flow rate of a
liquid in an open duct or channel such as a river or canal is required, a weir can be
used. There is no other alternative. As I said it can be used, but in fact actually there is
no other alternative than using weir. It is extensively used to get an estimate of flow
of water in the canal or duct for irrigation purposes. So, whenever there is irrigation,
so how much water is to be given to a particular land and all those things, details I
mean, this estimate is only possible by, if we use weir, right?
So, it is very much suitable for measurements of the liquids in the open. It is not for
gas. So, it is only for the liquid, mostly for water and principle operation of weir and
3
moreover one thing I forgot to tell you that it is, you see that it does not depend on
the, depend on the temperatures and all those things. Many other sophisticated, as you
know there are many complicated, as you know flowmeters are there, so but it does
not depend on the temperature. Suppose if the water is coming out of a boil or if I
want to measure or if you are discharging that water somewhere, so that type of
measurement is also possible in, by the, by means of weir, because the, if the
temperature increases it hardly matters to the weir or the flow coefficient also does
not change much.
You see, the flow rate, excuse me, the flow rate over a weir is a function of the weir
geometry and the weir head. You will see that basically there are two types of weirs
used. One is rectangular weir and the V notch weir. Usually the rectangular weir is
used for the large flow measurements and the V notch weir is used for low flow
measurements and however, we will find that the V notch weir is actually more
accurate than the, more accurate than the rectangular weir. In this particular lesson, so
we will discuss or we will derive the flow velocity as well as upon volume flow rate
for the, in the case of rectangular weir, but we will give the expression for the volume
flow rate, whenever we are using the V notch weir, anyway. So, it depends on the
weir geometry, right, geometry and weir head.
4
What is wire head let us look at. The weir head is defined as the vertical distance
between the weir crest and the liquid surface in the undisturbed region of the upstream
flow. So, these terms we will use repeatedly, the undisturbed region of upstream flow.
So it looks like this, we will discuss.
We will see here, if I take a blank page you will see the weir, actually in that is a
restriction like, see it is a restriction like this. We will give the figure, so water is
flowing through this side. There is a channel, right, so water is flowing from this side,
right? So, if it is closed on this side suppose, so liquid will flow over this one, right,
this. So, this is called the crest of the weir, right? If you look at, it is, I mean if I take a
side view, it will look like this, sharp edge. So, water is falling over this liquid, right?
So, this is the ….. In fact actually, I mean you, you will see that this will be like this
one.
5
That means I have a vertical side view, so the water is falling, so falling like this. So,
this is our undisturbed region of the upstream, right? So, this will be our H, which will
be shown in the next figure, right? So, here if I take the velocity V 1 and at the crest if
I take the velocity V 2, then we can easily write the Bernoulli’s equations, right? Our
goal is to measure this height, H by any means. So, what are the different means that I
will show you later on. So, by measuring H, I can calibrate this instrument in terms of
flow. That means H can be calibrated in terms of flow and mostly it is indicated type
of instruments or monitoring type of instruments. It is hardly used for transmission
purpose, right?
6
So, that is I am saying, the weir head is defined as the vertical distance between the
weir crest and the liquid surface and the liquid surface in the undisturbed region of the
upstream flow, right?
The schematic of a weir is shown in Figure 1. The rectangular weir is shown in Figure
2 and flow of liquid over the sharp edge of weir is shown if Figure 3. So, in
subsequent three slides, we will find that, we will show you the weir geometry and
7
rectangular weir and flow of liquid over the sharp edge of the weir. So, these are all in
Figure, Figure 1, 2 and 3, respectively.
Now, you will see that it is the schematic of a weir, as I have shown you. So, this is
the duct and assuming that this is a river or a canal, so the liquid is, if I take a different
pen, if it is flowing in these directions, right and flowing out in these directions. So,
restrictions we have put like this, right? So, in a college dictionary you will find the
meaning of the word weir is basically a dam, but it is, actually it looks like a dam, I
mean restrictions, but so there is, as it happened in the case of dam there is a level of
difference in the upstream and the downstream, the similar thing is here also, right?
Let us look at very carefully. You see this is our crest of the weir, right? Liquid is
coming here and falling down, right? Now, if I put a float on this side which can
move only in the vertical direction, then what will happen? This, according to the, if
the flow velocity is high, so this height will go up, right? This H, if you look at H,
because liquid is, suppose the liquid is in this, up to this level, right, so from the crest
to the undisturbed water level that is the H or head of the weir. I repeat that from the
crest that means the sharp edge of the weir to the level which is undisturbed water
level is the capital H that is the head of the weir. Actually, we measure H, right by
some means. So, this is the schematic of weir.
8
Now, you see, this is a rectangular weir. So, this is, we will discuss later on what is
small h and what is dh. So, this is our crest of the weir, right and this is the water
level. Not necessarily it will go to the top, right? Water level may go up to this
position. In fact, this should be not up to this, this should be, should be, it should end
here, right? So, this is the water level, this is h, right? So, from this crest to the
upstream, undisturbed upstream, water flow level that is the capital H, not this part,
right and for derivations, later on we will see we take a small section dh at a height of
h, right, from the top of the undisturbed flow level, undisturbed flow level upstream
that we have taken. So, at a distance h from the undisturbed upstream flow level we
take a section over width d. This will be needed for, later on to find the total volume
flow rate or volumetric flow rate in the rectangular weir, right?
Now, let us look at, let us look at how the weir, how the water is flowing on this weir,
where H is the head on the weir, as we told and L is the length of the weir. It is
basically length of the crest of the weir and small h is the distance below the free
surface of water where V 2 exists. That will be shown in the, more clearly in the next,
next slide, right? You see here, yes it is clear now.
9
You see, this is the upstream, undisturbed region of the upstream flow, right? This is
the water level, this is the undisturbed region of the upstream flow. If I take a pen, this
is the undisturbed region of the upstream flow, so this is, everywhere this is same,
because water is, I mean coming like this one, so this level, so it is undisturbed, so
after that you see that water height start to fall, right? Now, in the previous slide if
you see, what we have seen, here you see what will happen? Let us go back, anyway,
see this is the distance h. In the previous slide you see here, h is a, we have taken a
section of, small section dh, of height dh at a height from the undisturbed region of
the upstream flow of h.
That means if you look at, so this our h, this is from the upstream. This is from the
undisturbed region, right? Fine. Now see, what will happen? So, this is h and the
velocity above the crest is V 2 and velocity upstream is V 1, clear? If it is there, so I
can immediately, so this sharp edge is very important, we will discuss later on,
because if it is not sharp the calibration constant will be modified. So, we can, usually
after certain months we can just change the weir keeping all other dimensions
constant, so that the, our flow coefficients will remain same as before. Let us look at
weir.
10
Now, if I apply the Bernoulli’s equation at the undisturbed region of upstream flow
and at the crest that means the sharp edge of weir, we can immediately write H plus V
1 square by 2 g capital H minus h plus V 2 square by 2 g is not it, because the height
is, if I take from the datum level it is H minus small h, right?
If I go back, you see H minus small h. At this position, my V 2 exists and H is

existing. In that this case we are taking this one as the datum level, is not it? You see,
11
this is we are taking as datum level. So, this is H height plus V 1 square by 2 g, so
add. This one is V 2 square by 2 g plus if I take this one as a datum level, so it is H
minus h where V 2 exists, is not it?
So, if it is there, so I can write these equations. You can, interestingly you can see
here that all are dimensions of height, because if this is in meter, so this will be in
meter square per second square. So, g is meter per second square, so obviously it will
be in meter, clear? So, no problem in that side, so dimensionally it is correct, right?
12
Where, V 1 is the upstream flow of the liquid and V 2 is the flow at the crest of the
weir that we have told one, once we have shown in the diagram. So, obviously if I
manipulate the equations, so the equation number 1, so it will come as a, come up as
V 2 equal to, because we say H and h will cancel out, V 2 root 2 g h plus V 1 square
by 2 g, right? Now, in most cases we will find that the V 2, the speed at the crest will
be always very, very large that we have seen. Those who have visualized near the
dam we will find that whenever in a lock gate, especially we will find that the, when
the water is flowing just over the lock gate, the speed of the water is very high
compared to the, I mean calm upstream flow, right? So, in that cases, so I can
obviously tell that the V 2 will be much higher than the V 1. In that case what will
happen you see, we can simplify this expression like this.
If V 1 is small compared to the V 2, then we can write V 2 equal to root over 2 g h,

right? This is equation number 2, right? However, the ideal flow rate, if I want to now
find the ideal flow rate, because this is the velocity at, because it is not very easy to
measure the H, right, because this is the, we are getting the flow at the crest, so it is,
please remember another thing whenever it crosses the, if I take a blank page it is easy
to understand. You see here, what will happen?
13
Whenever the liquid, I have a crest here, so if I take some other pen, so you see here,
so here the velocity is V 1. Some liquid is falling down here is V 2 at the crest, right?
So, here you see the, again the velocity will reduce. So, at this position only the
velocity will be high. This velocity will not be maintained. Once it falls down, the
velocity will also be reduced in this region. It is already less than V 2. Here also this
velocity will be less than V 2, right? However, the ideal flow rate over the weir is
obtained by integrating the quantity V 2 into dA. What is dA? over the area A of the
flow plane just above the crest of the weir? What is dA, let us look at. What is dA?
You see, dA is, actually we are finding dh multiplied by L that is the dA. So, dh
multiplied by L will be the dA. So, we will integrate over the height, h. So, we will
get the total flow of velocity.
14
If I come down here, now you see, so theoretically what we can write that if I go
back, so however the ideal flow rate over the weir is obtained by integrating the
quantity V 2 into dA, velocity at the crest multiplied by the dA, dA is the small h
section at the height h from the top, small h of width dh, so dh into L will be the dA,
over the area A of the flow plane just above the crest of the weir, right?
15
So, we can write Q theoretical that means theoretical volume flow rate equal to V 2
into dA. So it is meter, meter cube per second, because this velocity is meter per
second, it is in meter square. So if I, that is if I convert, so it will be 2 g h, right, root
over 2 g h, already we have seen, multiplied by, what is dA? L dA that I told you
earlier. So, this will be equal to, integrated over instead of A will be from, since it is
all converted, L is constant, only H is the variable, so I converted this integration
values from zero to H. So, if I do it, so I will get the expressions equal to 2 by 3 under
the square root 2 g into L H to the power 3 by 2. Please note, it is H to the power 3 by
2, this is equation number 3.
Now, however this is, as it is happened in all the flowmeters, so you see that every
time we have to multiply by a flow coefficient. Sometimes we call flow coefficient,
sometimes we call discharge coefficients. So, it is very, sometimes it is very difficult
to measure actually, to find theoretically the value of all the discharge coefficients.
The best thing is to find experimentally with some standard flowmeters, right? So,
however as it happened in any other flowmeter, the actual flow rate is less than the
ideal flow rate due to the vertical contraction from the top and the friction loss and the
presence of liquid flow that is not horizontal. Therefore, the actual flow rate is given
by, so vertical contraction this is the slow contraction from the top. If you look at, you
see here, so there is a slow contraction from the top, right?
16
I have a weir, so there is a, it is not that suddenly it is coming, so it is a slow
contraction from the top and we are assuming that all the flow velocities is like this
one, right? If I have a weir, we are assuming that all velocities is in this direction. This
is not necessarily true. Velocity will be in this direction, some velocity will be in these
directions also, right?
So, we have to take care while we are making the, if you, if we are interested to find
the actual flow rate, right, for that reasons we are telling, you see, however the actual
flow rate is less than the ideal flow rate due to the vertical contraction from the top
and the friction loss and the presence of liquid flow that is not horizontal. Therefore,
the actual flow rate is given by Q actual equal to C D Q theoretical 2 by 3 square root
of 2 g C D LH 3 by 2.
17
This is equation number 4, where C D is the discharge coefficients of the rectangular

weir and it lies between 0.62 to 0.75 and the weir must be sharp for this coefficient to
be valid over a long period of time before the flowmeter is recalibrated, right? This I
told you earlier also. That means this H should be very sharp, otherwise what will
happen? You see that this, the calibration constant or discharge coefficient will be
modified. That happens in all the instruments. So, most of the instruments like, you
see in the case of Venturi, when we discussed the Venturi meter, in the case of
Venturi meter this type of calibrations is not necessary. But, we have seen in the case
of orifice meters this, the H is not, is not 90 degree rectangular orifice we will find
that the discharge coefficients will be modified.
So, the best way, during the routine maintenance you change the orifice. If you know
the thickness and everything is same, dimension is same, the orifice hole diameter is
same, you do not have to recalibrate, right? Similarly in the case of discharge
coefficients or in the case of weir also, so during the routine maintenance, suppose
after using 1 month or so or 2 months, depending on the how much …., how much is
the corrosion in water and all those things, we can just change this weir. Weir, it is not
very expensive equipment. That is also you must mention, I mean this is to be, I mean
this is to be explained, because it is not expensive compared to the, I mean if you look
18
at the other open flowmeter it is not very expensive. So, in that sense it is, replacing a
weir, it is not very expensive affair, right?
So, when the flow rate is low, now we talked about the rectangular weir. But, there is
another type of weir, as I told you, this is called V notch weir. When the flow rate is
low, a triangular or V notch weir is more suitable for measurement of flow.
This type of weir is more accurate than the rectangular weir, right? It is for the low
velocity flow rate, but it is more accurate, right? A V notch weir, as you can see, is
shown in figure 4, right? You see here also, so this is the length of the V notch weir.
This is also restrictions. Instead of rectangular, this type of weir we will put it and so
this is, it is slowly increasing. This is the included angle theta. As before, we have
taken a section dh from the top of, from top of the undisturbed flow region, h and H is
the height head of the weir. As the liquid flow increases, this head will increase, will
go up. H will increase, please note because in the previous case also we have seen that
this H has increased, right?
19
So, total, excuse me, the ideal flow rate for V notch weir is given by Q theoretical
equal to 8 by 15 under the square root of 2 g tan theta by 2 H to the power 5 by 2
4 L by 15 H. If I replace tan theta by 2 under the square root root 2 g H to the power 5
by 2, this is equation number 5, where L is the width of the weir or length of the weir
or width of the weir; whatever you say, it is same thing, right? So, it looks like this
you see. We have seen that, anyway. You see here.
20
So, this is included angle theta, so tan theta will be how much? Tan theta will be L, L
by 2. This total is L, so it is L by 2 divided by H, if I take the tan theta.
So, exactly that thing I did in the next slide. So, tan theta by 2 is replaced by, I mean L
by 2 divided by H. So, it will get 4 by 15 H root over 2 g H 2, where L is the width of
the weir. So, what is the variable here? In the Q theoretical also you see H. As the
liquid flow, volume flow rate increases, H will increase, right? The height or head of
the weir is increased. This H is to be measured by some means or the other, right?
All other things are constant, L are constant, g is also constant, so it is quite nonlinear,
as you can see. So, this H is to be measured and this height is calibrated in terms of
the volume flow rate.
21
However, the actual flow rate, as it happens in the case of the rectangular weir also,
we will find the actual flow rate will differ from the theoretical flow rate. So, it is to
be multiplied by discharge coefficient C D 4 L by 15 H, so under the square root 2 g
H to the power 5 by 2, equation number 6, where C D is the discharge coefficient that
lies between .58 to .6, because this is experimentally found. It is very difficult to find
the theoretical value of C D. So, what the people did, they have taken some standard
and compared and found the C D value.
The equations 4 and 6 indicate that the flow rate of the liquid depends on the height or
the head capital H, right? So, that capital H is to be measured. What is the capital H?
That is the upstream height of the, upstream flow from the crest of the weir, right?
What is that? Very clearly, let us quickly look at. H, I have to measure, some way or
the other.
22
What is that? That means I have, a liquid is flowing, right? So, I have a liquid flowing
like this one, undisturbed. So, this H is to be measured. As the flow rate increases, this
H will increase. If the flow rate decreases this H will decrease, right, clear? Even
though it is nonlinear, I mean, relation but if the flow rate decreases, this H will
decrease, if the flow rate increases this H will increase, right? So, I will measure the H
and calibrate H in terms of the flow rate, fine, right?
23
Now, as I told you earlier also, you see, the weir head is measured by a float activated
displacement transducer and it is placed in the upstream side and it is usually used as
an indicating instrument. It is not for the, I mean transmitting instrument. People tried
even though, I mean transmitting, but it is not for the, I mean actually it is indicating
or monitoring instruments.
Usually what they do, you see, in the upstream sides, I mean they are making a,
sometime they are making a float like this one, right? So, one like this one, so there is
a, it is put on a cylinder, right, a transparent glass. This is actually spheres which act
as a float. So, what will happen? So, this is put on the upstream side, right? What will
happen? If the water level increases, so this float will go up. This float will go up like
this one, right? This float will go up like this one and if you have a graduated scale
here on the glass itself, so I can measure this height and calibrate the height in terms
of the flow.
So, these are basically, for balancing if you use single that …. problem of balancing,
so they use three. Actually if we take a top view three spheres like this one, 120
degree apart, three spheres made of plastics and put. If I take top view you cannot see.
So, I take like this one, right? If I take a side view, it will be like this one; three
24
spheres, so one float. So, as the liquid, I mean if the flow increases this will go up. So,
this is calibrated in terms there, so this is all about the weir.
Now weir, as I told you, there is no other alternative for irrigation purposes for
making the and it is quite robust. It is not that accurate obviously, but it is to get an
estimate of the flow of the fluids especially in the irrigation purposes, right?
Now, our next transducer which we will discuss is the, basically we will discuss the
turbine flowmeter which is direct electrical leading flowmeter. But, to know the
turbine flowmeter we must first look at the variable reluctance tachogenerator and
what is the, some basic terminology of the reluctance and magnetic force we must tell
in brief and its units in SI, then we will go for turbine flowmeter. First we will look at
the variable reluctance tachogenerator, because exactly that principle is used to make
the flowmeter.
So, let us look at magneto motive force. Magneto motive force is a force that causes
flux to be established and it is analogous to the electromotive force for electric
circuits, right? The unit of mmf, we write, magneto motive force we write mmf in SI
units is the ampere. It is actually ampere turns, but it is actually telling about the
single turn, but it refers to a coil of one turn, right?
25
The opposition to the establishment of magnetic flux is called the reluctance. Now
reluctance, whenever reluctance increases your flux decreases. If the flux increases
reluctance decreases, right? So, the reluctance is defined as R equal to mmf, magneto
motive force divided by the flux phi, right? Rearranging, I can write mmf equal to R
multiplied by phi, right, where mmf is in ampere turns, if the number of turns is more
than 1 and phi is the flux, is in Weber.
26
Now, principle of operations of this variable reluctance tachogenerator let us look at.
Now, if the time varying flux phi linked by a single turn of coil, all of you know now
the basic principle, then the back emf developed in the coil can be expressed as e
equal to minus d phi by dt, right?
Here you can see the variable reluctance tachogenerator shown in figure 5. You see
this, we are showing a wheel with teeth or blade, whatever you say. There are several
teeth you can see or blade, several teeth here and this wheel is made of, this teeth
actually is made of ferromagnetic material. Please note, this is made of ferromagnetic
material and a coil is wound on a permanent magnet and this pole piece is made up of
soft iron. So, what it that? This coil is wound on a permanent magnet. This pole piece
is made of soft iron and this teeth is made of ferromagnetic material.
Now you see, what will happen? There is, a flux will be developed, is not it? How it
looks? You see, the flux will just go like this one. Sorry, if I take a new page, I think
we must go back, so if I take some other colour, I think let me choose this colour, so
the flux will look like this one. So, it will go like this, go like this, go like this, go like
this, in this …., is not it? Then, what will happen you see? You see, what will happen
that when the, when these teeth or blade comes to the, very close, actually this gap is
very small. Just for, I mean sake of understanding we have drawn in such a, I mean
27
larger gap. Theoretically this, I mean actually, in the actual, in the variable reluctance
tachogenerator this gap is very small.
Then, what will happen? In this case whenever it comes to, very close to this,
reluctance will be small, flux will be high and in between the teeth. Suppose this is,
this wheel is rotating in a direction, in a, in a, in an anticlockwise directions with an
angular velocity of omega, you see this one, you see this is the angular velocity.
Wheel is rotating with an angular velocity of omega, right? Now, what this angle is
theta. What will happen? You see if this rotates, if the wheel rotates like this one,
right, then what will happen?
If it rotates, so whenever this wheel, I mean in between the two teeth, suppose the
teeth has gone down this direction, this teeth has moved to this, now this pole piece
faces the, the gap in between the two teeth. Then, what will happen? The reluctance
will increase. So, there is a change of magnetic flux, right, right? So, what will
happen? This output voltage, there will be variable output voltage. Time varying
output voltage we will get if it rotates with the speed of omega, right? So, using this
principle we will find, we have made the turbine flowmeter. So here, again I will
repeat this is made up of permanent magnet, right and this pole piece is made up of
soft iron and this teeth is made of the ferromagnetic material.
28
Now, the teeth of the wheel moves in close proximity to the pole piece. It is, I mean
very close to each other as I told you, because for the sake of understanding we have
drawn that much of gap, but it is, actually there is no gap.
This gap is very, very small, right? This gap is ultimately, there is no gap at all, so it
should come very close to this, right? This will come very close to this. That means if
29
I take it, so this will come very close to this, right? That you please note. So, it should
come very close to this like this one anyway.
So, the teeth of the wheel moves in close proximity to the pole piece, as I told you.
Therefore, the flux linked by the coil changes with time and the voltage is developed
across the coil. The total flux phi subscript capital T linked by a coil of m turn is
given by phi T equal to m phi m mmf into divided by R. R is the reluctance, the
number of turns of the coil is m. So, it will given by m into mmf by R.
30
Moreover, we know that when the reluctance is minimum the flux is maximum and
vice versa; that I have explained to you already. The variation of the flux phi T with
the angular position theta of the wheel will be expressed as, this will be expressed as,
since it is a sinusoidal variation, we can express as phi T theta equal to alpha plus beta
cos n theta. What is this abbreviation let as look at, where alpha is the mean flux and
beta is the amplitude of the time varying flux and n is the number of teeth of the
wheel. There might be 4 wheel, there might be 4 teeth, there might be 8 blade. So,
here in this case we assume that the number of, number of teeth is n. So, we can write
phi T theta is time varying, theta total flux with respect to the angular position of the
wheel is equal to alpha plus beta cos of n theta, clear?
31
(Refer Side Time: 40:07)
So, e the output voltage which we will receive at the coil is given by minus d phi T by
dt. Total flux equal to d phi by minus d phi T by d theta into d theta by dt; that always
we can write, is not it? So, again we know that the d phi if I differentiate by previous
expressions that means if I differentiate this expression, so I will get, if I differentiate
this expression I will get d phi T by d theta, minus alpha will cancel, alpha, d alpha by
d theta is zero, so minus beta plus into n sin of n theta.
Now, theta is the, omega is the circular frequency and theta is an angle at instant of
time. So, if we multiply by, omega by t, I will get theta. Obviously, theta should be in
radiance. So, d theta by dt equal to omega. Quite obviously, if I differentiate this with
respect to t, I will get omega. So, I will substitute all these in this equation, in this
expression, right? In this expression, I will substitute this value of d theta d phi T by d
theta and d theta by dt I will substitute in this expression. So, I will get, where omega
I should say is the rotational velocity of the wheel. Now e equal to beta n omega sin n
omega t.
What does it mean? It means, you see that the output voltage from the variable
reluctance tachogenerator is a function of the angular frequency, amplitude of this. It
is sinusoidal signal, the output voltage. If it rotates at a constant velocity of circular
frequency omega that means the wheels rotate at the constant velocity of circular
32
frequency omega, then you will find that the output voltage is expressed as beta n
omega. So, this is very, very important expression. So, I should write, so what is
amplitude? Amplitude is beta n omega of this output signal and with the frequency of
n omega. That means I can say the amplitude of this signal that means I will get an
output voltage if it rotates at a constant angular velocity beta n omega and a
frequency. This is amplitude that means A and the frequency is equal to n omega by 2
pi, is not it, I mean a linear frequency. So, output voltage depends on both.
So, later on we will see that, we will little, make some little manipulations of the or
little signal processing by which I want to make that output voltage will be only
proportional to the frequency of rotation; that the frequency increases our output
voltage also increases. But here it is not, because if the frequency increases amplitude
also increases. You have your, the, I mean flow velocity, if the, if the number of
rotation increases, number of the wheel rotating that is increases, this omega increases
your output voltage also increases. But that does not depend directly only on the
omega. Your amplitude increases your frequency also changes. So, two are changing
simultaneously. That is not very desirable property, right?
Now, in turbine meter exactly this principle is utilized, right? This principle is utilized
in turbine meter.
33
You see, what will happen? It has direct electrical output. It is a great advantage. In
the, the other flowmeters we have seen that we have to make, lot of manipulation of
the signal processing we have to do, like differential pressure flowmeters, rotameters,
we have seen that there is a lot of, pitot tubes or we need a different lot of signal
conditioning. Sometimes it is not signal conditioning circuit, in that sense it is a
electrical. Sometimes you will find this pneumatic signal is coming. You have to
convert with DP transmitter and so many things are necessary which is not necessary
in the case of this type of direct electrical reading meter.
No, it is not much in use, because of some disadvantages anyway. It has a direct
electrical output. The flowmeter consists of a wheel with a multiple blade. Now, I am
telling here blade, because the shape is like a blade in the case of turbine meter. But
the blade, whatever I am talking now it is similar, exactly similar to the teeth in our
wheel of the just discussed variable reluctance tachogenerator, right? So, there is no
distinction between the blade and the teeth, please note it and it is installed inside the
pipe in which clean fluid is flowing. This clean fluid, the term clean is very important,
you see it should be clean. Liquid should be clean, otherwise there is a problem. That
we will discuss later on.
So, the flowmeter consists of a wheel with the multiple blades and it is installed inside
the pipe in which the clean fluid is flowing. A schematic of turbine flowmeter is
shown in Figure 6. I will show you the turbine flowmeter.
34
You see, this is our turbine flowmeter. So, we have a magnetic pick up. What is that
magnetic pick up? This is a permanent magnet on which the coil is wound. In fact,
there should be a, two terminal should come out. So, two terminal will come out,
right? These two terminals we will measure the voltage E, right?
What is this inside the magnetic pick up? This is the magnet. These blades are, are
made up of ferromagnetic materials. We have shown the 4 blades, right, 4 blades. You
see, these are in angle, little angle, so the, when the flow is, when the liquid is
flowing, it is impinging on the blade. If it impinges on the blade, so this, this will start
to rotate. If it starts to rotate, then what will happen? If its angle is like this one, if it is
like this one, then what will happen?
If I take a camera like this one, you see here if the blade is slightly like this, likely it
happens in the case of, so liquid will flow over this one, right, in this direction. So,
what will happen? You see here in this direction, so it will start to rotate, right? So,
similar thing, So, this is made of ferromagnetic materials. I have a magnetic pick up
here. What is that magnetic pick up? It is made up of permanent magnet on which the
coil is wound with soft iron of pole pieces. Since it is a ferromagnetic material,
whenever it rotates, for each rotation or whenever this blade passes this permanent
35
magnet, I will get a pulse at the output or I will get a time varying signal at the output.
So, that will be calibrated in terms of the flow, right?
Let us look at the details of this one.
The turbine meter must be installed along the, along an axis parallel to the direction of
fluid flow in the pipe, right? The flow of the fluid past the wheel causes it to rotate at
rate proportional to the volume flow rate. It should be proportional to the volume flow
rate. The wheel or blade of the meter is show in Figure 7.
36
You see this little bit, this, each wheel causes an angle of alpha, right, angle of alpha
and you see here the wheel. Actually if I take from the, along the axis of the pipe, the
wheel will look like this one. This blade will look like, …. the thickness t and it has
an angle alpha. It is actually, it looks like this, so that it will have an angle like alpha,
like this one, right? So, this is angle alpha, right? So, the water is, water or liquid is
flowing like this one. It is hitting this blade, so it starts to rotate in this direction,
right?
37
Construction, you see the construction of a turbine flowmeter is made of, similar to
variable reluctance tachogenerator, no difference. The blades or teeth are made or
teeth are made of ferromagnetic material.
The magnetic pick up consists of a coil wound on a permanent magnet. A voltage

pulse is obtained at the output of the pick up, whenever a turbine teeth or blade passes
the pickup coil, as I told you. The flow rate can be determined by measuring the
pulses by a pulse counter. So, if I can measure the pulse count, I can measure the flow
rate, as simple as that. Moreover, I can find the total volume flow rate if I count the
number of pulses over a certain period of time. That is the great advantage of the
turbine flowmeter that means the total volume of liquid flown over a certain period of
time, ….
38
If the number of rotation of blades per second is f and the volume flow rate is Q, then
we can obviously write f equal to k into Q, so the Q can be determined by measuring
f. The total volume of liquid V T flown through the pipe over a given time T will be
expressed as V T integral zero to T Qdt.
The total number of rotation of the blade over the time T is expressed as N T. Now we
have, total number of count we have expressed T to zero f d t k Q dt into, so it is
39
coming kV T, right? So, Q dt, so we are getting kV T, right? The total count is
proportional to total volume flow of the liquid. Obviously, this we can measure by
some counter, so which will give me the total volume flow of the liquid, right?
The voltage induced in the magnetic pick up will be periodic in nature, whose
frequency is proportional to the angular velocity of blades, right? So, assuming the
drag force due to bearing and the viscous friction negligible, the rotor angular velocity
omega is proportional to the volume flow rate given by, obviously what will happen?
As the volume flow rate increases, your, the velocity increases, obviously the omega
also will increase. Omega depends on what? The omega depends on the velocity of
the fluid flowing through the pipe. So, in turn if I, velocity if I multiplied by the A, the
area of the cross section of the pipe, so obviously, I mean, I mean minus the area
taken by the hub and all these blades and all those things, I will get the flow velocity,
I mean the angular velocity.
So, I can say that the angular velocity will be proportional, angular velocity of the
turbine flowmeter is proportional to the volume flow rate. So, that is actually we have
written.
40
So, the velocity of the fluid is given by u equal to Q by A, right? A is the area of the
cross section. Now you see, where A is the area of the cross section of the pipe minus
area of the cross section of hub and area of the blades. So, that we have written pi D
square by 4 minus pi d square by 4 n h minus d by 2 into t. So, this will be very clear,
you see here. As you can see from the, I mean in the, in the, from the constructions
you will see it is very much clear.
41
You see here, so this is h. h will not be exactly d by 2. Please note, h is not exactly d
by 2, because there is a gap. There should some gap to, because you have to give
some space to rotate, right?
So, this is our expression. We have shown pi d square by 4. The total area of cross
section of the pipe minus the hub, the area taken by the hub minus the area taken by
the blades, right? So, all these things if I, so where n is the number of blades and it is
and t is their average thickness.
42
So, moreover we have seen that omega h by u equal to tan alpha by in fact, omega is
equal to tan alpha, is not it? Exactly, so omega h is the velocity of the blade tip,
because that is the circular cross section, circular frequency multiplied by h is the
velocity of the blade, blade tip and is because h is the, actually if you look at, h is the,
h is the height of the blade from its central position, is not it? See, if I multiply by the
omega, so it will give the speed of the tip of the blades, right, per second. It is the
velocity of the blade tip perpendicular to the direction of flow.
Obviously that is very much clear, because the liquid is flowing in this direction, so
the blade is rotating in this direction. So, if the direction of the flow is this, the blade
will rotate in this direction. So, I can write that alpha is the inlet angle at the tip.
So, k 1 equal to omega by Q equal to tan alpha by Ah, right? So, the output voltage
from the magnetic pick up will be given by beta nk 1 Q sin nk 1 Q by into t, equation
number 7.
43
Beta is the amplitude of angular variations of magnetic flux and n is the number of
blades and Q is the volume flow rate. The equation 7 indicates that the output of the
magnetic pick up is a sinusoidal signal of amplitude beta nQ and the frequency is nk 1
Q by 2 pi, where, which can be written M f into Q, where M f is the static sensitivity
or the meter factor of the equipment, right?
44
Now, you see, one thing you must note that if I look at here, if I look at in this
expression, you will find this is a, again this is the expression what we are getting.
You see that it depends on the frequency of number of, sorry, it is a, beta depends on
what are the factor? Q. As the Q increases, this amplitude also increases. Q is the
volume flow rate increases the amplitude increase. So, simply what they do actually?
They take a, if I take, I have a turbine flowmeter, if I take a, sorry, so they take a
meter then they pass through an integrator, our turbine meter. Then the integrator,
then a Schmitt trigger. What will happen? You see, once you integrate, I will get a
signal which depends, I will get a constant amplitude signal. I will get a signal which
looks like this. You see here I will get a signal. This output of this Schmitt trigger is a
constant amplitude signal, because Schmitt trigger is a pulse shaper, as we know. So, I
will get a signal that depends only on the frequency. As the, that means that means as
the flow velocity Q changes here both the amplitude of the signal also will change.
Here it will not, because the …. of this amplitude will cancel out, flow will cancel out.
The Q will cancel out.
So, at the output of the Schmitt trigger, I will get now only the frequency. So, the
constant amplitude frequency I will get. So, that frequency can be calibrated in terms
45
of the volume flow rate. Let us look at, let us go back again. So, that is actually they
do in the case of meter.
So, provided that the turbine wheel is mounted in low friction bearings, the
measurements accuracy can be as high as plus minus 0.1%. It is less rugged and
reliable than the restriction type differential pressure flowmeter.
46
The blades and the bearings can be damaged if solid particles are present in the liquid.
This is the most important point; liquid should be very clean, right? There is another
problem. It is more expensive and it also imposes the permanent pressure loss on the
measurement system. Even though it is not as high as, because we are, I mean putting
some restrictions in terms of the blades and all these things, so extra pumping is here
also, so it is not that it is totally open channel like, I mean it is not totally the
permanent pressure loss free flowmeter.
So, there is a, some certain amount of permanent pressure loss in this type of
flowmeter. So, the typical range is at usually 10% to 100% of full scale output reading
below the 10% of full scale reading the bearing friction makes …
Turbine meters are particularly prone to large errors when there is any significant
phase change in the fluid being measured. Turbine meters have a similar cost and the
market share to positive displacement meter and the former are smaller and lighter
than the latter, right? Now, this is the some advantage, but as I told you in the turbine
flowmeter, the basic principle is that I am getting an output. The flow velocities which
I am getting you see, it is I am getting a sinusoidal signal at the output of the turbine
flowmeters. Since the basic means is similar as the tachogenerator, so as the flow
47
velocity changes, amplitude changes as well as the frequency of the signal also
changes, frequency of the sinusoidal signal.
Now, if I pass this signal through an integrator, then what will happen? So, this will
cancel out. So, I can explain this again. Let us go back.
You see here, yes, here you see here, what will happen? So, if I integrate this signal, if
I integrate this signal then what will happen, you see? So, I will get beta nk 1. If I take
the different colour, please note so beta nk 1 Q cos of nk 1 Q into t divided by this nk
1 Q, so this nk 1 Q nk 1 Q will cancel out. So, after the integrations I will get, e equal
to beta cos of nk 1 Q into t, is not it? You see here, right? I will write it clearly, e
equal to beta cos of nk 1 Q into t. So, what does it mean? It means that you see that
beta is a constant. It does not depend on the flow velocity. So, amplitude is directly
proportional to the frequency.
Now, what will happen? It, I will pass that signal. After integration I will pass that
signal through a Schmitt trigger. I will get a constant amplitude pulses, but the
frequency will change as the flow velocity changes. So, if I measure the frequency,
that frequency will be calibrated in terms of the flow, right? This is the advantage of
this type of meter. So, with this I come to the end of the lesson 14.
48
Prof. A. Barua
Lecture - 15
Flowmeter - IV
This is lesson 15 of Industrial Instrumentation, we will continue with the flowmeter. So,
this lesson 15 and we will basically discuss the Flowmeter IV, we have given the name
like that.
So, in this lesson we will cover, the contents of the lesson - electromagnetic flowmeter,
ultrasonic flowmeter. These two basic flowmeters we will discuss in this particular
lesson. Now, let us speak about what the viewer will know at the end of this lesson.
1
At the end of the lesson, the viewer will know principle of operation of electromagnetic
flowmeter, type of pipe used, because various types of pipes you have to use, then
magnetic field excitation for the flowmeters which will cover, because electromagnetic
flowmeters in needs of magnetic field excitation, then we have its advantages and
disadvantages. Principle of operations of ultrasonic flowmeter, this also … We will
discuss the ultrasonic flowmeters in details. So, it will come one by one.
2
Now, let us first start with electromagnetic flowmeter. Electromagnetic flowmeter is
limited to the measuring the volume flow rate of electrically conductive fluids. The
conductivity of the fluids should be greater than 10 micro mho per centimeter. This is a
typical restriction of the electromagnetic flowmeters and if it is less than this, so this
flowmeter will not work. However, there is no higher limit on the, on the greater
conductivity of the flowmeter. That you will see that in some situations the, the pipe as
well as the electrode installation will be different if the conducting liquid is very, is very
good conductivity of the fluid which is flowing through the pipe.
The schematic of an electromagnetic flowmeter is shown in Figure 1. Let us look at. The
instrument consists of a cylindrical tube made of stainless steel fitted with an insulating
liner. So, there should be insulating liner, because if it is a metal tube, so the, it will have
a short circuit pulse. So, for that reasons what will have that we should have insulating
liner.
Now, typical you see this is the electromagnetic flowmeter. It looks like this. You see, if I
have a camera on this side, camera on this, it looks like this, right? So, the liquid is
flowing in through it and then it is coming out; flowing in, coming out. So, there is a
3
magnetic field. So, if there is a coil on top and a coil at the bottom, so it will produce
magnetic field which acts perpendicular to the flow of direction, direction of the flow of
the fluid and we will put two sensors on both the sides - one sensor here or electrode,
another electrode here. So, you will see that due to Faradays law of electromagnetic
inductions, some voltage will be developed across this electrode. That voltage will be
carried by it in terms of flow.
Now, the flow of the fluid, this position of the electrodes, if the line connecting the two
electrodes and the direction of the magnetic fields are mutually perpendicular to each
other, right? You will see here fluid flowing is through the pipe. There is insulating liner,
so we are producing magnetic field with proper coil and all those things and two sensors
are there which are installed diametrically opposite side. One sensor is here, another
sensor the other side. I am sorry, one sensor is in this side, other sensor on the other side,
you cannot see that and we are getting the output voltage. This output voltage is to be
carried by it in terms of the flow.
The typical lining materials are neoprene, polyterafluroethane ethylene, which is

abbreviated as PTFE and polyurethane. A magnetic field that is perpendicular to the
4
direction of flow of the liquid is created in the tube by installing energized field coils in
diametrically opposite side of the pipe. This already we said. I mean two energized field
coils are there. What type of I mean electric, what type of energy we will apply that we
will discuss later on, but to produce magnetic field we must have some two coils which
are diametrically opposite side. The voltage induced in the fluid is measured by two
electrodes inserted into the opposite sides of the tube. We have shown just right now,
right? Two electrodes are there. These are also installed in diametrically opposite sides,
but all are mutually perpendicular to the, each other.
The ends of the electrode are flush with the inner surface of the fluid carrying pipe. This
is important for the magnetic flowmeter, specially if the conductivity of the fluid is not
very high. If it is very highly conductive, then it is not necessary to put it, to be flushed
with the pipe, right? The material used for making the electrodes are stainless steel,
platinum iridium alloys, titanium and tantalum. You can see thus electromagnetic
flowmeter, though it has many advantages that we will discuss what are the advantages
later on, but it is not very cheap instrument. It is quite expensive instrument in the sense
that the type of electrodes you will use, the pipe you use and the pipe is not expensive,
but inside the pipe you should have a liner. So, those are quite expensive. That cause, that
5
makes the instrument very expensive. Moreover individual calibration of the instrument
is necessary, very precise calibration.
To make this, if you add all these things, you will find that the initial installation of this
tube is quite expensive compared to the other flowmeters like differential pressure
flowmeters and all those things. But, in some situations we cannot afford to have a
differential permanent pressure drop. In that type of situation, this magnetic flowmeter is
used. The material used for making the electrodes are stainless steel, platinum iridium
alloys, titanium and tantalum. In accordance with the Faraday’s law of electromagnetic
induction, the voltage e, induced across the length L of the flowing fluid at a velocity V
in a magnetic field of flux density B is given by e is equal to VLB volts, right?
So, let us look at, what is this. This is our pipe, so you see what is capital L? Capital L,
basically if you look at here capital L will be the, if I take some pen, so capital L will be
the diameter of the pipe, is not it, inner diameter of the pipe, capital L. V is the flowing,
velocity of the flowing fluid and B is the magnetic field which is magnetic field intensity
or flux density which is working on the conductivity fluid, right? So, we you see the, in
accordance with the Faraday’s law of electromagnetic induction the voltage e induced
6
across the length L of the flowing fluid flowing at a velocity V in a magnetic field of flux
density B is given by BLV, where you can find this is in volts, obviously. See, this is in
volts, right?
This output voltage is in volts provided that where B is the flux density is in weber per
meter square, which is in Tesla, L is a distance between the electrodes or inner diameter
of the pipe which is in meter, V the velocity of the fluid that is in meter per second, right?
See, if you use all these notations, we will find that the velocity of the, I mean the output
voltage of the fluid will be in volts and you can see here if BA and L remain constant for
a particular liquid, obviously e will be directly proportional to the flow velocity. See, if
you now multiply that velocity with the area of cross section or the inner area of cross
section of the pipe, we will get the volumetric flow rate, is not it?
Now advantage; there are several advantages of this type of flowmeters.
7
Number 1, there is no obstruction of the fluid flow. Therefore, there is no pressure loss
associated with the measurement, right? This is very important. Then we have, it is
suitable for installation that can tolerate only a small pressure drop. In many situations
you see, this pressure drop cannot be allowed when the liquid is flowing in a slow
moving velocity and pressure across is very small. We will find it is, we cannot afford to
have this type of pressure drop, right? So, in that type of situations, this magnetic
flowmeter is quite popular.
The absence of any internal part is very attractive for measurements of velocity of
corrosive and dirty fluids. You will find that in the other type of flowmeters, whether it is
electric, whether it is differential pressure flowmeter, turbine flowmeters, everywhere we
have to install all the systems inside. In the case of turbine flowmeter, entire turbine is to
installed inside the pipe in the, though in the case of orifice meter, Venturimeter, the,
there is no such installation. But there is, orifice plate is there and also in the case of
Venturimeter there is a slight deductions of the pipe diameters which is actually the throat
of the pipe that also makes problem. So, the absence of any internal part is very attractive
for measurements of velocity of the corrosive and dirty fluids.
8
So, the operating principle is independent of fluid density and viscosities are very
important and the output voltage is proportional to the average velocity of the fluid; this
average velocity of the, this is very important point. Please note the voltage is
proportional to the average velocity of the fluid. The operating principle is independent of
fluid density and the viscosity, right? That what does it mean that is it independent? For
all the liquid, the output voltage will remain same if the velocity is same? It is not true.
Why? We will explain after some time. Got it, what I am saying?
9
That means I am saying that our equations, if you remember, is e equal to BLV. So, we
are saying that e is proportional to V. So, for all liquids, if the velocity is same will I get
the same value of the output voltage? Not true; you won’t get the, why? The reason is L
will remain fixed that is the inner diameter of the pipe, because we will install, this is a
basically distance between the pipe. It assumes like this one. You can assume the, like
this one. If I again take this example that a conductor is moving inside a, conductor of
length is moving inside a magnetic field, if I assume it is of, of length L, so that
conductor of length L is moving inside this pipe with a velocity V that we can assume.
Now, we apply the Faraday’s law of electromagnetic induction also by using that
principles; the conductor of length L which is the inner diameter of the pipe moving with
a velocity V on that the flux density B, clear? So, there is no difficulty of measurement of
the either laminar or turbulent flow. So, this is very important. Since it is independent of
viscosity and density, so obviously we can measure both laminar and turbulent flow.
10
It has accuracy of plus minus 1% of indicated flow. This is also quite high, I should say.
The flux density, now what I said sometime back that the, whether for the same velocity I
will get the, for the different liquids will I get the same value of the voltage? No, that is
not correct. Why, it will be clear now. You see, the flux density B is related to the field
intensity H by the relation B equal to mu H, right?
11
Mu is called the absolute permeability and is expressed in Henry per meter and the
absolute permeability of the another material can be expressed relative to the
permeability of the free space by mu equal to mu r into mu naught, where mu naught is
the permeability of the free space. Its value is 4 pi into 10 to the minus 7 Henry per meter
and it is a constant. The mu naught is a dimension, dimensionless quantity and it is called
the relative permeability and its value depends on the material of the fluid, right? So, as
the material of the fluid changes, value of the mu R will change. So, value of mu will
change. So, that will lead to the value of B will change, so the output voltage also will
change, right? So, for different liquids I will get different outputs depending on how
much the value of the mu for the particular liquid.
Now, the other forms of flowmeter, the electromagnetic flowmeter requires a minimum
length of straight pipe work upstream from the meter in order to guarantee the accuracy
of the measurement. We have seen that this type of restrictions we have especially in the
differential pressure flowmeter. That means we have 20 upstream pipe diameter. There
should be no obstruction, no pipe bendings, that type of things also here saying that like
other forms of flowmeter, electromagnetic flowmeter requires a minimum length of
straight pipe work upstream from the meter in order to guarantee the accuracy of the
12
measurement. No pipe bending or pipe fittings are usually allowed in 5 pipe diameter
upstream. So, it is quite less compared to some other flowmeters. We have seen that in
many flowmeters it is 25 diameter upstream. Please note that one thing is important, it is
measuring the average flow velocity, right?
Now, disadvantage is, one of the greatest disadvantages that the conductivity should have
some minimum value. If it goes below that magnetic flowmeter cannot work. So, this
excludes by this all the hydrocarbons. That means electromagnetic flowmeter is not
suitable for measurements of the fluid in any petroleum industry. There is some
restriction. We have restriction of the petroleum industry of using any electrical type of
system, because it is hazardous and I mean it is inflammable. But still, if you do not
consider that part you will find that electromagnetic flowmeter will not work for the
hydrocarbon, because hydrocarbons conductivity is much less prescribed for the
electromagnetic flowmeter. So, that I will discuss one by one.
The electromagnetic flowmeter requires a minimum conductivity of 10 micromho per

centimeter, right? Therefore, it is not suitable for measurements of flow of gases and
liquids, of liquid hydrocarbons. So, it is not suitable for measurements of liquid, any oil
13
industry or petrochemical industry, right? It is suitable for measurement of flow of
slurries provided the liquid phase has adequate conductivity. Adequate, liquid phase must
have adequate conductivity, then only the measurement is possible, otherwise it cannot
be. So, that restriction is there, micro mho per centimeter, it should be there, right?
Now, instrument is expensive, as I told you earlier, in terms of initial purchase cost,
because I told that the cost of the inner liner of the pipe, cost of the electrode, these are
causes of expensive and cost of calibration is also quite high. So, one of the reasons for
high cost of the need, is a need for careful calibration of each instrument individually
during manufacture as there is considerable variations in the properties of magnetic
materials used. You see, that is very difficult to get very, actually predict the B, I mean
the amount of value of B you will get. See, it will slightly vary depending on, even
though you are giving some coil a slight reorientations of the, slight misalignment of the
coil will give you different value of B, so your output voltage. So, each and every meter
should be calibrated on the site and separately it is to be, that is the one meter you are
calibrating you can say the other meters are of same calibrations value that is not correct,
right? So, these are the difficulty along with the initial purchase cost.
14
That means the, your pipe is quite expensive as well as the electrodes are also expensive.
So, keeping these three in mind and keeping these three points electromagnetic flowmeter
more expensive than the other conventional flowmeter. But, there are some advantages,
as I told you. One is the greatest advantage that it does not create any pressure loss, right?
That is the greatest advantage of this electromagnetic flowmeter. The field excitations,
what type of field I will give, because I need, I have to give some field excitation for the,
this electromagnetic flowmeter to work, right?
There are three possible ways of energizing the magnetic field coils and they are DC, by,
you can direct current you can give to the field coil. It will produce some DC voltage. It
does not matter; the pipe is, they have a DC magnetic field and the fluid is moving inside
the pipe. So, there is no harm. I will get Faraday’s laws of ……, so there is rate of change
of flux, so that will be converted into the voltage. We can, as well as AC, with 50 Hertz
AC we can have also. That is another way of energizing the magnetic coils.
Two magnetic coils, one on the top, other at the bottom, right? So, what type of electric
energy we will give? Either DC or 50 Hertz AC or pulsating DC or interrupted DC. This
15
is another method. So, we will discuss this pulsating DC in details, right? Because, that is
the most modern method of energizing the field coils for an electromagnetic flowmeter.
The three common difficulties faced by the DC excitations are polarization effect, any
DC has the polarization effect we know, because any DC has the polarization effect. This
is a very typical problem whenever you use any DC. There is a electrochemical effect,
because there is a, if there is a liquid there is a chance of electrolysis. So, the positive ion
moving one side, negative ion moving another side, so all these problems will be there
and also thermoelectric effect, because there is a DC wire metals. So, whenever you
apply DC, see if there is difference of temperature, so I will get a different, will get a
additional voltage for which we have not accounted for, accounted for, right? So, that
will give some error in the measurement.
16
So, all these will influence the DC output voltage. So, what is the solution?
Solution - these effects can be overcome by energizing the field coils with alternating
current of 50 Hertz. We can apply a 50 Hertz signals, alternating signals, right, so and it
is used for over the years, right? For many years AC excitation was most common. That
signals can be easily amplified by high gain AC. This is another advantage of the AC,
17
because DC amplifier was not popularly available sometime back also. So, now DC
amplifiers are available, but as you know op amp has some restrictions of the currents
and how much maximum voltage we will get that type of things are there. But, if we have
AC amplifier, it is easy to amplify the signal.
Now, the major disadvantage of AC energized field coil is, the output voltage is subject
to 50 Hertz interference voltage generated by the transformer action in a loop consisting
of the signal leads and the fluid path, right? Signal leads are there, we have a fluid path is
there, so in between there is a transformer action that will create problem in the case of
AC magnetic field, whenever we are using AC 50 Hertz.
The problem of DC and AC magnetic fields can be solved by energizing the coils with
pulsating DC or interrupted DC. We will use some interrupted DC, right, so that to solve
this type of problem. Neither there will be any DC problem polarization effect, then
electrochemical effect and then thermoelectric effect where those transformer feedback
action which is common in the case of 50 Hertz AC energized magnetic field. So, this
can, we can discuss in details.
18
It is a direct current that is pulsed at a fixed interval, right? The pulsed magnetic field and
the induced or output voltage of the flowmeter is shown in Figure 2.
This is Figure 2, we can see here. So, magnetic flux produced by the field coils and the
output voltage of the flowmeter. There is a non-zero voltage here, so that is called the
zero error in the case of electromagnetic flowmeters, we will discuss this.
19
So, the output voltage I can write e equal to e 2 minus e 1 plus e 3 by 2 whole minus e 4
minus e 3 plus e 5 by 2, right? Here, the dc field is switched in a square wave fashion
between the some value and zero at a frequency of 3 to 6 Hertz, right? When the field is
zero, any non-zero output from the flowmeter is considered to be an error. It is called the
zero error. When the field is zero, you see here when the field is zero, so we have some
non-zero output voltage. When the field is zero we have a non-zero output voltage, when
the field is zero we have a non-zero … So, these are the zero error, clear? When the field
is zero, any non-zero output from the flowmeter is considered to be an error. It is called
zero error, right?
By storing this zero error, the storing the zero errors and subtracting it from the output
obtained when the field is next applied, the error can be removed and the principle is
similar to the autozeroing of digital multimeter when the input is momentarily grounded
and the non-zero output voltage charges a capacitor. This capacitor voltage can be
utilized later to correct the output in the case of actual reading, right? In the case of
digital multimeter, as you know also we have A to D converter in other type of situation.
So, we will find that even though if I short the output, I will be getting some display on
the meter, so which is undesirable.
20
So, what they do actually in the case of digital multimeter, the same principles we are
utilizing here also. They are charging one capacitor by this offset voltage by shorting the
input. So, that voltage will remain intact. So, whenever I will make the actual
measurement, so we will add algebraically that voltage to the measured voltage. So, it
will be either subtracted or added, so I will get the actual reading. In practice, the zero
error correction is done several times in a second, in the case of flowmeter
electromagnetic flowmeter.
Now, there is situations when the electromagnetic flowmeter you can have metal pipe
also. You do not need lining, all these things which is expensive. That type of situation
will arise when, when the, the, the velocity of the fluid you are measuring and the fluid
has a very high conductivity. One of the good examples is suppose you are measuring the
fluid velocity of mercury. Mercury may flow through a stainless steel pipe that is quiet
desirable it should flow through a stainless steel pipe. In that situation you do not need
any liner, any lining, inside lining, because that the conductivity of the mercury is so
high, it is higher than the stainless steel. So, even if you put some, I mean sensor outside,
the probe outside, so it will pick up the signal that shortest path, right?
21
For measurement of flow of liquid that has high conductivity the metal pipe without
insulating liner is used that I said. For example, if the liquid is mercury a stainless steel
pipe will be used, where the conductivity of the mercury is much higher than that of the
stainless steel, right? The electrode in this can be installed on the outside surface of the
pipe that is very important. You can install the electrode just outside the pipe, right,
because it won’t shorten, right? So, that is the, I mean we talked about the
electromagnetic flowmeters, their advantages, only disadvantage that it cannot measure
the flow of the hydrocarbons and all those things. Other thing is, it is expensive also.
Well, how much it is rugged that depends. Obviously, it will not be as rugged as the
differential pressure flowmeters which is used over the years. It will need, I mean it need
very little attention also.
So next flowmeter, also nonlinear … type, we will discuss is ultrasonic flowmeter, right?
We will go to the board, digital board, yes ultrasonic flowmeter. You see, the ultrasonic
signal is a short burst of sine waves when the frequency is above the available range of
the frequency which is 20 kilo Hertz. This is non audible range, right? So, above 20 kilo
Hertz even we have to use ….. So, above that frequency we call it ultrasonic frequency.
22
Now, typically the ultrasonic frequency may be around 10 mega Hertz, but we will see
that we can have frequency anything between 0.5 mega Hertz to 10 mega Hertz.
The ultrasonic measurement of the volume flow rate is a non-invasive method. This is
again a non-invasive method of measurements. This is a great advantage of this type of
method, we will see, I mean later on. There are two types of ultrasonic flowmeters, so
basically. They are Doppler shift flowmeters and there are time shift, transit time
measurement flowmeter. One is Doppler shift flowmeter that is another is transit time
measurement type flowmeters. We will discuss both the case. There are some and
situation where you have to use Doppler shift, in the some situation we have to use transit
time, transit time measurement technique.
Both the method depends on the transmitting and receiving of acoustic energy. Acoustic
energy is to be transmitted and received and usually we use piezoelectric crystals for that.
The piezoelectric crystals are used for both the functions. In a transmitter, electrical
energy in the form of a short burst of high frequency voltage is applied to the crystal
causing it to vibrate. But, vibration now communicate to the, through the liquid and it
23
ends the, another diametrically opposite side of the pipe where there is another sensor
which will pick up the signal or receive the signal, right?
The crystal is in contact with the fluid and the vibration will be communicated to the fluid
and propagated through it to, what far? To the point, where the other sensors are there,
diametrically opposite side. The vibration reaches the receiving crystals which produces
an electric signal as an output, right?
24
Now let us, this is the basic principle so far I have discussed about the, I mean both the
case of Doppler shift ultrasonic flowmeter and transit time measurement flowmeter, this
is the basic principle. We will use some ultrasonic piezoelectric crystals. We will transmit
the signals and we will receive the signals. The Doppler shift ultrasonic flowmeter, it
looks like this, principle of operation let us look. The fundamental requirement of this
instrument is the presence of scattering elements within the flowing fluid, very clean
fluid.
If the liquid is very clean, if the gas or gas we cannot make the flow measurements, in
that type of situation what I am saying, the Doppler meters will not work unless sufficient
reflecting particles and or air bubbles are present in the fluid itself, which, in which we
are interested in measuring the, because it is to be reflected, right? We will show that
thing in the figure.
25
These flowmeters usually employ clamp on configuration as shown in Figure 3. You see,
this is our configurations. I have a, I am transmitting the signal from this, right? You see
here I am transmitting signal from, this is coming down here. This is reflecting particles,
so it is getting reflected with an angle theta and I have a receiver here, right? So, there is
a change of frequency. We will measure that frequency and calibrate in terms of flow. So,
fluid is going through velocity V, right and this is a clamp on measurement. That means
you can externally clamp on these meters. Even though, we are showing the, there is
some deviations, some refractions that is I am not showing in details here, right? This is
Doppler shift ultrasonic flowmeter.
26
The transmitter propagates ultrasonic, ultrasonic wave of frequency ranging from 0.5 to
10 mega Hertz in the fluid, which is flowing with a uniform velocity of V. The particles
or bubbles moving with the same velocity will reflect some of the energy to the receiver,
right? Some of the energy will be, I mean received by the receivers which is again are
piezoelectric crystals. The reflecting elements cause a frequency shift between the
transmitted and reflected ultrasonic energy.
27
The frequency is measured and it is calibrated in terms of flow velocity. It is calibrated in
terms of flow velocity and the flow velocity V is given by c f t minus f r 2 f t cos theta.
What are those legends? f t is the frequency of the transmitted ultrasonic wave, f r is the
frequency of the received ultrasonic wave or reflected ultrasonic wave.
c is the velocity of sound in the fluid being measured, theta is the angle of incidence and
reflected ultrasonic wave that make with an axis of the flow in the pipe. c, now you see
here the, we have to know the c very accurately in the medium itself, because your
accuracy of the entire measurement depends on how accurately you know the velocity of
the sound wave, ultrasonic wave in that particular liquid, right? That is true, if, I mean
found earlier, then only you can find the velocity, but whereas you will see in some other
methods this is not necessary. It will cancel out, right? That is in those methods are
preferred. Transmitter and receivers are made of piezoelectric oscillator technology. The
Doppler shift flowmeter is relatively inexpensive. This is quite inexpensive instrument.
28
The measurement accuracy depends on the flow profile, the constancy of the pipe wall
thickness, the number, size of the reflecting particles and the accuracy with which the
velocity of sound in the fluid is known, right? So, these are the all things which will be
necessary for measure, accurate measurements of the fluid velocity in a pipe by ultrasonic
method of Doppler shift. The accurate measurement can only be achieved by carefully
calibrating the instrument in each particular flow measurement application, right? That is
also in, see you have to again like electrometric flowmeter we have to calibrate
individually in all cases. That restricts the use of these instruments a lot, but it is, is quite
popular for some other application, because it is totally non-invasive technique.
29
Recently, the Doppler shift ultrasonic flowmeter has been developed with the transmitter
and receiver flush in the inner surface of the pipe. In this, the problem of variable pipe
thickness can be avoided, because in the clamp on method you have to install the
ultrasonic flowmeters from the outside, right, ultrasonic flowmeter outside the flow pipe.
But recently, the Doppler shift flowmeters has been developed with the transmitter and
the receivers flush in the inner surface of the pipe. In that type of situations, the variable
pipe thickness cannot be avoided, because you see the, if there is a variation in the pipe
thickness that will cause the problem, because you will ….. the ultrasonic waves outside.
So, it is coming through some wedge, then it is coming to the liquid.
But, this type of problem can be solved if we install the, the, if the, if the, if the sensor,
ultrasonic sensor is flush with the liquid itself, right? However, we must note following
points while measuring the fluid velocity shift, velocity shift flowmeter with the clamp on
design.
30
Dependene of c, velocity of sound in fluid cause compensating change in cos theta. For
such a design, theta is transducer wedge angle and c be the propagation velocity of the
wave in the wedge material.
What is that to we can look at that if I look at ultrasonic flowmeter, it will, this is here.
You see here, this is the wedge inside which we have put. So, the velocity of the fluid in
31
the, inside of the ultrasonic, in the wedge is more important ……….. We must know that
accurately also, right? So, those are the problem with the ultrasonic flowmeter. That is
people now, because there is a reflection also of the ultrasonic wave, so if they want to
flush it with the liquid itself.
Dependence of c velocity of sound in the fluid causes the compensating changes in cos
theta. For such a design, theta is transducer wedge angle and c is the propagation velocity
of the wave in the wedge material.
Now, let us start to discuss the transit time ultrasonic flowmeter. This flowmeter has
some tremendous advantage over the or conventional Doppler shift flowmeter. So, let us
discuss that. The transit time ultrasonic flowmeter is designed for measuring the volume
flow rate in a clean liquid or gases. This again the problem, in one case you need dirty
liquids and another case you need a clean liquids or gases. If the liquid is dirty, we cannot
make the measurement. It consists of a pair of ultrasonic transducers mounted along an
axis aligned at an angle theta with respect to the fluid flow axis, right? It will be more
clear once you see the diagram of this.
32
Transit time ultrasonic flowmeter is shown in Figure 4, next figure. Each transducer
consists of a transmitter receiver pair. So, each transducer is a transmitter receiver pair
and transmitter emits ultrasonic energy which travels across to the receiver on the other
side of the pipe, right?
33
You can see here, so this is a transmitter and this is also a receiver. So, this comes here,
receives and it again transmits, right. Fluid is flowing through the pipe. This is angle
theta.
The fluid flowing through the pipe causes a time difference between the transit times of
the beams traveling upstream and the downstream and measurement of the difference of
time traveled gives the flow velocity. What is that?
34
So, when the liquid is coming, when the, I mean this ultrasonic wave is traveling in this
direction, this upstream direction, because it is moving and when it is flowing the, when
the ultrasonic waves it is coming back which is in the opposite side of the beams, so there
is a time shift between the two, right? Received signal transmitting between the time
lapse between the signal transmitted and the received it will be different in two cases,
right?
35
So, using that principles I can measure the, you see here the fluid flowing through the
pipe causes time difference between the transit times of the beams traveling upstream
downstream and the measurement of the difference of time travel gives the flow velocity.
Typically, the time difference is 100 nanoseconds in a total transit time of 100 micro
seconds. There is again a problem. How you will measure such a small value of time? We
have to measure it …. very accurately, right? So, accuracy of the measurement of fluid
velocity comes down, falls down for the measurement of time difference of 100
nanosecond.
So, how will you measure this time shift? So, methods of measuring the time shift - there
is direct measurement, conversion to a phase change, you have a conversion to a
frequency change. We will discuss the frequency change in more details. The third
method is attractive since it does not need to measure the speed of the sound of the
measured fluid, because in previous case, in Doppler shift flowmeter we have seen that
we need the, we have to measure the fluid velocity which is not necessary in this case,
right?
36
The method also multiplexes the transmitting and receiving functions, so that some
transducer can be used both as a transmitter and receiver. So, multiplexes signal should
be there. When the transmitter is working as a transmitter, during that time you should
not receive any signal, right? So, some multiplexing functions your transmitter …., I
mean you have to make; we should say again the cost of the signal processing circuitry of
this type of transducers. The forward and reverse transit time across the pipe is t f and t r
are given by c plus v cos theta, right? If I take this, I can draw it here nicely and I have a
pipe here, have a signal here, we are receiving the signal.
So, this is the transmitter. When it is going back, forward and reverse …., so how much it
will be? This is the l, is the length between the two sensors is l actually. Please note, this
is l, small letter l. This is small letter l, right? This difference between the positions of the
two sensors is the velocity of sound plus v cos theta, because this is cos theta we have
shown, is not it? So, this is theta, right? So, this is cos theta, so v cos theta, so that is a
time. Similarly, here also it will be c minus v cos theta, right? Because in that case, what
will happen? In case of return, because signal is going in this direction, is not it, it will
take more time obviously to reach the transmitter here itself.
37
c is the velocity of sound in the fluid, v the velocity of the fluid, l is the distance between
the ultrasonic transmitter and receiver, theta is the angle between the ultrasonic beam and
the axis of the fluid flow, right? The time difference delta t is given by delta t equal to t r
minus t f. If you subtract 2 vl cos theta upon c square minus v square into cos square
theta, clear?
38
However, in actual practice, the receipt of the pulse is used to trigger the transmission of
next transmits, ultrasonic energy pulses. This is ……, otherwise how will have a machine
will know? So, the receipt, whenever we will receive the pulse we will trigger some
signal conditioning, so that it will launch the next ultrasonic pulses to the transmitter
itself. Even that time it is a transmitter and the transmitter where I have received the
signal from is actually now receiver. Thus, the frequency of the forward and return pulse
trains are given by inverse of F f is equal to 1 by t f c plus v cos theta by l, F r equal to 1
by t r c minus v cos theta by l, right?
See, if I take this frequency shift, thus the frequency, I mean if the two frequency signals
are now multiplied together, the resulting beat frequency is coming, difference and the
sum frequency will …. 2 v cos theta by l if the difference frequency, so v is l delta f 2 cos
theta. If l and 2 cos theta are same, fixed, so delta f we can see is directly proportional to
the velocity of the fluid flow, right which is flowing through the pipe.
39
The two frequencies F f and F r are mixed in such a way that the received signal contains
the beat frequency that are their sum and difference frequency. So, that difference, sum
frequency we can easily, I mean, I mean pass through a low pass filter, so only the beat
frequency will remain, difference frequency will remain. The other frequency will go
away, right and FFT analysis of this output will lead us to a peak corresponding to the
difference frequency which is a measure of the flow velocity, right? FFT analysis we will
do, so we will get a peak. So, that will give you about the flow velocity.
Peak corresponds to the difference frequency, so we will get a peak corresponding to the
difference frequency which is the measure of the flow velocity. Now, transit time
measurement is shown in Figure 5 that we can see.
40
So, this is our transit time measurements. You have multiplexer. It will activate whether
this will be transmitter or receiver and this is the receiver signals and we will have a
circuit.
If the pipe diameter is large, then the transit time flowmeter is preferred over the Doppler
shift flowmeter, because we are, at that time the transit time will be large and it will be
41
easy to measure the large time difference. That delta f also will be large in that case,
because we are not measuring the time. Ultimately we are converting to frequency and
time, frequency and then measuring the frequency. So, if the transit time is large, so
obviously what will happen? Your delta f also will be large. The accuracy is typically
0.5%. That is a typical accuracy of our, our flowmeter.
The instrument cost more than a Doppler shift flowmeter, because of the greater
complexity of the signal conditioning circuitry, right? The circuit is more complex than
the Doppler shift, but the advantage that we do not have to measure the velocity of sound
in particular liquid. So, that is I am saying the instrument cost more than Doppler shift
flowmeter, because of the greater complexity of the signal conditioning circuit needed to
make accurate transit time measurement, right?
Now, advantage of this ultrasonic flowmeter, let us look at one by one. This is in general;
I mean not only the, the flowmeter, Doppler shift flowmeter or transit time measurement
flowmeter, I am talking of the general advantage of the ultrasonic flowmeters, right?
There are several advantages of the ultrasonic that is the reason it is, slowly it is coming
up. You can say now the market share is around 10%; market share of the flowmeter in
42
the processing industry, you will find it is 10%, right? The fluid flowing through a pipe is
not necessarily to be conducting fluid. This is very, I mean important advantage you
should say that, because in case of, you will see that electromagnetic flowmeter it must
be a conducting fluid, that restriction is not there, right?
It is particularly useful for measuring the flow of corrosive fluids and slurries, right? See,
we have corrosive fluids and slurries. Even though it is not conducting, because in the
case of fluids, I mean slurries we have seen that in the case of magnetic flowmeters, the
liquid portions of the slurries should be conductive, otherwise you cannot use the flow
measurement of the slurries by the electromagnetic flowmeter. Now, in this case any
slurries, if it is conductive, non-conductive it does not matter, I can make the
measurements, right, by the ultrasonic flowmeter.
A further advantage, again I can repeat this is ultrasonic flowmeters advantage in general,
right? A further advantage of the, over other flowmeter is that the instrument is one
which clamps on externally to existing pipe rather than being inserted as an integral part
of the flow line, right? It happens in the case of electromagnetic flowmeter also. You see
here also you do not have to install inside like orifice meter, turbine flowmeter, all the
type of flowmeters, you do not have to install anything inside the pipe or rotameter also,
the bob is there which is inside. So, further advantage over other flowmeters is that the
instrument is one which clamps on externally to existing pipe rather than the being
inserted as, excuse me, as an integral part of the flow line.
43
Unlike other flowmeters, the ultrasonic flowmeter can be installed without any pipe
cutting and breaking, you see that previous slide, to existing pipe rather than being
inserted as an integral part of the … It is a clamp on. You have seen the Doppler shift
flowmeter it is a clamp on system we can have. Just outside you install the two sensors, it
will through the wedge, so the ultrasonic signal will enter the pipe. So, it is totally clamp
on system. That means outside of the pipe you install the two ultrasonic sensors. Even
though there is some problem created because of the pipe thickness and all this …. non
uniform pipe thickness that creates the problem.
So, for that reason people started to make the ultrasonic flowmeter Doppler shift where
the sensors are flushed with the liquid. But if you consider only the clamp on method, so
it is more advantageous that it is nothing to be installed inside. So, it is a totally non-
invasive technique. There are some advantages of this type of flowmeters, because of
non-invasive method. Unlike other flowmeters, the ultrasonic flowmeters can be installed
without pipe cutting or breaking. Therefore, it has enormous cost advantages. The cost
advantage is also there.
44
Its clamp on mode of operation has significant safety advantages in avoiding the
possibility of personnel installing flowmeters coming into contact with a hazardous fluid,
such as toxic, radioactive and flammable ones. If you have that type of liquid flowing
through the pipe, so the person who is working on this one, so this is also advantageous
that means if it is clamping on outside, just have a pipe, have a pipe and you just, I mean
installing outside like this to clamp on. So, it has the advantage that the person also may
be not exposed to this hazardous environment, right, where the fluid may be toxic or
radioactive, inflammable ones.
There is another advantage of this one type of meter also in food and drug industry.
So, also any contamination of the fluid being measured can be easily avoided. In many
situations, I have to measure the fluid of the, suppose in the case of beverage industries,
soft drink industries, where the liquid, the beverage or soft drink are flowing in the pipe,
so any installation of the, any meter outside that is you have to, I mean, I mean sterilize
the sensors and all those things you have to …. you have to do. But, if it is clamp on the
outside, especially in that case of ultrasonic flowmeter it does not matter whether it is
conductive or non-conductive, I can make the measurement very easily, right?
45
So, in that sense it is also very popular in the food and drug industries. Ultrasonic
flowmeter have been used mainly for liquids, but are now being applied to gases and
streams also. Basically it was previously used for liquids, but it is used for gases also and
streams. Now, there are, obviously you cannot say everything is very, I mean shiny and
very good. So, there are many disadvantage of this type of systems also.
What are the disadvantages? Ultrasonic flowmeter is sensitive to velocity profile of the
flow, right? In the case of electromagnetic flowmeter it measures the average flow
velocity. But, the ultrasonic flowmeter is not doing actually that, right? That means what
will happen? If the velocity profile changes, for the same average velocity I will get
different output. This is the problematic case, excuse me, right?
So, ultrasonic flowmeters will give different readings for axi, axisystematic profiles of
different shape, but with identical average velocity. If the average velocity is same, but
the profiles are not same, so obviously what we will find that the, you will get different
value of the reading, right, the output, which is undesirable. If you assume that the meter
coefficient to be 1 for uniform profile, this would drop to 0.75 for the laminar flow and it
will vary between 0.93 and 0.96 for turbulent flows. So, these are the two typical
46
disadvantage of the ultrasonic flowmeter, which is not there in the case of the
electromagnetic flowmeter, but for the non conducting fluids we have to use ultrasonic
flowmeter, because of its typical or invasive type of techniques of measurements, right?
So, with this, I think I, today I come to the end of the lesson 15 of Industrial
Instrumentation.
Welcome to the class of Industrial Instrumentation. So, this is lesson 16.
47
In this lesson we will consider flowmeter, because we are continuing for last several
lessons flowmeters, this lesson also will cover basically the flowmeter and the contents of
this lessons.
Hot wire anemometer one of the most widely used industrial flowmeter both for the
liquids and gases that we will discuss in this particular lesson. Also, we will solve several
48
problems on the flowmeter, various types of flowmeters. We will solve problems, we will
give the problems and also we will provide the solutions to these problems.
Obviously at the end of the lesson, the viewer will know the working principle of hot
wire anemometer, both constant temperature and constant current type. This is a basic
two types of hot wire anemometer you will find, so we will cover both. How it works,
what is the signal conditioning circuitry and advantage, disadvantage everything to be
covered in this particular lesson, along with the problems and solution to the different
flow meters.
49
Now, hot wire anemometer if I look at, the principle of operations it is based on the
principle of heat transfer by convection between the resistance wire and the flowing fluid.
The wire is kept at a temperature which is much higher than the surrounding temperature
and hence the name hot wire is given. It is capable of measuring average velocity and
turbulent flow of the fluid. It can measure both, I mean average velocity as well as
turbulent flow of the measurement. This is a great advantage of this particular flow meter.
50
So, there can be two types of schematic arrangements of hot wire anemometers. That is I
am telling either it will be constant current or constant temperature, right? So, constant
current type and constant temperature type or constant R w type.
A schematic diagram of the constant current type, you see here actually what we are
doing that initially we are taking all the resistance equal. R 1, R 2, R 3 and R w and this R
s is quite high, right and we can measure, we can see the current which is passing through
the, through this hot wire. Now, this resistance is so high compared to this resistance, you
will find due to the change of flow, the resistance even the resistance will change, the
change of current will be insignificant, right?
51
So, current is kept fixed at a certain value that can be measured by that resistance that by
the ammeter. What is that ammeter? This ammeter, the, with the ammeter we can see
whether the current is getting or remaining constant or not. The resistance R 1, R 2, R3,
are of the same order as R w. Typically it should be quite small compared to the R, R s
and the value of R s is high, right?
Generally used to, this is, this constant current type is generally used to measure the
steady flow or low fluctuations in the velocity. Not good for very high frequency
turbulent flow. We need some compensation. If you do the compensation, then it is
possible to that mean lead lag network using I can make the circuit in such a way, so that
the, I can measure the much higher frequency. Otherwise, normally it is not suitable for
the low, high frequency fluctuations. So, high frequency turbulence, turbulent of the
flow.
52
Now, this is the schematic diagram of a constant temperature type of a hot wire
anemometer. You see the, here it is a auto balancing system. This R w is put on a one arm
of a bridge, which is a our hot wire and all other resistance are put, all are same. Now see,
what will happen that if there is a, if there is a unbalance that means suppose due to
resistance change, then what will happen? You see that if there, if there is a unbalance, so
you will get a, initially it is balanced say there is no output voltage, no error voltage, but
some steady state current will flow through this resistance. Now, what will happen you
see that if there is a change, so that change will be a, if the, there is the unbalanced
voltage here and that will amplify this voltage amplifier and power amplifier. So, it will
give a current, so it will change the current through the, through this bridge and it will
continue till we achieve the balance, right?
53
So, it is auto balancing systems, we can see here. Here it keeps the resistance R w
constant by incorporating the feedback. Now, as the velocity increases R w decreases,
thereby creating an unbalance voltage and the current increases which brings back the
resistance to the initial value, right? It increases the bandwidth and thereby suitable for
turbulent flow measurement also. So, it is also for increasing, since we are using
feedback, so it is a, it is suitable for measurement of the high frequency swirls and
turbulent flow.
54
Now, this is a block diagram of a constant temperature type of sensors, we can see. We
have K B K, inert is going out. So, this is our typical systems.
Thus we get, you see delta e naught by v, this output voltage divided by change of output
voltage due to the change of velocity K V, K V, K B, K A, R O, tau V S plus 1 plus K B
K A K I which is equal to K dash equal to upon tau dash v S plus 1. Previously it is only
55
tau V S plus 1. So, what is tau, tau dash v, let us look at. So, K dash equal to K V K B K
A R O 1 upon 1 plus K V K A K I and tau dash v is equal to the, previously we have tau
v S and naturally, what will happen if I decrease a time constant of the system? All my
frequency responsible increases, is not it? You have the time constants, I can measure the
high frequency. This is the basic, fundamental, I mean fundamental equations of any, any
instruments or any systems.
Even if the …. time constant is large, I can measure the response very fast for a step
input. If the cyclic, obviously also that advantage is there. Thus we see that the system
becomes faster, since we are dividing this tau v by this factor, so obviously what will
happen, by this factor what will happen? This frequency is tau, tau dash v will decrease
and I can measure the higher frequency.
Now, comparison of the constant current and constant temperature type measurements,
disadvantage of the constant current types, the current has to be kept large and with a
sudden drop in fluid velocity it may lead to the burn out of the wire. So, this is a typical
problem of the constant current type. That means the current has to be kept large,
otherwise what will happen? So, usually it is large and if there is sudden fall in the fluid
56
velocity, so temperature may rise. Because it is constant temperature, some constant
current must flow through these ones, it will burn out the wire. For dynamic
measurement, separate compensation networks are required. That means I need a lead lag
network, as I told you earlier, to make the phase compensation, so that I can go for the
higher frequency of measurement.
Constant temperature type removes the above problem with the introductions of the
feedback loop. However, there are other problems. Instability and the drift problems of
the there, because drifts of the amplifier will come in picture, right? With this, I come to
the end of the lesson 16 of Industrial Instrumentation.
57
Prof. Dr. Alok Barua
Lecture - 16
Flowmeter – V
Welcome to the class of Industrial Instrumentation. So, this is lesson 16. In this
lesson, we will consider flowmeter. Because we are continuing for last several lesson
flowmeters, this lesson also will cover basically the flowmeter and the contents of this
lesson:
Hot wire anemometer, one of the most widely used industrial flowmeter both for
liquids and gases that we will discuss in this particular lesson. Also, we will solve
several problems on the flowmeter, various types of flowmeter we will solve
problems. We will give the problems and also we will provide the solution to these
problems.
1
Obviously at the end of the lesson, the viewer will know the working principle of hot
wire anemometer, both constant temperature and constant current type. This is the
basic two types of hot wire anemometer we will find, so we will cover both. How it
works, what is the signal conditioning circuitry and advantage, disadvantage,
everything will be covered in this particular lesson, along with the problems and
solution to the different flowmeters.
2
Now, hot wire anemometer if I look at, the principle of operation, it is based on the
principle of heat transfer by convection between the resistance wire and the flowing
fluid. The wire is kept at a temperature which is much higher than the surrounding
temperature and hence the name hot–wire is given. It is capable of measuring average
velocity and turbulent flow of the fluid. It can measure both, I mean average velocity
as well as turbulent flow of the measurement. This is a great advantage of this
particular flowmeter and moreover, time constant of this flowmeter is also quite small
compared to the other flowmeters and it is direct electrical output flowmeters. So, if
you compare this flowmeter with the DP transmitter, DP transmitter based
flowmeters, so it is a great advantage of this particular flowmeter.
Now element used is, basically hot wire filament is usually a fine wire of platinum
and tungsten. This is a, basically two wires used. I should say the tungsten is most
widely used as a hot wire probe, right? Typical dimensions are: diameter 5 millimeter,
5 micrometer to 300 micrometer. So, it is quite thin wire. It is very difficult to even
see in a bear eyes, it is a very thin wire and the length is usually 1 millimeter to 10
millimeters. It is to be installed inside the pipe. Please note thate one thing is very
important that it is to be installed, installed inside the pipe. It is not a invasive type of
flowmeter, so you have to cut the flowmeters of the pipe and install with a fork sort of
arrangement, so that it will, the length of the flowmeter will be opposite at 90 degree
to the direction of the flow of the fluid in the pipe. You see, this is our pipe.
3
We can see here this is the pipe and this is a holder and this is our hot wire
anemometer. This is our hot wire. You can see here this is a hot wire and liquids are
coming in this and liquids will go down, right? So, what is the, this basically we will
see th, this basically the heat balance equations, I mean is involved in this particular
… So, that is basic principle of this one. That we will discuss later on.
Care needs to be taken, so that the loss of heat energy due to conduction or radiation
is minimized, because basically we are depending on the convection type of heat
4
transfer, so the conduction loss and the radiation loss is to be minimized. The element
is thus jacketed and placed inside the pipe with the help of a fork type support as
shown in the figure, last figure we have shown. So, this figure we can see, so it is a
forked type of arrangement, so it is placed inside the pipe, right? At equilibrium the
heat balance equation can be written as I naught square R WO equal to U capital U v
naught multiplied by A T WO minus T F. This is equation number 1. That is heat
generated in the wire equal to convective heat loss of heat from its surface, right?
What is this?
The legends are I is the current flowing through the resistance wire in ampere, R WO
is the resistance of the wire in ohm, A is the heat transfer area that is in meter square,
T WO is the temperature of the wire in degree centigrade, T F is the temperature of
the fluid in degree, in degree centigrade, U v naught is the convective heat transfer
coefficient which is in Watt per meter square per degree centigrade. You will find this
is for the, when there is no, I mean flow in the liquid what is the resistance of the wire
we are talking about, so that will come later on, right? In addition to that, some
resistance will come whenever there is a change of flow or the flow starts to flow, I
mean flow starts to act.
5
Now, this also can be written as that means this equation I am talking about, this
equation this U v naught can be written as that means I am saying that it looks like
this, U v naught so if I take a yellow pen, U v naught equal to a b under the square
root V, where a b depends on d k row and eta, viscosity. This we will discuss. See
here, next is let us come here, where d is the sensor dimensions that is in meter, k is
the thermal conductivity of the fluid which is in Watt per meter per degree centigrade,
row is the fluid density that is in kg per meter cube and eta is the viscosity, fluid
viscosity that is in the Pascal second, right?
For the sake of the, I mean SI units we are using Pascal second. Even though in the
industry we will find still people are using the terms centipoise, poise and all those
things. Since we are, here we are actually following SI unit, because the SI unit of the
viscosity is the Pascal second.
6
Now, dynamic characteristics, now we are talking about when it is in steady state.
When it is not in steady state condition what will be the dynamic characteristics?
Steady state, there is a convection heat loss equal to the heat generated in the,
thermally heat generated in the wire. But, what will happen when there is a certain
change? Liquid is flowing in a pipe in a steady state condition, now sudden there is a
change in the fluid velocity. So, what will happen? So, there is a rate of change of the
heat generation.
So, this is no more in steady state condition. So, during that time what will be our, I
mean equation, heat balance equations, let us look at that. Taking into account the
dynamics of the system we can write, you see, rate of heat input minus the rate of heat
dissipation equal to the rate of rise of heat energy or rate of rise of temperature. So,
rate of heat input which we are giving thermally and rate of heat dissipation due to the
convecting heat loss from the wire to the surrounding that is in the fluid medium
equal to the rate of rise of heat energy in the wire itself, right?
So, we can write here, I square R W minus U(v) AT W minus T F MC dT W dt. You
see, all the units are slightly different previous equations, we can see here. You see
these are all different there. If you look at this R W, T W, T F, T F will remain same
because, it is the fluid temperature it does not depend on the temperature of the wire.
But the, all these units, all the legends you will find R W, U(v), T W, all things have
7
been changed there if I look at the previous expression, so where M is the mass of
heating element and c is the specific heat of the element.
At equilibrium or steady state, we can have the equation 1 as, for incremental change
we can write first there is a change in the wire temperature, so change of velocity.
What will happen? Let us, for incremental change I has changed to I naught plus del I.
T W, wire temperature previously it was constant, steady state condition U (v zero) in
the first equation we have written like this. Then plus delta T change of temperature
of the wire, R W is the resistance of the wire. In the first equation, in the steady state
equation we have written in terms of R WO, here we are writing del R W plus del R
W, then U(v) equal to U(v naught) delta v naught.
You see, what was our first equation let us look at. If I, I think this is already chosen
or very first equation if I look at, it was like this I naught R W equal to U (v naught) A
2 minus T F. This is equation 1, is not it? Now we have a change, because we say that
it is no more in steady state condition. That means there is heat loss, heat generated
minus heat loss equal to rate of heat energy I mean generated (inside or in site). So, it
has previously now we have, so now it has new values. We are, so it is time varying
value. So, this T W, R W are all time varying. I, R are also time varying. This, all
these elements are time varying. These are steady state conditions.
8
If it is time varying I can write incremental increase I naught plus delta I T W plus
delta T, R W equal to, R W equal to R WO plus R W. So, all these expressions, you
see these expressions I naught, T W, RW, V naught were used in the steady state
conditions, right? When it is not the steady state condition, condition it will be
something different. That is in that case you have to use these time varying
parameters, right?
Sigma we write delta U del of U upon del of v, where v equal to v naught, T W equal
to T naught. Simplifying, we are getting this expression, but we must derive how we
are getting this expression. After simplifications we will get these expressions, right?
So, let us look at how you are getting these expressions. Expression looks like this.
You see initially that you take a white page that will be more clear. You see, we had
previously if I take the, change the colour of the pen, yes our equation is, looks like
this.
9
Equation is I square, I square R W, R W is the time varying quantity, U (v) A T W

minus T F equal to M C dT W by dt. This, this was our new equations or dynamic
equations of the hot wire anemometer. Now, I replace I by I naught plus del I I
replaced T W by T W O plus delta T, yes, R W I replaced by R WO plus del R W and
U(v) I replace U(v ) naught equal to plus sigma into delta v, right? See, if I put all
these I here, I will get a expressions we have to take a new page.
10
I naught plus delta I whole square R WO plus del R W minus U v naught plus sigma
delta v A multiplied by T WO plus delta T minus T F equal to MC derivative of T
WO plus delta T by dt, right? So, I can simplify this equation neglecting all as
multiplication of the terms like delta I, delta R W, so these are all quite small. So, we
can neglect this that means we are neglecting delta I into delta R W, small quantity;
we are assuming these are all equal to zero. So, I will get a expressions I naught or I
will get sorry or I will get I naught square 2 I naught delta I R WO plus I naught
square delta W minus U v naught T WO minus T F into A minus U v naught A delta
T minus sigma A T W minus T F delta v equal to MC delta t, sorry, this will be delta
T, see if I take the eraser, right, so I take the pen again, pen again, C delta T by dt.
Now from this expression, I subtract, excuse me, I subtract the equation number 1.
What is that equations? Let me write down the equation if I recall the equation
number 1 that equation number 1 was, so from this equation if I give it equation
number 2 and our equation number 1 was I naught square R W equal to U v naught A
T WO minus T F. This was equation 1, right? So, what we are doing?
Taking a new page, so I making 2 minus 1, I will get the expression which looks like
this, which looks like 2 I naught del I R WO plus I naught square delta R W minus U
v naught A delta T minus sigma A T WO minus T F del v equal to MC delta T by dt,
11
right, clear? If this is our expressions, so let me go back to the power point
presentation again.
So, this is, so this is our expression, right? Yes, this is the sigma equal to partial
derivative of U with respect to v when v equal to v naught and T W equal to T naught.
If we have this expression, putting, we put delta T equal to delta R W by K t, we have

delta R T K i 1 plus S tow v into delta I minus K v by 1 plus tow v into delta v, so
12
where tow v is given by MC U v naught into A I square I naught square into K T,
excuse me.
Where, K V we can write is equal to K T sigma A T W minus T F when U v naught A

minus I naught square into K T and K I equal to 2 K 2 K T I naught R W naught U v
naught A minus I naught square into K T.
13
Now, you see the block diagram representation of the hot wire anemometer. This
delta I I am giving, this is change of velocity delta v and this is my change of
resistance of the hot wire, right? This is the first order system as you can see.
So, there can be two types of schematic arrangements of hot wire anemometers. That
is I am telling either it will be a constant current or constant temperature, right? So,
constant current type and constant temperature type or constant R W type.
14
A schematic diagram of the constant current type, you see here actually what we are
doing that initially we are taking all the resistance equal R 1, R 2, R 3, and R W and
this R s is quite high, right and we can measure, we can see the current which is
passing through the, through this hot wire. Now, this resistance is so high compared to
this resistance you will find, due to the change of flow the resistance even though the
resistance will change, the change of current will be insignificant, right?
So, current is kept fixed at a certain value that can be measured by that resistance, that
by the, by the ammeter. What is that ammeter? This ammeter, with the ammeter we
can see whether the currents are getting, remaining constant or not. The resistance R
1, R 2, R 3 are of the same order as R W. Typically it should be quite small compared
to the R s and the value of R s is high, right? Generally used to, this is, this constant
current type is generally used to measure the steady flow or low fluctuations in the
velocity, not good for very high frequency turbulent flow.
We need some compensation. If you do the compensation, then it is possible to, with
the lead lag network using I can make the circuit in such a way, so that I can measure
the much higher frequency. Otherwise normally it is not suitable for the low, high
frequency fluctuations. So, high frequency turbulent, turbulent of the flow
15
Now, this is a schematic diagram of a constant temperature type of a hot wire

anemometer. You see here, it is auto balancing systems. This R W is put on arm of a
bridge which is a, our hot wire and all other resistances are put, all are same Now, you
see what will happen that if there is a, if there is a unbalance that means suppose due
to resistance change, then what will happen? You see that if there is, there is a
unbalance, so you will get a, initially it is balanced, there is a no output voltage, no
error voltage. But, some steady state current will flow through this resistance.
Now, what will happen? You see that if there is a change, so that change will be a, if
there is, there is unbalanced voltage here and that will amplify this voltage amplifier,
then power amplifier. So, it will give a current. So, it will change the current through
this, through this bridge and it will continue till we achieve the balance, right? So, it is
auto balancing systems we can see here.
16
Here, it keeps the resistance R W constant by incorporating the feedback. Now, as the
velocity increases, R W decreases thereby creating an unbalance voltage and the
current increases, which brings back the resistance to the initial value, right? It
increases the bandwidth and thereby suitable for turbulent flow measurements also.
So, it is also for increasing, since we are using feedback so it is, it is suitable for
measurements of the high frequency …. and turbulent flow.
17
Now, this is a block diagram of a constant temperature type of a sensor we can see.
We have K B, K A, e naught is going out, so this is our typical system.
Thus we get you see, delta e naught by delta v, this output voltage divided by change
of output voltage due to the change of velocity K V K B K A R O tau v S plus 1 plus
K B K A K I, which is equal to K dash equal to upon tow dash v S plus 1. Previously
it is only tow v S plus 1. So, what is tow, tow dash v let us look at. So, K dash equal
to K v K B K A R O 1 upon, upon 1 plus K B K A K i and tow dash v equal to, now
previously we have tow v S and now actually what will happen if I decrease the time
constants of the system or my frequency response will increase, is not it?
Lower the time constants I can measure the high frequency. This is the basic
fundamental, I mean fundamental equations of any, any instruments or any systems.
Even if the thermodynamic time constant is large I can measure the response very fast
for a step input. If it is cyclic, obviously that advantage is there. Thus, you see that the
system becomes faster. Since we are dividing this tow v by these factors, obviously
what will happen, by this factor what will happen is frequency is tow, tow dash v will
decrease and I can measure the higher frequency.
18
Now comparison of the constant current and constant temperature type measurements,
disadvantage of the constant current type - the current has to be kept large and with a
sudden drop in fluid velocity it may lead to the burn out of the wire. So, this is a
typical problem of the constant current type. That means the current has to be kept
large. Otherwise what will happen? So, usually it is large and if there is a sudden fall
in the fluid velocity, so temperature may rise. Because it is constant temperature,
some constant current must flow through these ones. It will burn out the wire.
For dynamic measurements separate compensating networks are required. That means
I need a lead lag network, as I told you earlier, to make the phase compensation, so
that I can go for the higher frequency of measurement.
19
You see the drift of the amplifier will come in picture. If I have a drift here, suppose if
I have a drift here, drift here, so that will come in the amplifier output and we will
take output from these terminals anyway.
Instability and drift problems and for small velocity change noise has to be taken care
of separately or the measurements are erroneous, right? That also we have to think of.
20
Now, let us solve few problems on the flowmeter. Let us start with the problem
number 16.1. We will solve, we will give the questions, then we will solve the
problems. So, problem number 16.1 you see here, a measurement of the velocity
profile prior to the installation of a permanent flowmeter in a pipe is needed, right? So
we, we first want to get estimate of the flow velocity, right? The mean velocity of a
high pressure incompressible fluid through the pipe is measured using the pitot tube.
At max, at maximum flow rate the mean differential pressure was found to be 310
Pascals, right?
21
We have to measure, number a, calculate the mean velocity of the gas at maximum
flow rate, calculate the maximum mass flow rate, calculate the Reynolds’s number at
the maximum flow rate and the data given are: the diameter of pipe which is 0.18
meter, then density of the fluid 6 kg per meter cube, viscosity of the gas 5.0 into 10 to
the power minus 5 Pascal per Pascal second; this is given, right? So, let us go one by
one, let us take a white page here, so we will take a pen.
Assuming, solution 16.1, we have assuming energy conservation and no frictional or

heat loss, the sums of pressure, kinetic and potential energies, potential energies, at
the impact of the static holes are equal. This already we discussed in the case of, in
the case of when we discussed actually the pitot tube. Anyway, let me write away,
straight away write the equations: p 1 by row plus zero g z 1 p 2 by row plus v 2
square by plus g t 2, right?
So, p 2 is a static, static pressure and this is a stagnation pressure. Obviously, p 1 will
be greater than p 2, right, where z 1, z 2 are the elevations of the holes above the
datum levels that means static hole and the and the stagnation hole. So, we assume
initially that z 1, now we assume z 1 equal to z 2. Hence, we can write that v equal to,
velocity will be equal to, velocity will be equal to 2 p 1 minus p 2 by row, right?
22
Therefore the mean velocity will be given by, the mean velocity will be given by
equal to 10.16 meter per second. The maximum mass flow rate is given by dm by dt
equal to vA into row, where vA is the volumetric flow rate. So it will become, sorry,
so it will become m dot equal to v pi D square by 4, where d is the diameter of the
pipe into row. Now, D here is 0.18 meter. If I put everything there, because row is,
row is given by 6 kg per meter cube, density of the fluid, so I will get that mass flow
rate equal to 1.55 kg per second. This is one answer, this is another answer, right?
Now, we have to find the Reynolds number. Let us take a new page.
23
Now, Reynolds number is given by R e equal to row vD by viscosity. Viscosity is eta,

we are writing, where eta is the viscosity. So, this will becomes 6 into 10.16 into 0.18
by, already given 5 into 10 to the power 5, 5 into 10 to the power minus 5 Pascal
seconds. So, this will give you 21.94 into 10 to the power 4. So, obviously you can
see this is a turbulent flow, right? So, this ends the problem number 1. Let us go to the
problem number 2.
24
Problem number 2 is on turbine flowmeter, very straight forward problem; just plug in
the equations, you will get the answer. A turbine flowmeter consisting of a rotor with
6 blades is suspended in the fluid stream, the axis of the rotation of the rotor being
parallel to the direction of the flow of the fluid, right? The blades are rotating at an
angular velocity of omega radian per second, where omega equal to 4.2 into 10 to the
power 4 Q, right, where Q meter cube per second is the volume flow rate of the fluid
and the total flux phi T linked by the coil of the magnetic transducer is given by
4.637, phi T is 4.637 plus .92 cos 6 theta. This is in milliwebers, right?
Theta is the angle between the blade assembly and the transducer. These are all very
standard notation which we have given in the, our particular lessons. So, exactly we
use the same notations here, right and the flowmeter can measure the flow rates
ranging from 0.22 into 10 to the power minus 3 meter cube per second and 3.74 into
10 to the power minus 3 meter cube per second, right? Find the amplitude. Find the
amplitude and the angle, sorry; find the amplitude and the frequency of the transducer
output signal at minimum and maximum flow rate, right? So, this is our question
number 2 on the turbine flowmeter. Let us look at the solution, right? Let us take a
white page.
25
So, this is problem number 16.2, right? Here we know that phi T theta equal to alpha
plus beta cos n theta, right, where alpha is the mean flux, beta is the amplitude of the
flux variation and n is the number of blades. So, E equal to, we know equal to minus d
phi T by dt minus d phi T by d theta into d theta by dt. So, it will give you now d theta
by dt by dt is equal to minus beta n sin n theta, right and d theta by dt is equal to
omega, which is angular velocity. So, theta equal to omega t assuming that at theta
equal to zero, at t equal to zero right? Now we can obviously write, let us take a new
page.
26
That e equal to beta n omega sin n omega t, right? So, the amplitude of this signal is
beta n omega and the frequency is equal to n omega by 2 pi, linear frequency not
circular frequency, by 2 pi, right? So, omega will be, frequency for the minimum flow
rate will be, if I plug in all the expressions will become 9.24 radians per, per second,
right and amplitude will be given by 0.92, right, 0.92, multiplied by number of blades,
6 into 9.24 which is, will be 51 milli volt, right and the frequency will be, this should
be milliwebers, so this should be in millivolts. Frequency will be 6.6 into 9.24 by 2 pi.
This will be equal to 8.82 Hertz. So, this is te amplitude, this is the frequency, right
and this is for the minimum flow, right, minimum flow rate, minimum flow.
Now, let us do for the maximum flow. Take a white page.
For maximum flow, for maximum flow rate I will get omega equal to, if I plug in all
these values, 157.1 radian per second and the amplitude will come up, whereas if I
calculate equal to 867.2 milli volt and frequency equal to 150 Hertz, right? So, this is
one answer, this is another answer, right? So, these are another problem, right? So,
this is 800, I do not know whether is legible, 867.2 milli volt, right? So, the frequency
is 150 Hertz.
Let us do the other problem. Let us go to the problem number 6.3.
27
Problem 6.3 is on the Doppler flowmeter. Doppler shift, as you know there is two
types of Doppler flowmeters. Time shift measurements and the ultrasonic and the
Doppler shift measurements. So, let us first do the Doppler shift ultrasonic flowmeter,
problem on the, on Doppler shift ultrasonic flowmeter. A Doppler shift ultrasonic
flowmeter is shown in figure below which its, with its signal conditioning circuitry.
Two piezoelectric crystals each having a natural frequency of 2.4 mega Hertz are used
as transmitter and receiver. The transmitting crystal directs an ultrasonic wave into the
pipe which makes an angle of 45 degree with the flow. Calculate the received
frequency as well as the cut-off frequency of the low-pass filter. Why should I ….
low-pass filter, let us look at.
28
This is our diagram of the system that means you see that RF oscillator. So the, we are
launching the ultrasonic wave. It is getting refracted from liquid, some suspended
particles. As you know this Doppler shift ultrasonic flowmeter is not for the clean
liquids that there should be some particles inside the liquids for which the ultrasonic
wave should be reflected by the particular particle. These ultrasonic waves are
launched from this transmitter. It is coming here, it is getting reflected and coming in
this directions.
29
You see we can write here from our previous expression, f r, frequency of the
reflected wave equal to f i 1 minus v cos theta by C upon 1 plus v cos theta by C.
right? This we can write, simplify, equal to f i 1 minus v cos theta by C whole square,
where f i is the frequency of the incident beam and f r is the frequency of the reflected
beam, right, of the ultrasonic wave. So, we can obviously write f r equal to f i equal to
multiplied by 1 minus 2 v cos theta by C, right? Now, one thing you please note that v
cos theta is much, much less than C. So, f r will come up as 2.366 mega Hertz, right?
So this is our, even though it is small, so it is, 2.4 is f i. So, received signal is like this
one.
Now, the input signal we can write like this one. Let us take a blank page.
Input signal is equal to v sin 2 pi f i into t, right and the reflected signal equal to v sin
2 pi f r into t. So, output of the adder, adder is equal to v sin 2 pi f i t plus sin 2 pi f r t,
right? So, this I can simplify as 2v cos of 2 pi f r minus f i by 2 into t multiplied by sin
2 pi f r plus f i by 2 into t. So, this will give you 2v cos of 2 pi delta f by 2 into t
multiplied by sin 2 pi f r plus f i by 2 if I put in the first bracket, this will be in the
second bracket into t, right? So, you see these equations I have modulated with, with
the carrier frequency of this one. This is the carrier frequency, right and the
modulating frequency is delta f by 2, right?
30
Now let us suppose we write that, take a blank page again.
Let me suppose that we are telling delta f by 2 omega m and f r plus f i by 2 is equal
to, we are writing, excuse me, this is actually by 2 pi, I think f r plus f i by 2 is equal
to omega c by 2 pi, omega c by 2 pi. If it is there, so quite obviously I can write here
the input to the demodulator is equal to v sin omega c plus omega m t plus sin omega
c minus omega m, omega m into t. If I demodulate with, demodulate with k sin omega
c t, so the output of the modulator will be K v sin omega c t sin omega c plus omega
m into t plus sin omega c t into sin omega c minus omega m into t, right?
So, this will give you K v by 2 cos of 2 v c omega m into t minus cos of omega m t
plus cos of 2 omega c minus omega m into t minus cos omega m t.
31
So, thus when passed through a low-pass filter, see this signal is passed through a
low-pass filter, output I will get, K v by 2 cos of omega m t. So, that means we have
only signals which consists of only delta f by 2. Now, the cut-off frequency of this
low-pass filter must be greater than that of delta f by 2. So, what is delta f? Delta f we
have seen f r minus f i is actually coming as 33.93 kilo Hertz. So, delta f by 2, I can
write 16.96 kilo Hertz. So, the cut-off frequency we have chosen, frequency we have
chosen is equal to 20 kilo Hertz, right say, so this is answer of the third problem.
Now, you see the, I mean fourth problem is very straight forward.
32
The problem is 16.4. A transit time ultrasonic flowmeter is used to measure velocity
of a gas flowing in a pipe. In such a case it was found that the zero flow transit time as
1.2 milli second. That means when there is a, there is no flow in the liquid, so transit
time, the transit of the signal between the transmitter and receiver is 1.2 milli second,
whereas when there is a flow the differential transit time was 115 micro seconds. The
angle between the line connecting the transmitter and the receiver and the direction of
flow of the fluid is 30 degree.
Find the velocity of the gas and by what will the transit time change for percentage of
2% change, plus minus 2% change in the velocity of the sound. Velocity of the sound
in the gas is 500 meter per second. So, this is our problem. So, this obviously is very
familiar with, so let us take a blank page.
33
So, it looks like this. I have a pipe, so I have a transmitter here, I have a transmitter
here. So, this angle making with this is 45 degree. What is the angle? No, this is 30
degree, right? So, this is our L, length. So, we see that, see if you say A, transmitter B,
so T BA is coming L by minus v cos theta. T AB equal to L by c plus v cos theta.
So, delta T, I can write L 2v cos theta by c square, right? So, this we can write 2 V cos
theta by c into L by c. So, L by c is the zero flow transit time.
34
L by c is the zero flow transit time. That means when there is no flow, transit time is
equal to T naught. So v equal to, I can write delta T by T naught into c by 2 into 1 by
cos theta. So, if I put, plug in all the value, I will get v equal to 27.66 meter per
second, right because T naught also we know, T naught already given. Now, we see
that there is a persistent change of the, I mean flow of the fluid, right? A persistent
change in the measure, no sorry, there is a persistent change in the velocity of the
sound, how my measurement will be affected by that. So, let us look at that.
So, initially when it is, there is no change that times I …… with T 1. So, delta T 1
equal to 2v cos theta into L by c 1 square, right? So, L by c c 1 equal to 1.2 milli
seconds, is not it, right and L, I can write taking that the, there is no change on the
velocity of sound that means if it is 500 meter per seconds, so I will get L equal to 1.2
into 10 to the power minus 3 into 500, so it is coming as 0.6 meter, right? So, c 1 will
come up as 1.02, 1.02 c 1. Actually this should be c 1 I think. So, because this is a no
change so let me take the eraser, so c 1 equal to 1.02 c. Let me take the pen again,
right? C 1 equal to 1.02 C. Now, there is a change, right? So, there is a change of the
fluid velocity. So, let us take again a new page.
So, second case delta T 1 equal to 2v cos theta L by c square. So, this will become or
c 1 square rather, this should be c 1 square equal to 2v. So, this will be, if I take, sorry
if I take, I am sorry, so it will be c 1, if I write delta T 1 equal to 2v cos theta L by c
35
square. So, this will give you, if I plug in all the values, 110.5 micro second. So, for c
2 equal to 0.98 c, so we have delta T, T 2 equal to one upon 0.98 whole square into
delta T equal to 119.7 micro second. So, this will lead to the transit time variation. So,
the transit time variation, transit time variation, transit time variation is equal to plus
minus 3.9%. This is our answer, right?
Now, let us look at the problem number 5. Problem number 5 is on the

electromagnetic flowmeter. In an experiment, an electromagnetic flowmeter was used
to measure the average flow rate of a liquid flowing in a cylindrical pipe. The flux
density of the electric field applied to have a peak value of 1.4 Tesla Weber per meter
square. The output from the electrodes was fed to an amplifier with a gain of 10, an
input impedance of 2.2 mega ohm. The internal resistance developed due to the fluid
between the electrodes is found to be 200 kilo ohm, right?
36
Now, determine the velocity of the liquid when peak to peak output voltage of the
amplifier was found to be 4 volt and find the percentage change in the reading of the
amplifier for a 10% increase in the conductivity of the flowing fluid, given the
diameter of pipe as 0.1 meter.
So, it will look like that E is equal to here 2 volt. Now, the output voltage will be
given by e minus i R internal equal to i into R amplifier, right? So, e equal to E 1 plus
R internal by R amplifier, where e is the voltage induced to the magnetic field and
37
also the open circuit voltage. This basically is the open circuit voltage. So, this will
become, if I plug in all the values it will become 2 1 plus 200 into 10 to the power 3
upon 2.2 into 10 to the power 6 equal to 2.18 volt.
Now, we know that the e equal to BLv into the gain of the amplifier. If I have the
amplifier, amplifier gain or the usual notation, flux density, length or diameter of the
pipe and v is the velocity of the fluid, so this will give you obviously, if I manipulate
that v equal to e BL multiplied by gain of the amplifier. So, it will give you 1.55
seconds, sorry meter per second. I will write again, 1.557 meter per second, right? So,
if I have that thing, we will take a new page.
Conductivity increases by 9% that means conductivity increases, increases by 10%.

So, the resistance change, resistance decreases by 9.1%, right? So, R dash will be 0.91
into 200 equal to 182 kilo ohm. So, the output voltage will be equal to 2 multiplied by
2.18 upon 1 plus 182 into 10 to the power 3 divided by 2.2 into 10 to the power 3, like
this one. So, this will give you 4.026, right?
38
So, the change of output voltage, if I take a white page again, so percentage change of
output voltage is equal to 4.026 minus 4 by 4 into 100. So, this will give you 0.67%.
This is the answer.
So, with this I come to the end of the lesson 16 of Industrial Instrumentation.
39
Prof. A. Barua
Lecture - 17
Problems on Temperature Sensors

will see that we will solve some problems on the, on the, the basic temperature
sensors like thermistor, thermocouple, RTD and like change in the signal conditioning
circuitry, what are the different types in signal conditioning circuits are there, if they
need some particular accuracy what are the things we have to do that is all those
things we will discuss in details in this particular lesson.
Now contents of this as I told you, this is a problem on the temperature sensors.
Contents - the problems and solutions on RTD, thermistor and thermocouple. So,
these are the basic things which we will discuss in this particular lesson. So, let us go
to the problem number 1.
1
So, this is the problem number 17.1. This is basically on a RTD. You can see one
Wheatstone bridge is there, so this is RTD. You can see this is RTD and there are
some resistors of R, 2R; R and 2R and this is excitation voltage of this Wheatstone
bridge.
The question is, determine the sensitivity of the bridge for small value of x, right? If I
go back, you can see what is x. So, this is our x. So, determine the sensitivity for
2
small value of x when it is less than 0.1 and also determine the linearity of the output
voltage with respect to x, when x is large. So, this is the part b of the problem and part
c is that what happens to the linearity and sensitivity when the following circuit is
used instead of using the Wheatstone bridge signal conditioning circuit? So, instead of
Wheatstone signal conditioning circuits we will use one op amp based circuit and we
will, we will consider the linearity, sensitivity, all these things in details.
We will find there are some advantages of using the op amp based circuits. So, we
will discuss that thing in detail.
So, let us look at the circuits. This is the circuit, right? So, our solution starts here.
3
You see the, the solution of problem a. You see we have the bridge which looks like
this. This we have excitation to the bridge and we are taking the output voltage. So,
this is the output voltage, e o. So, this is the resistance R, this is 2R, this is R and this
is our RTD. So, its resistance is 2R multiplied by 1 plus x, right? Now, solution looks
like this. Now you see, in this particular case e A is equal to 2 by 3 of E ex. e B, if I
take this terminal A, this as B, so e B equal to, I can write 2 1 plus x upon 3 plus 2x E
ex.
Therefore, the output voltage e o will be equal to 2 by 3 minus 2 1 plus x upon 3 plus
2x multiplied by the excitation E ex, right? So, I can write here that output voltage by
the excitation will be given by 1 by 3 6 plus 4x minus 6 minus 6x upon 3 plus 2x, 3
plus 2x, so which will you give you, let me take a new page.
4
That e naught by E ex equal to minus 2x upon 3, 3 plus 2x. So, this is equal to 2 by 3x
upon 3 1 plus 2 by 3x, right? Now, when x is small, x is small, this, mod of this if I
take that means mod of output voltage by the excitation equal to 2 by 9. This will be
all small, so I can write 2 by 9. So, this will be neglected compared to this. So, this
will be only 3 in the denominator, so 2 by 9 into x.
5
So obviously, if I take a new page, sensitivity I can write here in this particular
Wheatstone bridge is e naught by E ex upon x equal to 2 by 9. It is an approximation,
obviously. Fine, now let us look at when x is large.
Solution b, if I take a new page, when x is large, when x is large e naught upon E ex
will be equal to minus 2 by 9x upon 1 plus 2 by 3x. So, this will be equal to K x, I
write upon 1 plus 2 by 3x, right where K is equal to minus 2 by 9. Now let us put, for
an example let us put x equal to 1, right, which is large compared to x less than 0.1,
right? So this is, usually we have taken x equal to 1. That means x is quite large
compared to x less than .1. So, we will get the expression, excuse me, we will get e
naught by E ex equal to 3 by, by 5 into K.
So, this is basically a straight line passing through the origin 0,0 and the point 2 by 3
K and 1, in actually the, if I plot e naught by E ex in the y-axis and x in the x-axis,
right?
6
So, this will have a locus which will look like, if I take a new page, that e naught by E
ex will be equal to 3 by 5 Kx, right, which I can write again e naught equal to 3 by 5
K 1 into x, where I can write K 1 is equal to K E ex and it is a constant, right? But the
actual value of the output voltage, actual value of the output voltage is if I take that
one as e naught dash equal to K 1 x, a very first equation 1 plus 2 by 3, this is
generalized equation, is not it, because that is the actual value we have
approximated in some cases as Kx less than 0.1, x less than, greater than, I mean
equal to 1 like all this.
7
The error voltage obviously will come up as, the error voltage, so I give it name delta
e naught equal to K 1 x 3 by 5 minus 3 by 3 plus 2x equal to 3K 1 x by 5. So, it is 3
plus 2x minus 5 3 plus 2x. So, it is coming 3K 1 x by 5 multiplied by 2x minus 2 upon
3 plus 2x, right? Now, let us take a new page.
Now, to see the errors, to see whether the error is maximum with respect to x, we
differentiate delta e naught, right? So we differentiate with respect to e naught, delta e
8
naught with respect to x. So, you will get d of dx 3K 1 by 5 2x square minus 2x by 3
plus 2x, right? So, this I will give, give you the value 3K 1 by 5 del of del x x minus
2.5 plus 7.5 upon 3 plus 2x, right? So, del of delta e naught by del x will become 3K 1
by 5 1 minus 15 3 plus 2x whole square. Now that is therefore, for maximum value of
delta e naught we will get 3 plus 2x whole square equal to 15.
If I take a new page that means x is equal to root 15 minus 3 by 2, which is equal to
0.436, right, right? Yes, so delta e naught max equal to 3K 1 by 5 2 into 0.436 minus
2 upon 3 plus 2 into 0.436 equal to minus 0.127 into 3K 1 by 5, right?
9
Now, linearity we will define. So, the percentage linearity is defined as, I should say,
percentage linearity equal to maximum drift shift divided by full scale deflection into
100. So, we will get the linearity. So the percentage linearity in this case, in this case
is equal to minus 0.127 3K 1 by 5 by 3K 1 by 5 into 100. So, this is coming as minus
12.7%. So, this is the answer, right?
Now let us solve the third, I mean c. In c we are saying that instead of Wheatstone
bridge we are connecting one op amp. The circuits already we have shown, so for that
circuit how the sensitivity will change or the linearity will change, let us look at that.
Let us take a blank page again.
10
So, this is the solution to the problem 17.1c. So problem, solution now let us redraw
the circuit again. The circuit looks like this. This is an op amp here, 2R, R, R. This is
A point and this is, all the four resistances are as usual as before, but we have taken 1
plus x, right? So, this is our RTD, so this is our RTD. So, this is 2R, now we are
connecting the excitation and taking the output from this terminal, right? So, voltage
at the point A which if, if I define as e A will be equal to 2 by 3 of excitation, is not
it?
Now, if I apply the KCL, apply the KCL at node A I will get E ex minus e A by R
equal to e A minus e naught by 2R 1 plus x or quite obviously it will be, if I take a
new page …
11
I will write again, E ex minus 2 by 3 excitation equal to 2 by 3 E ex minus e naught

upon 2 1 plus x or 1 by 3 E ex minus sorry, equal to 1 by 3 E ex 1 plus x half of e
naught 1 plus x or I can, if I take 1 by 3 E ex common, so it will be 1 upon1 plus x
equal to half of e naught upon 1 plus x, right? So, this will, I mean lead to an
expression e naught by E ex equal to minus 2 by 3x, right?
12
If take the mod, so it will look like, if I take a new page, so if I take mod of this E ex
the excitation equal to 2 by 3x, right? So, sensitivity in this case I can write that e
naught upon E ex by x equal to 2 by 3. We have seen that the sensitivity has been
improved in this case and also it is perfectly linear. That is most import…. In first
case it is, we have, the final output also becomes very much linear. So, this is the
improvement over the previous case, right?
So, now let us go to the problem number 2, 17.2.
So if I, problem number 17.2 is telling that the resistance temperature detector RTD,
this is also an RTD, a resistance temperature detector is required to measure the
temperature correctly in the temperature range of 75 degree centigrade and 125 degree
centigrade with a resolution of 0.5 degree centigrade, right? This is the range. I will
not measure below 75. This is very common in industrial problem. That is we are, not
always necessary we will measure from zero degree centigrade or 30 degree
centigrade.
There is a particular range, so within that range how will be my, what will be my
signal conditioning circuitry that we will look at. So, it is 75 degree centigrade and
125 degree centigrade with resolution of 0.5 degree centigrade. Design a signal
13
conditioning circuit for the above, for this above requirements and some data are
given, so they are that alpha that is temperature coefficient of resistance of RTD is
0.002 per degree centigrade and resistance of RTD is 350 ohm at 25 degree centigrade
and PD, power dissipation factors we are using 30 milliwatt per degree centigrade,
right? So, this is the problem. Now, let us look at the solution, right? So, let me take a
blank page.
See, the signal conditioning circuit should give a zero volt at 75 degree centigrade and
5 volt at 125 degree centigrade. That means we define our output range is 0 to 5 volt,
0 to 5 volt, right and we need this, you see basically one Wheatstone bridge along
with the instrumentation amplifier. The Wheatstone bridge is necessary to give an
output of zero volt at 75 degree centigrade and output of the Wheatstone bridge will
be very small and hence an instrumentation amplifier is necessary.
Now we have proposed the following circuits, right? Let us look at the circuit, let us
look at the circuit. The circuit looks like this. Let me take a blank page again.
14
So I have a, I have an instrumentation amplifier here and this resistance we have taken
variable, so this RTD is connected here, clear? We have this excitation, this is E ex, so
this is our RTD. This is R 2 and this is R 1. This is also R 1. This is our RTD sensor,
right? Now, instrumentation amplifier is necessary, because the output voltage is very
small, right? So, this is our e A, I should rather name it e A. This is our output and this
is the instrumentation, instrumentation amplifier, right?
So, variable resistance R 2 is necessary, so that the bridge remains balanced at 75

degree centigrade. We take that E ex equal to suppose 10 volt, right? Let us take a
blank page again.
15
So, if it is there, so you see now resistance of the, our RTD will look like this. R To,
resistance of the RTD at 25 degree centigrade, R T is the resistance of the RTD and it,
at temperature of T degree centigrade, RTD at T degree centigrade and R T obviously
we know, this R To at zero degree centigrade plus alpha we have taken the first order
approximation, so it will look like this. So, quite obviously that means R T at T equal
to 75 degree centigrade will be 350 is already given, 350, right it is given multiplied
by 1 plus 0.002 into 50. So, this will give you 385 ohm, right?
Now, to keep the bridge balanced at 75 degree centigrade, RTD is to be adjusted to

385 ohm, right? So, in that case bridge will be balanced. Now, the question is we need
a resolution of 0.5 degree centigrade, right?
16
So, if you have a resolution of 0.5 degree centigrade, the output should change by
0.05 volt for every change of 0.5 degree centigrade in temperature. Now, the error
may be introduced due to self heating of the RTD. To ensure that this is not, so we
choose the temperature change due to self heating to be 10% of the given resolutions.
So, we have taken the delta T for self heating or self heating equal to 0.05 degree
centigrade. Now, delta T we know equal to P by P D, right, where P is the power
dissipations in the RTD, right, in watt. It is a power dissipation in the RTD in watt and
P D is the dissipation constant of RTD. It is in watt per degree centigrade, right? So,
this is the, all the unit of now maximum, maximum allowable, let me take a blank
page.
17
Maximum allowable power dissipation in the RTD is equal to P equal to delta T into
P D. This is the maximum allowable power dissipation in the RTD. So, this will be
equal to, quite obviously 0.0015 watt, right which is equal to, obviously this will be
equal to I square and R T, right, I square R T loss. So, I can equate these things. So, if
I write I square R T equal to 0.0015, so the resistance at 75 degree centigrade will be,
as you know, it is 385 ohm. So, current I which is flowing through RTD at 75 degree
centigrade is equal to 0.00197 ampere or equal to, almost equal to 0. 002 ampere. This
is the current which is flowing through the RTD.
18
Now, from the Wheatstone bridge circuits quite obviously I can write, from
Wheatstone bridge circuit I can write I equal to R T plus R 1 equal to 10 volt, quite
obviously. So, this will give you that R T plus R 1 equal to 5000. So, R T we already
know it is 385 ohm. So, therefore R 1 will be equal to 4615 ohm, right? We choose a
standard value, because this resistance will not be available, so we choose a, some
standard value. We choose a standard value, value of, excuse me, value of 4.7 kilo
ohm, right?
Now, as temperature increases the value of the resistance increases, but still that does
not introduce much of error, I mean this resistance may also increase, but it is not that
error is temperature sensitive. Now again from the Wheatstone bridge circuit we see
that, let me take a blank page.
19
That output voltage e A equal to 385 plus x upon 5085 plus x minus 385 by 5085 into
10, right? So, x is the change in the resistance of the RTD. As x is very small
compared to 5085, so we can write, approximate e A equal to 10x upon 5085, right?
Now, what will be the gain of the instrumentation amplifier? Now, you see that output
of the instrumentation amplifier we can write e o equal to K into e A, so where K is
the gain of the instrumentation amplifier.
20
Now, e naught equal to 0.05 volt for 0.05 degree centigrade change of temperature,
change in temperature and x equal to 0.35 ohm for a gain 0.5 degree centigrade
change in temperature. So, K equal to, we can find K equal to 72.64, right? So, the
gain of the amplifier, instrumentation amplifier should be like this. So, this completes
our design of the problem number 17.2. The next problem will be on thermistor. Let
us solve the problem.
This thermistor problem, it is telling that find the temperature versus output voltage
relation for a thermistor using the following linearization techniques: Number a, using
a resistance in parallel to a thermistor, right, resistance in parallel to a thermistor and
b, using a thermistor in a Wheatstone bridge, right? So, this is the two problem, two I
mean thermistor with two different problems we have given. So, let us take a digital
……, we will take a board, right?
21
So, this is our problem. Now, we know that in the case of thermistor that you all know
the thermistor will have a relationship R equal to R naught exponential beta 1 by T
minus 1 by T naught, right? Now, suppose T is equal to T naught plus delta T. So, we
can write R equal to R naught e to the power beta 1 upon T naught plus delta T minus
1 by T naught. So, R naught e to the power minus beta delta T by T naught square,
so this I can write R naught into e to the power minus x, where x equal to beta delta T
by T naught square, is not it? Let us take a new page.
22
Using a parallel resistance, so if I take a parallel resistance, so this is our thermistor, R
T, I am using with a parallel resistance, so this is our thermistor, thermistor symbol
typical. We are assuming that in this case that parallel resistance has a same resistance
as the thermistor at the temperature of the t zero. So, we will get the equivalent
resistance of the, in this case R equivalent is equal to R T R naught upon R T plus R
naught equal to R naught square e to the power minus x upon R naught 1 plus e to the
power minus x, excuse me.
This I can write R naught 1 plus e to the power x, right, clear? So, R equivalent I can
write equivalent equal to R naught 1 plus e to the power x to the power minus 1,
right?
So, quite obviously this I can expand and tell that, sorry, so this R equivalent
becoming R naught 1 plus half minus x by 4 plus x cube by 48 like this one, right,
clear? Fine; now we see that x equal to beta delta T by T naught square which is very
small value. Therefore, x to the power cube, 48 and higher order terms become
negligible. So, this becomes very negligible. Therefore, we see that more or less a
linear relationship between the resistance and R equivalents and temperature change,
delta T, right? So, this will be almost equal to zero. So, these two terms will
predominate, right? Because this is very small, therefore we can predict that R more
23
or less linear relationship between the resistance R equivalent and temperature change
delta T, right? If you put delta T here, so it will be that means this thing here, so it will
get this.
Now, part b is, again let me draw the, so it is, I put the thermistor now in the
Wheatstone bridge, right and look at linearity business. So, I have the output e naught,
so I have a power supply. Now, in this scheme, all the resistances are made equal to
the thermistor resistance R 0 at temperature T 0, right? So, I assume that the R 0 plus
if all the resistances are equal, this is equal to the thermistor resistance at temperature
T 0. From the circuit we can write obviously, so there we can write that e naught by E
ex equal to half minus R naught e to the power minus x upon R naught 1 plus e to the
power minus x, right?
24
So, so now if I take a blank page, e naught by E ex equal to 1 plus e to the power
minus x minus 2 e to the power minus x upon 1 plus minus x. So, this will give you e
to the power x minus 1 and e to the power x plus 1, is not it? So, this is e to the minus
x. Now, e to the power x minus 1 we know; x plus x square by 2 plus x cube by 6 plus
so on and e to the power x plus 1 is equal to 2 plus x plus x square by 2. So, I can
write here that e naught by E ex it will become x by 2. If we make a long division, if
we make a long division of this, it will become x by 2 minus x cube by 24 plus so on.
So, we can see that the, using the Wheatstone bridge also the relation between the
output voltage and the temperature change is almost linear, right? This proves this
part. Now, let us go the problem number 17.4. This is on the thermocouple, right, so
this is on the thermocouple.
25
It is telling a K type thermocouple is used to measure a certain temperature change,

range and the signal conditioning circuitry is kept at a distance from the place of the
measurement. That means I need a cold junction compensation, because in the point
of measurement and the, and the, and the, the, the and the point of actual, the
installation of the sensor and the point of the voltmeter is far away, I need some
phase, I mean compensating cable, right or lead wires sometimes we call it. So, you
know actually from, when we studied the thermocouple that for K type of
thermocouple which is basically Chromel Alumel thermocouple, what should be the
lead wires, right?
Now, the question is design a signal conditioning circuitry taking into account the
noise as well as the cold junction compensation and use a thermistor for cold junction
compensation, right? So, instead of using, I mean 8590 and all these things, use the
thermistor. Because, we have seen that by using 8590 we can make cold junction
compensation. But here we are asking you that use the, make the cold junction
compensation using a thermistor, right? So, this is your problem, fine. If this is the
problem, so let us look at the solution.
Now, since the signal conditioning circuitry is placed at a distance from the place of
measurements, we need some lead wires for compensating cables. Here we are using
26
K type of thermocouple, so it is a Chromel Alumel thermocouple. So I, the lead wires
or compensating cables we will use is a copper constantan, right? There are two types
of, I mean lead wires. So, we are choosing copper constantan here. Now, the
thermocouple is an active transducer. It gives very few millivolts as the output
voltage. As they are highly prone to noise and environment, noisy environments, so
noise can, can also be introduced by the signal conditioning circuitry from the power
line and all these things.
To remove this we, we use a capacitance between the power supply and the circuit.
Noise is also introduced by the capacitive coupling and the magnetic coupling, so I
have introduced in the circuit, right? Now, to measure this we may use a shielded pair
of twisted cables with the shield grounded. It is then passed through the low pass filter
before feeding to an amplifier. Then we can see the circuit looks like this.
This is the circuit. You see this is the shielded cable, a type K thermocouple. There is
a compensating. This is the shielded, this is the shield and this is the twisted pair
compensating cable, because thermocouple will end here. So, after that it will go,
so it is going to a low pass filters. Ultimately it is fed into the instrumentation
amplifier, so we are using some capacitor. We will introduce, I mean eliminate the
power supply noises and all those things. So, this is our complete circuit, right?
27
An instrumentation amplifier with high CMRR needs to be used, so that any common
mode noise signal will be attenuated. So, this is our …. Now, see that for cold
junction compensation we use the thermistor, fine. Thermistor being a nonlinear
device, we need to use a linearization technique. This is then passed through an
amplifier to make the sensitivities of these two circuits identical, so that it will take
care of ambient temperature fluctuations or variation, right, right? So, the circuit will
look like this. Let me go down, yes.
So we have, it is a cold junction compensation circuit, right? This circuit will be used
along with the thermocouple circuit through some signal conditioning circuit, so that
to avoid any change in the ambient temperature. This is an amplifier. This is the
output voltage e 2. First case output voltage is e 1. We have chosen R o, this is R o.
This is R o and this is R T. Now, this cold junction compensation circuit, as it
happened in the case of 8590 also, should be placed close to the different junction
temperature, so that any fluctuations in temperature of the environment will be
nullified, right, so that, in that, in that case it will since the, the temperature of the cold
junction actually, right? Ideally they should be within the isothermal block and the
total circuit is shown in the next slide.
28
So, let us look at; so, this is our complete circuit diagram you see, right? You can see
here, this is the type K thermocouple, this already we have shown. These are figure,
the circuit 1, circuit 2. This is e 1, this is output voltage. It is not cold junction
compensated. We have a reference junction. Now, this we put on a, on a isothermal
block. You see here and this is our circuit which I have used, is not it? Just now we
have used this circuit, yes; so, this is our thermistor, this is our thermocouple. I need
cold junction compensation, so this is isothermal block. Then this is a, you see this
output is coming down here.
So, any change will be, obviously what will happen? It will be subtracted, so the
output voltage is e naught equal to e 1 minus e 2. So, the difference amplifier is
coming, because this will give you high input impedance. This is output to the, this is
coming to the instrumentation amplifier, right? So, this output is coming down here,
so difference I am getting output here, right? So, this way also I can make the cold
junction compensation, right? Fine.
29
So, we can now come to the problem number 5 or 17.5. It is telling that a simple
potentiometer circuit as shown in the figure is used to measure the, to measure the
emf of an iron constantan thermocouple. A fixed voltage of 1.215 volt is applied
across the points A and B. A current of 3 milliampere flows through the resistors and
the range of temperature variation is from 150 degree centigrade to 650 degree
centigrade. Find the values of R 1, R 2 and R G for an ambient temperature of 25
degree centigrade. Let us look at the circuit diagram.
30
So, this is our circuit, right? It is telling that the, we have to find the values of R 1, R 2
and R G for an ambient temperature of 25 degree centigrade. So, this is our iron
constantan thermocouple, right? So, the fixed voltage is 1.215 volt, standard voltage
which will not vary, applied across the points A and B, right and we have a
galvanometer to know the current. Now we have given also some data.
So, data are emf at 25 degree centigrade is 1.00, 1.022 millivolt emf at 150 degree
centigrade is 14.682 millivolt and emf at 650 degree centigrade is 53.512 millivolt,
right? That is emf at 150 degree centigrade, emf at 53, I mean 650 degree centigrade
fifty three point one, 53.512 millivolt. Now, let us look at the solution. Solution looks
like this. Now you see the current is 3 milliampere, is not it? Sorry, current is 3
milliampere. Now, I should go to the, now see what will happen here? Let me draw
the circuit again, so that will be more clear.
31
So, we have R 1, R G, R 2, it is connected to a thermocouple. It is galvanometer, so I

have point here. So, this is R 1, this is R G and this is R 2. This is A, this is B, right?
So, this is our iron constantan thermocouple, constantan thermocouple, right? So, it is
our range. So, we are saying it is 150 degree centigrade to 650 degree centigrade,
right? So, the current is of 3 milliampere, is not it, it is given. So, let me now make the
solution.
32
So, the current is 3 milliampere. So, I equal to the voltage E divided by R 1 plus R 2
plus R G. So, I will get that R 1, because this, I am sorry, so this should, you see, this
is plus R G. Now, E is given, E is equal to 1.215 volt. So, R 1 plus R 2 plus R G equal
to 405 ohm.
Now, the measured voltage, voltage at 150 centigrade is emf at 150 degree centigrade
minus emf at 25 degree centigrade. So, this will give you 13.66 millivolt. Similarly,
thermistor voltage at 650 degree centigrade, …… emf at 650 degree centigrade minus
emf at 25 degree centigrade. So, this will give you the value 52.49 millivolt. Now, at
150 degree centigrade, from the circuit we can see that R 1 into current is equal to the
measured voltage, voltage at 150 degree centigrade, right? It is not very good, I
should delete it. So, take the pen again, 150 degree centigrade, right?
33
So the or I can write R 1 equal to 13.66 by 3 equal to 4.55 ohm. Similarly, similarly at
650 degree centigrade, R 1 plus R G equal to 17.496 ohm. So, R G will become 12.95
ohm. So, R 2 will become 405 ohm minus R 1 plus R G equal to three hundred eight
seven five, 387.5 ohm.
34
So, if I write all the answers at one point, it will be R 1 equal to 4.55 ohm. R 2 equal
to 387.5 ohm and R G is equal to 12.95 ohm. So, this is our answer to the problem
number 17.5.
So, in this lesson we have seen that we have solved several problems on the
temperature sensors. When we discussed the thermocouple we have solved one
problem, while the, if, if the, the compensating cables terminals change, then what
will be the effect on the output? But these are the, we have solved several problems in
this particular lessons on the, only on the problems on the temperature sensor. This
ends the lesson 17 of Industrial Instrumentation.
35
Prof. A. Barua
Lecture - 18
Pressure Sensors
Welcome to the lesson 18 of Industrial Instrumentation. In this lesson, we will basically

cover the pressure sensors. As you know in the, in the process actually, there are three
important process parameters. In fact we call that is a pressure, temperature and flow. We
already discussed temperature and flow in details. Now, in this particular lesson, we will
discuss pressure sensors. Even though pressure has a different range, we have a low
pressure range, we have a medium pressure range and for high pressure range, but in this
particular lesson we will basically cover this high and medium pressure. For the low
pressures or the vacuum pressures we will have, that will be covered in the lesson number
19.
Now the contents of this lesson is: U tube manometer, then we have bourdon gauge,
which is basically used for high pressure measurement, excuse me. Then we have bellow
gauges and now we will find, mostly this U tube manometer, bourdon gauge and bellow
1
gauge, these are basically the monitoring instruments and it is not transmitting though in
the U tube manometer people tried to make a monitoring instrument. Then, we will
discuss the diaphragm gauge. It is very important instrument; it has a direct electrical
output, so it is widely used for the pressure measurements. We will discuss the diaphragm
gauge also. Then, we will discuss the semiconductor diaphragm gauge to some extent.
Now, at the end of the lesson the viewer will know the working principles of C type
bourdon tube, metals and alloys used for pressure sensors, detailed analysis of the
diaphragm gauge. These are basic things we will discuss, right? Now, before going to the,
let us first discuss a U tube manometer, what is the basic principle of the U tube
manometer? For this let me take a white page.
2
You will see the, in the pressure measurements we make the absolute pressure, absolute
pressure equal to gauge pressure plus atmospheric pressure, right? Now, typically this,
these U tube manometer it looks like a English letter U. That is the name, we have coined
it U tube manometer and I should say it is one of the oldest pressure measuring,
measuring devices so far the mankind developed. Still it is used for the monitoring
instruments in many process industries or the estimation of the pressures in many,
starting from your, I mean your, in the bioreactors or in the, in the, in the laminar hood or
in the actual process this actually, I mean is and it has a wide range also.
Now, typical range for this U tube manometers it range from, starting from the actually
1.013 bar, as you know. Now, 1.013 bar is equal to 1 atmosphere, as you know, right? It
starts from that range. Now, what is the basic principle? That let us look at. The basic
principle, it is like this.
3
You see that we have a U tube. Now, this relates to the atmospheric pressure. This is
unknown pressure which is coming in here and if the pressure is higher than this
atmospheric pressure, obviously what will happen? The liquid, we have a liquid,
previously we have a liquid like this, it is the same. Now, once the pressure increases, so
what will happen? This liquid column, so now it is both same. You see, what will happen
that so, liquid column will come down here, so it will come down here and liquid column
will go up. It will go like this, right? This is a simple principle. These particular
configurations will measure the gauge pressure; these particular configurations will gauge
pressure.
Now, suppose the difference of height in the two, two arms of the U tube is h, so the
pressure p 1, the gauge pressure p I should say, will be equal to hdg, where h is the height
or the difference of the columns of the U tube, d is the, that is in the kg per meter cube,
sorry h is in meter, d is density of the liquid of the manometer or manometric liquid
sometimes we call it that is in kg per meter cube and g is the acceleration due to gravity
which is in meter per second square, right? So, if we use all these units it will come
Newton per meter square, the pressure, right?
4
Now, same we, at the same time we can measure the differential pressure also with this
manometer. It looks like if I measure the differential pressure it will look like this.
So the, what will happen you see that we have a U tube as before. So, this is unknown
pressure, this is p 1; this is unknown pressure also, pressure p 2 I should say, this is p 1.
So the, actually the liquid, if the p 1 is higher than p 2, so what will happen? Liquid, there
is a height of difference, so it will be like this one, it will come like this, right? So, so this
height that means starting from this point is h. So, again this p 1 minus p 2, the pressure is
equal to hdg. As before, h is the height or the difference of the columns in the
manometer, d is the density of the manometric liquid and g is the acceleration due to
gravity.
Now, typical manometric liquid or cheap manometric liquid I should say is water; is
easily available and clean water is, you can use distilled water which is, it will be clean.
Now, there are two problems of using the water. Even though it is cheap and it is used in
industries, so the typical problem is that the water will be evaporated very quickly. So
you have, always the re should be toping that means you have to, whenever it is
evaporations you have to, I mean fill it to the point and secondly that there is a problem,
5
since it is transparent, water is also transparent, even if it is in the glass tube, if the U tube
is made of glass, so it is difficult of viewing. These two difficulties with the water, I mean
simple, I mean water, very difficult to use in the process actually.
Now, what will happen? You can make it colour. So, if you use colour water, so if you
put some, add some colour, so you can also use this particular, I mean you can use
manometric liquid as water. Now, there are two problem that the actual liquid which is
coming or the fluid which is coming if it get contact with the water, if it, it reacts with the
water that creates a problem In that type of situations, because you see the, in the, in the p
1 which we are measuring and the p 2, it will be connected to some liquid or some gas.
So, in that case that gas or liquid should not react with water. So, you have to prevent that
portions, otherwise that will create problem.
Now, typical other industrial manometric liquids are, are typically, these are aniline, then
we have carbon tetrachloride, then we have bromoform, then we have mercury,
transformer oil. These are all typical manometric liquid, right, typical manometric liquid.
Now, another problem with the water as you know that means it, because the density of
water is, it is, I mean less compared to the mercury and other liquid, so as you, with the,
6
as p 1 minus p 2 equal to hdg we have seen that thing, right, so the density of water is less
compared to mercury. Mercury is 13.6 times the density of water, so that if I use mercury,
obviously my pressure range will also be high compared to the water. So, that is the
reason people, I mean use different for that, as the case applies with, as a we have to
measure the different pressure range, obviously the manometric liquid also you have to
modify, right?
Now this is, we have used about the, we talked about the simple U tube manometer.
Now, there are other U tube monometers also which looks like this. You see we have
well type manometer or it is called the system manometers. So, well type manometer
looks like this. Instead of U tube what they do actually? So, there is a well. We have
manometric liquid here, right and liquid is here and if the pressure p 1 is, this is released
to the atmospheric pressure. So the liquid, manometric liquid is here. It is going up,
suppose this is the height.
Now, advantages of this well type manometer or the system manometer is that you see, I
have to measure the height only in this column, not here. The reason is this area is so
7
large, so change of height, because you see in the, all U tube manometers we have seen
that U tube manometer what will happen? If p 1, if p 1 is greater than p 2, so liquid, this
column will go down and this will go up, is not it? So, both the columns I must, have to
measure. This, sometimes it is very difficult. You have to measure this point, also this
point from some reference level; only that times I will get the height.
Now, here what will happen, you see the, if the pressure p 1 is higher than the
atmospheric pressure, this liquid will come down by infinitesimally small length. So,
what will happen? I can, I have to measure only in this one, h. I do not have to bother, I
should not bother for that, I mean difference of height of the liquid which is in the well,
right? So, whenever this pressure will be released, the liquid will come down and again it
will come to the, suppose this will be our previous position of the liquid, so this small
difference of height you should, should not bother. So, that is the advantage of the well
type manometer, right or it is called the system manometer.
Now, we have another manometer which is called inclined tube manometers or draft
gauge. It looks like this.
8
So we have here, I am sorry, so there should be, see here, now instead of, so they need
inclined tube, right? Now what will happen? This is the unknown pressure and this is our
atmospheric pressure. So, what will happen you see? The liquid will come down here.
Now, what is the difference between this and the inclined tube manometer and the well
type manometer? You see in the well type manometer this, this column is straight. Now,
here what will happen? The liquid, since the height is same, the liquid has to travel a
more distance here. So, ultimately what will happen? I am getting the advantage of the
well type manometer, but the sensitivity here in this case will be higher. So, I can
measure a smaller pressure difference, resolution will also be better, right? So, this is the
advantage of this type of manometer. This is called the inclined tube manometer or we
called it draft gauge.
Now, all these manometers as you know is basically indicating sort of instruments as I
told you earlier. So it is a, I won’t get any, I mean electrical output digital. So, this is
basically a monitoring instrument, neither transmitting instruments nor recording
instruments. Though people tried with, mercury in glass manometer with some electrical
output. How, let us look at. People tried with this, like this one. You see what they, what
they did?
9
Suppose there is a U tube like this one. Liquid is coming in, so it is mercury in glass, Hg
in glass. Now what they did? They put a resistance like this one and these are shorted.
Now, what will happen? You know that when the, this liquid column grows in height, so
one by one this resistance will come in parallel, right, otherwise it is not. So, there will be
change of resistance whenever there is a change of pressure . Suppose this is p 1, p 2. So,
this change of resistance, this R, equivalent resistance I should say, R equivalent, R
equivalent will be a function of p 1 minus p 2. So, that can be utilized to get unbalanced
voltage, if I put in a bridge circuit. So, that can be utilized to get unbalanced voltage of
the, unbalanced voltage as a output. So, that can be calibrated in terms of pressure.
But as you know this is a discrete form. So in between, suppose in between the two
resistance if there is a liquid column comes in, so that it won’t get any change until and
unless this liquid, manometric liquid touches this resistance and getting the change in
resistance, right? So, these are the, all about manometer. Let us go back to the other type
of gauges.
Now, as I told we will discuss a bourdon tube in more details.
10
The pressure transducer, you see that this as I told you, this we have, even though there
is, you won’t find the U tube manometer there. So, this is a entire gamut of the, our
pressure transducer. We have the bourdon tube. We have two types: we have a C type,
we have a spiral tube, we have a diaphragm. The diaphragms are flat, capsule or
corrugated and we have bellow. These are differential and absolute. So, these are the
entire gamut of the … Now, let us look at the bourdon tube in details, right?
11
You see, this is our common bourdon tube transducers and you can see here, so we have
a, we will see this one after sometime. We have a, now the cross section of this bourdon
tube transmitter is like this one, transducer is like this one. Cross section is, looks like,
sorry, its cross section looks like this. Now if the pressure increases, then what will
happen? You know this will make, try to make circular. That means it will have a, it will
try to make it circular. Due to this, I mean see, it is electrical initially, now it tries to
makes circular. This tip of this bourdon tube, you see this is a closed end, this is a free
closed end; this is free and this is closed. So, the pressure is coming here, it is going
through this and hitting this portion. This is oval cross section or elliptic cross section.
So, they hit cross sections.
Then what will happen? This will move up. It, it, it moves down. This tip moves out.
Then what will happen? So, the sector, we have a sector here, it moves in this direction.
If this moves in this direction, there is a gear in which this, you see this is a gear, you can
see this gear, right, this is a gear, so what will happen? There is a hairspring, because to
bring back this, our, this point up to the same position, what will happen? This needle
will go up, go in this direction. How? It is, again I am telling this is a, if I apply the
pressure, if the pressure is higher than the atmospheric pressure, this tip will go outside. If
it goes outside, then what will happen? This sector will move in this direction. You see,
this sector move in this direction. If it moves in this direction, this gear in which this
needle is, is connected, your pointer is connected, so it will go in this direction, right? So,
I will get a change of pressure.
It is a very widely used transducer I should say for industries, I mean it is thoroughly, I
mean everywhere, from a cycle store to any process industry you will find it is a very
wide range. It can measure the corrosive liquid, pressures of the corrosive liquids also,
gas also. So making this, starting from the liquid, gas, air, everything we can measure
with this and it is over the years people are using, even though this is basically a
monitoring instruments, not a transmitting instrument. So, since it is a C, you see this,
this tube is C, we call it C type bourdon tube and this, either it is written in PSI, pounds
per square inch or in the kilo Pascal. Again I am telling, you see the fixed open end so
12
liquids are and there is a threaded, so that it can put on a, actually the vessel in which I
am interested to measure the, measure the pressure and it is covered you cannot see from
outside. It is, the dial will be on this one so you cannot see anything from outside, right?
Now, let us look at the typical bourdon gauge of the, you see this is the dial of the
bourdon gauge.
You can see the dial of the bourdon gauge, right? You can see the dial here, fine; this is
the dial of the bourdon gauge. Now, I remove the glass cover. You see here, now let us
look at the back side.
13
See here, right? See, it is the back side. This is the C type bourdon tube; you see, this is
the C type bourdon tube. Pressure, unknown pressure is coming through this region; this
region unknown pressure is coming and it is coming and it is, this side is closed. You see,
this side is closed, right and you see with linkage it is connected to a sector; sector is
here. Now, what will happen you see that when I will move this one, sorry, if I move this
one, because this tip will move outside, is not it? So, what will happen? When the
pressure increases, it will move this way. You see, the sector will move. There is, air in
the sector will move and if I move what will happen you see here? With the sector the
pressure will increase, right? Again it is coming back to the original position.
Now we will see here, again I am telling, see we are increasing this position, because
now the, there is, it is in the atmospheric pressure. So, it is, can …., but if I increase the
pressure you can see here, you see this is the closed end. It is sealed here. This is a, this is
a typical bourdon, I mean your C type material we have seen. You see here, this C, this is
sealed end, right and here, you see here now you can see the sectors. Again I will show
you, you see here; here you see the sector is moving. You see the sector, you can see the
sector; the sector is moving. C tube is coming down and we have, so we have air spring
also here, here we have air spring which will bring back the needle and there is a gear
14
also inside. You cannot see, so I will show from the backside you can see there is a gear
also which is on the, the, actually the needle is connect, fixed on the gear itself, right? So,
I am getting the output. So, this is about the C type bourdon tube.
Now, what will happen? You know that in many situations we have a spiral. That instead
of one C I can make it a spiral, several spiral and at the tip that will be connected to the
sector. Now, advantage of that type of thing is the, you see, I am getting the more
deflections, because usually this is for the quite high pressure, right? But, if I have a
longer length of the C tube, because if I make a only single C, I won’t get much
deflections. So, instead of that single C if I make a spiral, but it does not end here. It
comes along this, we will have several turns and it ends here and that closed end now
again connected to the, to the lever. You see there is a lever here, you see there is a lever.
There is a sector here, sector on the backside; you can see here sectors. Again I am
moving it. You see here, you can see the sector is moving, right, sector is moving. So,
this way that measurements are taken, right? So, this is all about the and we have a
different shape of bourdon tube. We will see the twisted bourdon tube. So, this is all
about, of the bourdon tube.
Let us go back to the class again, digital class.
15
Now, see that this is the bourdon tube arranged for the absolute pressure measurements.
You see, what will happen here, we have a case open to atmospheric pressures. We have
a sensing bourdon tube here. This is the applied pressure here. So, there are two C tube
actually connected back to back. If I connect two C tubes back to back, now reference
bourdon tubes sealed at zero absolute pressure. This is sealed at the zero absolute
pressure. This is sealed, I mean totally evacuated and sealed it. There is no free end that
means there is no question of giving any input to this, right, whereas here I am giving the
input. So, this is, there is a free end and this is a closed end, right?
Now we, I mean liquid is going or feed is going inside. Now in this particular, what will
happen? I will get a absolute pressure, because this is, this pressure will be already
deducted. So, it is not a gauge pressure, so it is measuring the absolute pressure. But this
case is, there should be, there should be little opening, because with reference to the
atmospheric, atmospheric pressure. I am, this C tube will change its position, right? So,
this is another form of absolute measure, measurement of bourdon tube. Let us go back to
the next.
16
This is another one. You see, this is basically an axis or small angular displacement. This
is a twisted bourdon tube. Instead of, I mean C type bourdon tube we have a twisted
bourdon tube. What will happen if I applied pressure? So this, I mean side will try to
move. Again the same principle, this is a electric or oval shape. It will try to make it
circular. So, this will be rotated. So, this way also I am getting the, the small angular
displacement that can be amplified mechanically and get the output, right? This is another
form of bourdon tube. This is called twisted bourdon tube, right?
17
Now typically the, even though as I told you that in the academics we are using SI, in
industry still people are using this, I mean PSI. They, still they call it 1 PSI, 3 PSI. So,
you have two side, two things should be side by side. That means both the SI unit as well
as say PSI unit, right, because you know the pounds per square inch is basically PSI unit.
So, 1 PSI is equal to 6894.76 Newton per meter square, which is equal to 6.894 kilo
Pascal. Bronze pressure springs up to, now I talked about the pressure spring that means
what is the material of the C type bourdon tube? We have seen sometimes we are using
brass, sometimes we are using a phosphor bronze.
Now, depending on the particular range the different pressure transducer is used. Now,
bronze pressure springs up to 4136856 Newton per meter square. Berryllium copper
spring up to 68947600 Newton per meter square. Then we have a steel and alloy steel
spring for pressure range greater than 6894 into 10 to the power 4 Newton per meter
square. Sometimes you know that the, this range of this particular instrument depends not
only, not only we should consider the range, but also the, what the environments.
Suppose I am measuring the, I mean liquids, pressures of the corrosive liquids, in that
type of situations we do not have any use or we do not have any choice rather than using
18
the, your stainless steel C tube, right? So these are the different materials of the C type
bourdon tubes. These are the different materials. These are the different material of the C
type bourdon tube. It is typically used and the, the C type bourdon tube which I have
shown in the, I mean in the, just few minutes back is basically, the material they are using
is the bronze, right, easily available. The beryllium copper, these are not very, is
expensive also, much expensive than the, I mean the Bronze pressure gauge.
Obviously, you can see the range. The range of the Bronze is much less than the
Beryllium copper and steel alloy and all these things is necessary, depending on what
type of environments we are using, what type pressures, I mean in what situations or
what conditions you are measuring the pressure. In that type of the situations we use, I
mean different types of C tube. But basically, these three are the, the three basic C type
materials and among that bronze is the most widely used pressure C tube material.
Now let, next let us go to the bellow gauges. Bellow gauges is very widely used. You see,
what will happen here, so that you see here we have a bellows, we have bellows. Now,
what will happen? You see, this is a closed end, it is closed from outside. This is closed
from outside, this is closed, sorry. You see here it is closed, right? What happened? Here
19
you see, the bellows are, I will show you the bellows. Now if the, if the pressure
increases from this side, if I, pressure goes inside these bellows will squeeze. If it
squeezes, then what will happen? This pointer or link will go up, right? So, it will move
like this one. You see, it will move like this one. If it moves like this one, then what will
happen? So, I will get an indication. So, this can be calibrated in terms of the pressure,
right? So, this is all about the bellows. So, we can show in next slides something, you see
here.
The pressure range of the system is determined mainly by the effective area of the
bellows and the spring gradient. Commonly the bellows are of brass or phosphor bronze.
These are basic material used for the bellows and bellow, these are basically annular ring
and welded at two different ends. I will show you. This is basically annular ring welded
at the alternative points and bellow gauges are commonly employed for measuring gauge
pressures range up to 689476 that is in Newton per meter square that is 100 PSI typically
and bellow materials are brass or phosphor bronze. Brass is most widely used bellow
materials and we can show, show some elaborate arrangement of these bellows, let us
look at.
20
You see this is a bellow. This applied pressure is coming down here, so we have a pointer
here. Again there is a sector and this needle or pointer is connected on a, on a, on a, on a
gear. So, in this actually there is a teeth of this sector. So, we have not shown, we have
not shown this thing. Now, what will happen? You see, the pressure increases this bellow
will squeeze. This bellow will go up. If this goes up, then what will happen? The sector
will move in this direction. If the sector moves in this direction, then what will happen?
You see that this pointer will move in this direction, right?
So, this scale can be calibrated in terms of, in terms of the pressure, right? So, this is a
pinion and sector gauge movement, this is a connecting link. This is a spring, because
they …… because once you remove the pressure this, this bellow should come to the
original position. That is the reason the spring should be inside and this is a case and case
should be hermetically sealed. That means it should be sealed from all end that means
from this side there should not be any leak, this side there should not be any leak. As you
know, the bellows are also used for the, I mean pressure switch that will be discussed
later on. Let us go and see one bellows.
21
You see this is the, you see this is the bellow. See, I am squeezing, you see the pressure,
if the pressure increase, goes inside through this one, so it will be squeezed. It will be, it
will, if the pressure increases, then what will happen? You see, if that side it will be
squeezed like this one, right, it will be like this one, right?
22
You see, now you look at very carefully these are welded actually and join. This is made
of brass. You see, this side is welded. Again the inner side is also welded, right? This side
is welded; the inner side is also welded. So, we will get the movement like this one, right
and as you know, the bellows are extensively used for measuring the, for using as
pressure switch also. How? You see what will happen if this is hermitically sealed? In the
refrigerators and this is typically used for your temperature control, because I can use this
particular contact point for the temperature control. Then what will happen? You see, if
the pressure increases, I mean suppose this is sealed unit, the pressure decreases then
what will happen? It will squeeze like this one. If it squeezes, you have an electrical
output that will be disconnected, right?
So, you see here how the pressure like this one it is doing like this, right? If the pressure,
outside pressure with, the inside pressure is less than the outside pressures it will go like
this. If the inside pressure is less than the outside pressure, it will squeeze like this. If the
inside pressure, the pressure given, which is more than the atmospheric pressure it will
expand like this. You see, we cannot expand it. So, it will expand like this, clear? That
means in inside pressures which are pressure giving is more than the atmospheric
pressure, I will show you again it will expand like this, right? So and this particular
bellows is made of brass. This is used for some force balance system. We are taking that
thing to show you. Let us go to the digital class again.
23
Now, the diaphragm pressure transducers, now the diaphragm pressure transducers is
very widely used, I mean pressure transducers actually we will discuss this particular
pressure transducers in detail. Diaphragms are used for low and middle pressure ranges.
So. it is 1 to 20 into 10 to the power 4 kilo Pascal, right? Both tension and compression
stresses exist simultaneously in the diaphragm gauges and this allows the use of a four
active arm bridge in which all effects are additive which gives both the temperature
compensation and large output as well.
24
In computing the overall sensitivity, the expression of tangential and radial stresses
cannot be applied to find overall sensitivity of this. Since the diaphragm surface is in a
state of biaxial stress and both the radial and tangential stress contribute to the radial or
tangential strain at any point, the general biaxial strain relations are …
25
Epsilon r which is radial strain equal to S r which is radial stress minus nu S t upon E.
Similarly the epsilon t which is a tangential strain equal to S t minus nu S r upon E, where
S t is the tangential stress and nu S, S r is the tangential radial stress and nu is the
Poisson’s ratio and E is the Young’s modulus of elasticity. Now, S r can be given by, this
S r can be written as 3 PR square nu by 8 t square whole bracket 1 by upon nu plus 1
minus 3 upon nu plus 1 small r capital R whole square. S t that means the tangential
stress 3 PR square nu by 8 t square 1 by nu plus 1 minus 1 by nu plus 3 small r by capital
R whole square, right?
The P is a pressure difference across the diaphragm. Now, diaphragm is, looks like this
actually. If I see that looks like this. They have a stretched diaphragm, right? Suppose this
is P 1 and pressure on this side is P 2 P 1 greater than P 2, so this diaphragm will be
stretched like this, right? So, this centre point deflections either you can use LVDT or we
can use some force strain gauges to sense this particular, I mean deflections. So, that is
calibrated in terms of pressure, right?
26
Let us look at the legends. What are the legends? We have t is the thickness of the
diaphragm. This is very important. It relates to the linearity of the sensor, then radius of
the diaphragm, outside radius of the diaphragm, then we have Poisson’s ratio, position
parameter, I am sorry this is the position parameters, I will tell you what is that actually?
Then we have, S r is the radial stress, S t is the tangential stress.
27
Once the strains are calculated, the change of resistance delta R’s of the individual gauge
are obtained from the gauge factors. Then e naught can be determined from a bridge
circuit sensitivity. Now, epsilon t assumes the maximum value at r equal to zero and the
radial strain epsilon r is positive in some regions, but negative in others and assumes its
maximum value at small r equal to capital R.
You see here the diaphragm, how it looks? You see, this is the position parameter small r,
right? You see here if I take a, this is the position parameter, right? At any point what is
the deflection, right and height at this point, particular point is y and t is the thickness of
the diaphragm and R is the radius of, of the, the I mean total diaphragm or the inside
radius of the pipe. Almost it is equal to the inside radius of the pipe and at any point from
the center, center volt deflection will be the maximum. These deflections we are calling it
y c. These deflections we are calling it y c. So, at any point deflection we are calling it y.
From the center at a distance r the deflections we are calling it y. When r equal to zero at
the center we are calling y equal to y c, right?
28
Now see, this is the strain distribution. So, diaphragm gauge and its stress distribution
looks like this. In both the cases we have both tensile and compressive stress. Here you
see the S r or the radial stress is given like this one and radial stress if we look at very
carefully, radial stress is maximum at the circumference which is compressive. This
radial stress is compressive and it is maximum at the circumference, whereas this
tangential stress is maximum at the center point. Why I am telling that is very much
important. You see, again I am telling you see, this radial stress is maximum at the
circumference.
So, we will put two sensors at the circumference which will sense the radial strain and we
will put two sensors at the center which will sense the tangential stress, right and there is
some particular, I mean strain gauge available for that which is called strain gauge …..
We will come to the details of that. So, this is our diaphragm gauge and stress
distributions.
29
Now, flat diaphragms show nonlinearity at large deflections, whereas stretching action
adds to the basic bending causing a stiffening effect. So, this should be the, the deflection
of the center point which we have to keep minimum. Otherwise there will be distortions,
right? This nonlinearity in the stresses follows closely the nonlinearity in the center point
deflection y c of the diaphragm, for which the following relation is available. This is the
relation between the, this is the relation between the pressure P and y c, because you see
all other things are constant.
Young’s modulus of the elasticity of the material of the diaphragm is constant, thickness
of the diaphragm is also constant, t is constant. Now, nu is Poisson’s ratio that is also
constant, R is the radius of the diaphragm that is also constant. Radius of the diaphragm
is a constant that means if I have a diaphragm here, so this, this is radius R, right? This is
our diaphragm relation between P and y c and see, very interestingly this is our nonlinear
term, we have to avoid this thing. That means this part should be very, very small
compared to this part.
So, you can see very easily if the y c is very, very small compared to t, if the y c is
smaller than compared to, we can neglect this part. So, nonlinearity will be less. If the y c
30
is higher, then the nonlinearity will be higher, then we cannot neglect this term, whereas
say this term is much, much less than this term, so we can easily neglect that part, right?
So, you will solve some problem; how much nonlinearity will be introduced by this?
Obviously, but always you see the deflection of the, deflection of the center point or
center point of the diaphragm should be much less compared to the thickness of the
diaphragm. So, these two are the quite related, y c and t to have, to eliminate the
nonlinearity in the system. Let us go back again.
By designing for a sufficiently small value of y c by t, the desired nonlinearity may be

achieved. However note that small y c by t, please note, also gives smaller strains and
output voltage, right? Why? It is very clear. You see, what will happen if I have a
diaphragm like this one? So, I have a deflection like this one, right? So, either I have to
sense this center point deflection by some LVDT or there is some strain gauges will be
there. See, if the deflection is less, then what will happen? This strain developed also will
be less. So, I will get the less output, right? So, those things you must keep in mind.
31
Now, deflection at any point, any other point than the center point as I told you at the
distance r, y is equal to, is given by this expression y equal to 3P 1 minus nu square R
square minus r, small r square 16 E t cube. What is that, let us look at. Basically that
means on the central point suppose I have, at any point, I have a diaphragm, now it is
deflected like this one. This is our centre point. Suppose this I am taking about y, so this
distance is y, this deflection of the central point and this distance from the center is r, so
that deflections or height is here y at this point, which is at a distance r from the center.
For that reason our expression of the deflections y can be given by this particular
expressions.
32
A special purpose strain gauge rosette, which has been designed to take advantage of this
strain distribution and is widely used in diaphragm type pressure transducers. Two strain
gauges will be at the center of the diaphragm where tangential strain will be maximum
and two strains, similarly two strain gauges are installed at the edge of the diaphragm
where radial strain is maximum. We will show you.
33
You see, this is a bit typical. Now you see here what will happen? You see here, this is at
the circumference. You see here, this is R 2, it is not coming anyway, so I can draw it
here, I am sorry. What will happen you see here, this is the, this sensor, this strain gauges
and this strain gauges will sense, this R 2 and R 4, it is not coming; this is R 2 and this is
R 4. This will sense the radial strain, right, which is maximum at the circumference. So,
that is the reason it is, you see this is a total diaphragm.
This is the diaphragm, size of the diaphragm or I should say this is a, from this this is the,
distance is radius R. The two strain gauges are at the center. You see here this R 1 and R
3 will sense the tangential stretch, stress which is maximum at the center point which is
positive. Whereas, the radial stress which is at the two extreme positions is, is basically
compressive and it is negative, right? Because we have seen the strain, the stress
distribution curve, in some case it is positive, in some case it is negative.
Strain gauges R 2 and R 4 are oriented to read the radial strain and placed as close to the
edge as possible, since the radial strain has maximum negative value at this point, as I
told you earlier. Strain, strain gauges R 1 and R 3 are installed as close to the center as
possible and it will read tangential strain, since it is maximum at this point.
34
You see this is our strain gauge. R 1, R 3 which will basically sense the tangential stress,
R 2 and R 4 which will sense the radial stress and it is compressive. So, it will be
negative. So, resistance values are decreasing, here this is increasing. All are active
gauges. So, strain gauge rosette in a Wheatstone bridge, so our output will be quite
obviously ….
35
In Wheatstone bridge tangential elements are in arms R 1 and R 3 and radial elements are
in R 2 and R 4, right? Radial elements are R 2 and R 4 and tangential elements are R 1
and R 3. R 1 and R 3 is positive, right? So, the output voltage E naught with lambda
equal 2 for the diaphragm sensor using strain gauge rosette supplied by the manufacturer
is as follows: so, you are using advance. Since lambda equal to 2, obviously you can see
it is advance, so E, E naught equal to 0.82 PR square 1 minus nu square upon Et squared
ex.
Or I can say that P equal to 1.22 Et square E naught upon R square 1 minus nu square
here E ex, where C into E naught, where C is a calibration constant which is equal to 1.22
Et square, R square 1 minus nu square by excitation voltage of the Wheatstone bridge.
36
Diaphragm strain gauges rosette installed on backing materials are commercially

available in several sizes ranging from 0.45 centimeter to 3.2 centimeter and the
sensitivity of the diaphragm pressure transducers Wheatstone bridge combinations will be
given by S equal to E naught by P upon, equal to 1 by C equal to 0.82 R square 1 minus
nu square E ex square Et square. You know that we can measure this deflection by the
LVDT also, but we are not discussing. We are discussing the strain gauges, so this is the
sensitivity.
Sensitivity, you can see that it depends on the power supply, on the excitations, on the
radius, diaphragm, thickness of the diaphragm, Young’s modulus. Poisson’s ratio is
hardly, in India, in our country is mostly 0.3 for steel.
37
Unlike other transducers, diaphragm deflection rather than the yield strength determines
the limit capital R by t max. It can be mathematically shown that the relationship between
the pressure and the voltage will be linear to within 0.3% if the deflection y c at the
center of the diaphragm is less than t by 4.
38
Now, neglecting the non linear term, obviously deflection at the center of the diaphragm
y c can be expressed in terms of the pressure as this relation, right? This already we have
given in terms of P. Now we are neglecting the, that cubed term that the third I mean that,
so if we neglect that terms, so I will get a expression. So, over all I can, I should say it is
almost, y c is almost center point deflections equal to 3PR to the power 4 1 minus nu
square 16 t cube into E.
With the restriction that y c should be less than t by 4, because this is the restrictions we
have to follow, otherwise that nonlinearity term we have to introduce here, so that with
this restriction, the maximum pressure which I can measure should be less than 4 by 3 t
upon R to the power 4 multiplied by E upon 1 minus nu square. We can see this ratio is
very important. That means thickness to, thickness to the radius of the diaphragm is very
important to control the maximum pressure, right?
Pressure transducers that use a diaphragm sensors are well suited for either static or
dynamic pressure measurement, quite obviously. The diaphragm sensor has a very high
natural frequency with a small damping ratio, because of its low mass and relative
stiffness.
39
However, the high frequency limits of the diaphragm pressure sensor depends primarily
on the degree of damping provided by the fluid in contact with the diaphragm, because
not necessarily we measure the gas pressure, we can measure the fluid pressures or gas
pressures also. In that case also that damping will be basically controlled by that
particular liquid. The resonant frequency of the diaphragm should be 3 or 5 times higher
than the highest frequency of the applied dynamic pressure.
40
A diaphragm has an infinite number of natural frequencies. However, the lowest is the
only one of our interest, we have, we should be. For clamped edge diaphragm vibrating in
contact with a fluid having density of rho f that is the fluid density in which it is in
contact, the lowest natural frequency of the fundamental is given by f equal to 10.21 upon
pi R square under the square root Et square by upon 12 1 minus nu square rho into and
that is in Hertz. This rho is basically the density of the, density of the diaphragm material.
E is a modulus of elasticity in Pascal, t is the thickness of the diaphragm in meter, R

radius of the diaphragm in meter, rho is the density of material of diaphragm in kg per by
meter cube, rho is a Poisson’s ratio and for steel diaphragm, the equation 1 can be
simplified as f equal to 4.912, this is, I mean in Hertz we are telling, equal into 10 to the
power 4 into t upon pi R square.
41
Now, another type of, I mean diaphragm gauge is avilable, is called the semiconductor
diaphragm gauge, because as you know the semiconductor is very widely used. So, it is,
we are using the diaphragm as a semiconductor. Here actually, this, the entire diaphragm
is not made of semiconductor material and what will happen you know that where you
want to get, because there also the rosette, rosette concept is there, right? So, rosette
concept is there also. What they will do that , actually that, in that, in particluar what will
happen you know that where you want to make the gauge in that paricluar point they
dope it, right?
So they, that way they are making the semiconductor. Now semicondtor, the advatage
you know the sensitivity, because the, one thing you have noticed that in a diaphragm
gauge the less the deflections of the central point or the less the deflection of the pressure,
so it is better. Now the problem is if I want to make very low pressure, if the deflections
are very small, then I won’t get any output, because strain and because lambda is equal to
2 and what about the output unbalanced voltage in Wheatstone’s bridge I will get, that is
…… may not be sufficient to measure or calibrate in terms of pressure. Whereas,
assuming in the case of semiconductor strain gauges, thus the gauge factor is very high.
42
Now, if it is advanced it is 2, lambda equal to 2. In the case of semiconductor strain
gauge, the lambda is 130 or 140. With this high, such a high, I mean the value of the
gauge factor, advantage is that, advantage, what the advantages we will get that you
know I can measure very small pressure, because the deflection at that point can be very
small. But, I will get a larger unbalanced output of the bridge. So, helping us to measure
the low pressure there also; not that low, but the medium pressure, right? So, that is what
is the, inside the diaphragm I mean semiconductor diaphragm gauge let us look at.
Because of its small gauge length, a semiconductor strain gauge can be used for
constructing very small transducers. These transducers may for example be used for
measurement of the pressures and due to their small size, would allow measurements of
strain at higher frequencies.
In diffused semiconductor gauge the diffusion process employed in integrated circuit

manufacture is utilized. In a pressure transducer, the diaphragm would be silicon instead
of metal and the strain gauges are made by depositing impurities at specific locations, as I
told you. So, I will not use any metals, I am not using brass, steel or anything. I am using
the semiconductor wafer itself as a diaphragm and where I want the particular strain
gauge, particular that position will be doped, right? As I told that the pressure transducer
43
diaphragm would be silicon instead of metal and the strain gauges are made by depositing
impurities at the specific locations.
The use of semiconductor technology in pressure transducer construction has led to the
development of variety of very fast, very small, highly sensitive strain gauge diaphragm
transducers. Silicon piezo resistance strain gauges can be diffused, because as you know
that in the case of the, case of semiconductor strain gauges the piezo resistive effective is
more, right? So, silicon piezo resistance strain gauges can be diffused in a single silicon
wafer which forms the diaphragm, right?
44
Now, semiconductor strain gauges have a sensitivity that is 50 times greater, almost 50
times; sometimes even more. Because in the case it is I have told you in case of advance
it is 2, but if you use some other, suppose isoelastic which has around 3 to 3.5, in that
case it is 50 times, 50 times greater than the conventional metallic strain gauges and
because of piezo resistive gauges are integral part to the diaphragm, they are relatively
immune to the thermo elastic strains prevalent in the conventional metallic strain gauge
diaphragm constructions, right?
45
Now, further more a silicon diaphragm will be, will not creep with age. That is the typical
problem with the metal diaphragm, it is not creeping. But not always everything is good
in the case of semiconductor strain gauges. It has been minimizing the calibration drift
over time. However a gauge failure is catastrophic in the case of, which is in the case of
metal strain gauge is hardly it fails, whereas our gauge failure is catastrophic in the case
of silicon and its performance deteriorates with the presences of the moisture. That is I
am telling that however the gauge failure is catastrophic and silicon is not well suited to
wet environment, right?
So, that is, most times it is very difficult to control that the, what type of an environment
that there might be a wet environment also. So, in the case of, I mean if I have a metal
gauge that problem does not arise, but if we use a semiconductor, if it is the, measuring
the pressures of dry air, so in that situation there is no problem, right? But other cases it
will be a problem, right and since, but since it is, developments of the, it is not that
popular I should say, because after the development of the strain gauge rosettes force
strain gauge rosettes which is used for the diaphragm gauge, it is, I mean people are more
using that sensor than the diaphragm.
46
But diaphragm, the semiconductor diaphragm is the integral part. You do not need any
strain gauge rosettes there, which is to be installed on the metal diaphragm. Because you
have to wait for sometime, because you are drying, because strain gauge rosette will be
available separately on a backing material that is to be installed normally, installed on a
metal gauge, on the metal diaphragm and you have to keep using all these adhesives.
Here, we have to use that thing epoxy. Then you have to dry it, you give the time for
drying as it happens in the case of other strain gauges. But that type of situations does not
arise in the case of and since it is factory processed, this entire things, you are not using
in the laboratory or in the, in the, in the actual process. So, the advantage is its accuracies
is also being better.
This ends the lesson 18 of Industrial Instrumentation. In the lesson 19, we will start to
make the low pressure measurement.
47
Prof. A. Barua
Lecture - 19
Low Pressure Measurement
Welcome to the lesson 19 of Industrial Instrumentation. In this lesson we will

basically cover the low pressure measurement. In lesson 18, we have seen that the, we
have basically discussed the medium and high pressure measurements, but low
pressure measurements also is a interest, is a part of interest for, even though not
much for the industry, but for the academics and also scientific community. So, the
low pressure measurement also is very much necessary and is a different class of
instruments. We will find the principle of operations of these instruments is also
totally different.
So let us look at lesson 19, low pressure measurement.
Contents of this lesson are McLeod gauge, we will see the, we will cover the McLeod
gauge. Then we will cover the pirani gauge, then thermocouple gauge, ionization
gauge and Knudsen gauge. Now, we will see that the McLeod gauge actually, I mean
these are, all these pressure gauges we will find it, they were, they have different
1
application as well as different range of the pressure measurements. It is all lies
between 1 torr to 10 to the power minus 10, 10 to the power minus 12 torr. Torr
means 1 millimeter of …., so that we will cover. What are the units of the pressure
measurements that we will discuss after sometime.
Now, in some instruments we will find it is basically monitoring sort of instruments

or indicating instrument, but some instruments we will find, we will get electrical
output. So we can utilize to offer control for transmission, all those facilities will be
available. We will particularly see that the McLeod gauge and the Knudsen gauge, so
let us discuss for the McLeod gauge. So, at the end of this lesson, so we will know all
the details of these types of gauges. Also we will consider some problem solving. We
will probably make some problems and we will give the solution to some pressure
gauges, diaphragm gauges, especially and also on the McLeod gauges, right?
Now, before going to the details of the pressure measurement let us look at the units
of the low pressure measurements. The different units are 1 micrometer equal to 10 to
the minus 6 meters of mercury or sometimes we call it micro in million or micron
which is 10 to the minus 3 millimeter of mercury and now 1 micron means, sorry you
see that if I say that 10 to the minus 3 millimeter of mercury that means 10 to the
minus 6, if it is SI unit, 10 to the minus 6 into the density of the liquid multiplied by
2
the acceleration due to gravity that will give you the pressure. So, 1 micrometer is
equal to 10 to the minus 3 millimeter of mercury equal to 0.133322 Newton per
square meter, right and 1 torr we will find it is 1 millimeter of mercury. 1 torr is a
lowest pressure measuring units. So, we call it 1 millimeter of mercury, right? So, 1
millimeter of Hg or 1 millimeter of mercury.
Let us look at the McLeod gauge. The McLeod gauge is considered to be a vacuum
standard. We will find that the, the pressure can be computed from the dimensions of
the gauge. It is very important, it is independent of the so many parameters, so it is,
can use as a pressure standard. The principle of the McLeod gauge is the compression
of a sample of the low pressure gas to a pressure sufficiently high to read with a
simple manometer, right? We will compress the gas, low pressure gas to such a value,
so that we can, simple manometric principle I can measure that pressure. That is the
basic principles of the McLeod gauge.
A low pressure gas will be taken and it is systematically or very intelligently, we will
compress it to sufficiently high pressure, so that a simple manometric, manometer
method I can measure the pressure. That pressures will be calibrated in terms of that
low pressure actually in which we are interested, right? So, that is why I am saying
the principle of McLeod gauge is the compression of a sample of the low pressure gas
3
to a pressure sufficiently high to read with a simple manometer. Reading is most
important, reading is independent of gas composition. That means I can take any gas.
So, I do not need, so I do not have to recalibrate the instrument and that is the great
advantage of the McLeod gauge. This advantage is not there, we will find in other low
pressure gauge like ionization gauge and all those things. It is totally different. Let us
look at more details of the McLeod gauge.
You see, this is the schematic of the McLeod gauge. You see, what will happen? You
look at very carefully. You see here, we have a movable reservoir, right? So this
reservoir I can put, move it up and down. This can be moved up and down. This is a
flexible tubing; even though we have drawn like this one, this is a flexible tubing.
This is a, these are all flexible rubber tubing, right? It is connected to the, this
movable reservoir. This reservoir can go up and down. Actually there is a frame; there
is a frame like things through which this reservoir actually moves, right?
Now, we have a capillary here. You see this capillary we have. Now, principle is that
first this movable reservoir will go down. If I keep it go down you see the mercury
from this reservoir is coming and it is filling up all these. You see, it is coming
through this region. It is coming and filling up to this level. So, accordingly, this will
be the same, this should be at the same level. Now, what will happen? Once I move
4
this reservoir downward, is all these mercury in the, this tube as well as in this
capillary and the bulb will be collected in the reservoir, fine.
Now, if I put the mercury below this level, below this zero level, you see what will
happen? A gas which, in which we are interested to measure the pressure will enter.
So far it cannot come down, now once I push the mercury down here at this region,
then what will happen? You see this mercury will come, this gas will come down and
it will enter this region, this capillary. Slowly it will enter the capillary also. So, basic
principle is we will push this movable reservoir down as long as mercury comes down
to this zero level.
Now, I push this reservoir, move this up. Then what will happen? If I put, go this, the,
if I push this reservoir up, then what will happen? You will see the gas which is
trapped inside this capillary and also the bulb will start to compress. Once it start to
compress it will give some, because it is getting compressed and compressed it will
give some pressure for which the mercury column cannot rise to this level. Now, this
height, this blank height will be calibrated in terms of the low pressure. This is the
basic principle of the McLeod gauge, right?
Again I am telling, this movable reservoir will come down, all the mercury will be
collected here, because this is a flexible tubing; I should say this is a flexible tubing,
right? So, it will come down. All the mercury will be collected in the reservoir. The
gas will be trapped here. It will come down this position. Then it will come down, gas
will come down through this region. It will enter here, then it will go up. Now, I will
slowly move this reservoir up. Then what will happen? Simply the gas will be
trapped. Gas will be getting compressed and compressed, mercury column is going
up. As I move the reservoir up, this mercury column also will go up and this
compressed gas, gas is getting compressed and compressed and it is giving a back
pressure to this mercury column.
So, after certain time we will see that even though we have taken the reservoir at this
level, this mercury cannot go up, because of the pressure created by the compressed
gas, right, so that the region where that mercury, I mean cannot reach will be
5
calibrated in terms of low pressure. This is the basic principle of the McLeod gauge,
right? So, this will be elaborately, we will speak in the next slides.
The movable reservoir is lowered until the mercury column drops below the opening
zero or O, whatever you say. The bulb B and the capillary C are then at the same
pressure as the vacuum source p. Now, one thing is important. You see this, this bulb
B is purposefully made quite large compared to the capillary. We will see the
mathematical relation, why it is made large, right? So, this is made quite large
compared to that volume of the capillary. This is the volume of the total capillary, so
height multiplied by the area of cross sections of the capillary that should be much
less compared to the volume of the bulb, right?
The movable reservoir is lowered until the mercury column drops below the opening,
O. The bulb B and the capillary C are then at the same pressure as the vacuum source
p. This is the, actually the pressure which I want to measure. This is the pressure, this
p is the pressure which I want to measure. The reservoir is subsequently raised until
the mercury fills the bulb and rises in the capillary to a point where the level in the
reference capillary is located at the zero point. What is that reference capillary, let us
look at again.
6
This is our reference capillary. So, until and unless this reservoir is, this mercury
column in this reference capillary coming to this zero level. I will move the reservoir
up and up, right? Once it comes to this zero level, I will stop moving the reservoir in
the upward direction, right? The reservoir is subsequently raised until the mercury
fills the bulb and rises in the cap, in the capillary to a point where the level in the
reference capillary is located at the zero point, right?
7
Area of the capillary is denoted by, area of cross-section of the capillary I should say,
denoted by a. So, the volume of the gas in the capillary is V c into, V subscript c that
means capillary, equal to a into y where, you see this, where y is the length of the
capillary occupied by the gas, right? This is the length of the capillary which is
occupied by the gas. What is that? I should go back again.
You see, this is the length of the capillary occupied by the gas, sorry. So, I should say
this is the length of the capillary occupied by the gas, right? This is our y and this is
the area of cross section, is a. This is area of cross section. So, obviously this volume
V c, capillary volume, this volume will be equal to a into y, clear?
8
y is the length of the capillary occupied by the gas, right? That is true. Let the volume
of the capillary, bulb and the tube down to the opening be V B. So, you see that V B
is not only the volume of the capillary, it is volume of the capillary, bulb up to the
point zero. I should, I refer back again.
You see, the volume of the capillary that means I am telling the volume of the
capillary up to this level, so this total volume I am telling is V B, subscript B, clear?
9
So, let the volume of the capillary, bulb and the tube down to the opening be V B.
If we assume the isothermal compression of the gas in the capillary we can write that
p c that means the gas which is compressed equal to p actually that which I am to
measure equal to V B capital, I mean this volume divided by the V c, the capillary
bulb this is equation number 1, right? This is, actually gas was in pressure p. Please
note it was in the pressure p. Now, I have compressed to p c, right, so which is higher
compared to p, clear? So, it should be in such a high conditions or such a high value
that I can measure by the simple manometer. That is the basic principle of the, our
McLeod gauge.
10
Now, the pressure is indicated by the cap, now the pressure indicated by the capillary
is p c minus p that compressed pressure minus the pressure actually indicate, I, which
I am interested to measure equal to y, right? If suppose, if it is in, I mean if we divide
it by d and g, so obviously I will get that also in y that is in height. This is equation
number 2. Because, obviously it should be divided by d and g, the density and the
acceleration due to gravity to get the expressions of the pressure only in height, where
we are expressing the pressure in terms of the height of the mercury column, right?
Now, combining 1 and 2, we get p equal to the, unknown pressure equal to ay square
upon V B minus ay equal to yV c divided by V B minus ay.
11
For most applications I can say the volume of the capillary is much smaller than the,
much, much smaller than the volume of the, total volume of the capillary, then bulb,
up to the leveling, opening O of the mercury. Actually, in fact you know that is
actually the, what will happen? For most applications I can take that ay is much, much
less than V B. So, I can write y, now it is clear that why I have taken V B much, much
greater than ay, purposefully. You are using a bulb that is the reason there is no need,
otherwise without bulb also system will work. We have taken so that I can write that
unknown pressure p equal to ay square upon V B.
You see, a the area of cross section of the capillary is constant, V B is the area of the
bulb, then capillary as up to the opening of the, this McLeod gauge, this is also
constant. So, a is, p is directly proportional to p square. Even though is not linear
instrument, so p is y, y square, right? With this approximation we will introduce
obviously some error, because we have approximated by this, right?
12
Commercial McLeod gauges have the capillary calibrated directly in the micrometers,
right?
Now, McLeod gauge is sensitive to the condensed vapours that may be present in a
sample, because they can condense upon the compression and invalidate equation 1.
So this, we should be careful about that. For dry gases the gauge is applicable to the
13
10 to the power minus 2 to 10 to the power plus 2 micron which is equal to 0.0013 to
13.3 Pascal.
Now pirani gauge, now this is all about the McLeod gauge. Now, pirani gauge is
basically is, is based on the thermal conductivity. Basic principle is something like
that - a gas, a heated filament will be placed inside a chamber. But, there is a heat
transfer from that chamber, from that heated surface to the outer surface, colder
surface, colder surface is a glass envelope which is actually, in which actually the,
those heated filament will be placed and as the number of molecules decreases, there
will be less heat transfer from the heated filament to the cold, I mean cold valve, the
colder surface.
So, the temperature of the heated filament, the filament will rise. So that temperature,
so obviously if the temperature rises, the resistance of the heated filament also will
rise, so that if I put that in the Wheatstone bridge, so I will get a unbalanced voltage.
That unbalanced voltage it will be calibrated in terms of the low pressure. That is
basic principle of the pirani gauge. It is, looks like this; you see, at low pressure the
effective thermal conductivity of the gases decreases with pressure. How? That is I
said that if the pressure decreases in a vessel, then what will happen?
14
Why I am getting the pressure? Kinetic theory of gases says that the, I am getting a
pressure in the vessel, because this, it has a random motions of the molecules. So, it is
heating the surface of the walls which is actually creating the pressure. If the number
of molecules decreases, also the pressure will decrease, right? Because there is less
number of molecules in the, if the less or in other way, if the low, the pressure is low,
number of molecules in the vessel also will be low, right? So, what will happen that I
have, it looks like this. We will show in the next slide that what is the basic principle?
At low pressure the effective thermal conductivity of the gas decreases with pressure.
The pirani gauge is a device that measures the pressure through the change in thermal
conductivity, conductance of the gas. An electrically heated filament is placed inside
the vacuum space. So, it will be clear, you see here.
Heat loss from the filament is dependent upon the thermal conductivity of the gas and
the filament temperature. Lower the pressure lower the thermal conductivities,
because you see the, as the pressure decreases the number of molecules also will
decrease, is not it? As the pressure decreases the number of molecules also will
decrease, right and higher the filament temperature. So, temperature of the filament
will rise. Temperature of the filament if rises, its resistance will rise, so the bridge
unbalanced current also will, unbalanced voltage will also increase; so, that
15
unbalanced voltage will be, actually we will measure, right? The resistance of the
filament is measured by a bridge circuit, right?
Now you see, this is schematic of the pirani gauge. It is not actual, it is the pirani
gauge. You see what will happen here? I have a heated filament, so I have a number
of molecules here, is not it? It will be very clear from this diagram. Molecules are
here, so what will happen you see this is connected to the vacuum space where in,
which we are interested to measure the pressure. So the filament, this molecule will
heat this. This will come in contact with this heated filament and it will take away,
due to random motion it will go and this surface is cold surface compared to the
heated filament. So, this heat will be taken away in this direction, is not it?
So, as the number of molecule decreases, as the pressure decreases number of

molecules also decrease. So, there is a less chance of taking the heat away from this
heated surface to the colder surface. So, what will happen as the, as the pressure
decreases the temperature of the heated filament will rise. If the temperature of the
heated filament rises, then what will happen? Its resistance also will increase. Now, if
I put in a bridge, if I measure this change of resistance by some, in one way, by some
means or the other, I will get some change of resistance. So, that change of resistance
16
either I can calibrate in terms of voltage or current that will be calibrated in terms of
low pressure. That is the basic principle of the pirani gauge, right?
They have a coil tungsten wire, sorry, they have a coiled tungsten wire like this one.
Several wires are there, four wires, usually they will find there are four wires. These
wires are placed in a vessel. The wires cannot, I mean suspend like this. There must
be some arrangement by which it is to be hold inside this chamber, right? Let us go
back again. So, this is the schematic of the pirani gauge.
17
Now see this is the bridge circuit which is actually used for measurement of low
pressure by pirani gauge. Now you see, for because since it is a resistance devices, so
it should be changed. It should be affected by the ambient temperature variation. If the
ambient temperature changes, so what will happen? This resistance will change. To
avoid that you will use two pirani gauges and put on a two arms of the Wheatstone
bridge, right and one particular gauge which is compensating gauge, as it, as it
happens in the case of, in the case of Wheatstone bridge we have seen, in the case of
strain measurements we have seen this is called a dummy gauge. This will be
compensating element. So, this will be evacuated at very low pressure, then it is
sealed, right and this will be, as usual the normal pirani gauges which is the low
pressure will go inside through this region.
Now, what will happen? You see, if due to environment temperature change if the,
there is any change of temperature due to any change of resistance, so there will be a
similar change of resistance here in the same, because these two gauges elements are
also exactly identical. So, the environment ambient temperature variations or bridge
unbalance due to ambient temperature variations will be totally nullified if I use this
type of configurations, right? So, this is called the compensating element or dummy
gauge, dummy pirani gauge. You see the, this is evacuated and sealed. Now, there is a
balance pot. I will explain what is the, I mean what is the function of this balance pot.
18
Initially we need it, because once you put this gauge and this is also at low pressure,
so we will see, we will balance these bridges. That means we will make the resistance
in the two arms exactly equal, then I will make it null. Now, the unknown pressure
will go inside. Initially these two gauges are at the same pressure, please note. These
two gauges that means these gauges, this dummy gauges, this dummy gauge and this
gauge were at the same pressure, right, now and I will balance, using the balance pot I
will make these two resistances equal, because these two resistances may not be
equal, exactly equal, there will be slight difference.
So, I will use this balance pot to make these two resistance in the two arms that means
arm from this region to this and this region to this, exactly equal. If it is equal, then
what will happen? So, it will be totally balanced, right? Now the unknown pressure
will be, go inside. So, I will get a bridge unbalance. That bridge unbalance will be
calibrated in terms of the low pressure. That is the basic principle of the pirani gauge,
right?
Now, the resistance element is in the form of four coiled tungsten or platinum wires as
I have shown, right? The cold surface is the glass tube. The outer surface of the glass
is a cold surface. Two identical tubes generally are connected in the bridge circuit to
avoid the change in bridge output due to change in ambient temperature. One of the
19
tubes is evacuated to a very low pressure and then sealed off while the other has a gas
admitted to it.
The balance pot is then adjusted to produce null condition, right? We have used that
null condition; I have explained that thing. To balance the bridge initially, the
pressure in the measuring element is made very small and the balance pot set for zero
output. Generally the bridge is used as a deflection rather than a null device. Any
change in pressure will cause a bridge unbalance. So, this unbalance we will calibrate
in terms of …
20
You see, this is the bridge unbalance current in microampere of a pirani gauge by a
pressure. Now see, after some time as the pressure increases there is saturations of the
bridge unbalance. There is a definite reason why it is coming saturation. This is the
current in micro ampere, 100, 200 micro ampere, right? So, as the low pressure it is
quite good response, I am, what change I am getting, but at high pressure, when the
pressure is coming beyond 1 torr, so it is getting a saturation. There is no change and
in fact if the pressure increase further there is no change in the bridge unbalance
current, right?
21
The calibration you see, the gauge must, must be calibrated against some standard. It
is not suitable for pressure below 1 micron. The upper limit is 1 torr, 133 Pascal
giving an overall range from 0.1 to 100 Pascal. For higher pressures, the thermal
conductance changes very little with the pressure; thermal conductance changes very
little with the pressure.
22
It must be noted that the heat loss from the filament is also a function of the
conduction losses to the filament supports and radiation losses to the surroundings,
right? The lower limit of the low pressure measurement by pirani gauge is the point
where these effects overshadow the conduction into the gas, right? Now, pirani gauge,
one of the greatest drawback it has the large time constant. It is not the, I mean the
measurement is immediate, so and so that is one of the greatest drawback of the pirani
gauge, right? The establishment of the thermal equilibrium may need several minutes,
because it is thermal systems, it may take several minutes before it settles down.
Now, this thermocouple gauges we will find its response time is quite fast. Even
though it is slightly different principle, but its response time, since in the, if you
compare with the pirani gauge, it is much less in the case of thermocouple gauge. The
basic principle is like this. You see, it is also based on the thermal conductivity. What
is that? I am saying that there is a molecule, gas molecules are there. So, these gas
molecules will try to take away, this is the heated filament. So we have with the, with
the some battery we are heating this filament. This is a heated filament, note it this is
our heated filament, heated plate rather I should say. So, principle is same; instead of
coil I am using the heated plate.
23
So, this will be heated. So, as the temperature, as the pressure falls, there is a less
chance the molecules to take away heat to the colder surface. This is the cold surface
glass tube. So, the temperature of this, with the same current, the temperature, the
same battery voltage what will happen? The temperature of this heated plate will
increase. There is one thermocouple welded here. So, that will measure the, the fall or
rise of temperature of this heated filament. Because if the temperature rises, what will
happen? If the pressure increases, then what will happen that it, temperature, actually
the filament will fall, filament current will fall, if the temperature, pressure increases,
is not it?
Why you see, because there is more chance of the heat, heat to be taken away, so that
is the reason the pressure will fall, the current will fall or voltage will fall, right? So,
what will happen? You see in that type of situations, I will measure the temperature,
right? I will measure the temperature and this temperatures is, are in, this millivolt is
calibrated in terms of the low pressure. That is the basic principle of the pirani, of the
thermocouple. This is also basically depends on the thermal conductivity. That means
as the pressure increases the current will decrease or the voltage will decrease,
because it will have less temperature rise if the pressure increases, right?
So, it looks like this. The response curve will be, look like this one, so like this one.
As the pressure increases in torr, the current or voltage of the, this thermocouple
voltage will fall, right? So, as the pressure decreases what will happen? So, less
number of molecules will try to heat this one. So, its temperature will rise. So, this
filament temperature, this heated plate temperature will rise, so that so you see here,
so as the pressure decreases, here you see the temperature will, the output voltage will
rise. So, that is measured by the thermocouple, you see.
24
Hot surface is a thin metal strip. I am still discussing about the thermocouple gauge.
Its temperature may be varied by varying the heated current, heating current. For a
given heating current and gas, the temperature assumed by the hot surface depends on
pressure, obviously. This temperature is measured by the thermocouple welded to the
hot surface, right? Gauge is not independent of ambient temperature change, because
there is no such bridge arrangement or something like that, so that the advantage what
we have using a dummy gauges or compensating gauges in pirani gauge, in pirani
measurement systems that is not there, absent in particular, the thermocouple gauge,
right? So, range is typically 10 to the power minus 4 to 1 torr. Range is also I should
know, I should say is not that good; that good means it cannot measure that low
pressure, right?
25
Now, third pressure gauge I should discuss is ionization gauge. Ionization gauge is
the, one of the gauges which can go to extreme low pressures. This is the lowest
pressure measuring devices. It is, the basic principle is that it will try to ionize some
gases and that ion current, that ion will be collected by some plate and that current
will be measured. Now, let us look at the, this ionization gauge. You see, this is the
chamber in which we put the gas in which we are interested to measure the pressure.
We have a cathode, we have a grid and we have a plate. Now you see, there is always
some electron will lie here, right? So, this electron will be collected by this plate, is
not it? So, electron will emit here, it will be collected by the plate. There is no doubt
about that, fine, because there is, number of electrons will be large here.
Now, if I put some gas inside what will happen? If I put some gas inside, while these
electrons by this high voltage, you see this 400 volt will try to, will accelerate from
this to this position. It will knock out some of the electrons of the outermost shell of
the gas molecule, so that gas will be ionized. If that gas is ionized, then what will
happen you see? This, since this is a, you see this is the, connected to the negative
terminal of the battery, so ion will be connected by this grid or ion collector, we have
given the name grid or ion collector, will be collected by this grid. So, I will get a ion
current.
26
At the same time what will happen you see? When these knock out electrons will be
collected by this plate, so there is an increase of current, in the electron current.
Previously suppose there is no gas inside, there is no gas inside, then what will
happen? I will get a simple current. Suppose it is total vacuum, then what will
happen? I will get a simple and the current which is due to this electron which is, I
mean continuously emitted from this cathode, it is collected by this plate.
Now, the gas is ionized. Gas is ionized means we have electrons as well as ion. Ion
will be collected by the ion collectors and electron will be again collected by this one.
But the change of electron due to the ionized gas, due to the change of electron
current, due to the ionized gas will be, will be very, very small compared to the
current which is I am getting by i i.
What does it mean? Suppose previously I am getting some, some current here due to
the, some, these electrons are coming down here, I am getting some current i, which is
usually in milliampere range. Now, how many ions will be, how many gas molecules
will be ionized? Very small, is not it? Suppose ….., suppose if I say the 10 molecules
are ionized, then what will happen? What will be the change of the electron current
for that? Nothing; negligible, is not it? But, I will get a ion current. This ion current
even though small will be measured by some micro ammeter. So, I can say that this
electron, that pressure, low pressure will be, can be measured if I measured this i i,
because increase of the electron current will be negligible for the ionised gas, is not it?
You see, the different voltages here. This is the 6 volts we have given to the cathode.
The cathode can be directly heated or can be indirectly heated. So also as it is in case
of the picture tube of our television, these are basically indirectly heated cathode,
whereas say it will be fine, thus we are giving only 20 volt here for the cathode. We
have given 400 volt because the electron is to be accelerated through this region, so
that, I mean it moves very fast with high energy and heat the gas molecule to ionize it.
Basic principle is that with high voltage 400 volts, so we will ionize some gas
molecules inside. This is the basic principles of the ionization gauge, let us look at.
27
In ionization gauges, stream of electrons are emitted from the cathode. Some of these
electrons strike gas molecules and knock out secondary electrons leaving the
molecules as positive ions, as I said. For normal operation of the gauge, the secondary
electrons are a negligible part of the total electron current. Total electron current is
very, very negligible, because electron current is already existing there, is not it? That
could be due to the electrons which is around the space charge region of the cathode,
continuous electrons are moving around. So, whether the gas is ionized or does not
ionize or not ionize, it does not matter. Always there is some electron that will be
collected by the plate, right? That is I am saying for normal operation of the gauge,
the secondary electrons are a negligible part. So, the electrons which is generated by
the ions, by, in formation of the ions will be very, very negligible. So, it is a negligible
part of the total current.
28
Therefore, electron current i e is same for all vacuum pressures. This is a great
advantage. We will see the expressions. The number of positive ions formed is
directly proportional to i i, ion current i i, I am giving the name. Rate of production of
the positive ions, ion current i i for a given gas is a direct measure of the number of
gas molecules per unit volume and thus the gas pressure, right?
29
The positive ions are attracted to a negatively charged electrode which collects them
and carries the ion current. Therefore, the pressure of the gas is proportional to the
ratio of the ion current and electron current, right, even though the electron currents is
constant. The expression, this is a very famous expression, you see this is our
relations. This is ion current and this is electron current. This is almost constant and
this is a emitter constant, S. So, as the pressure decrease, I mean the pressure
decreases, I should say the i i will also decrease. If the pressure increases at that low
pressure region you cannot give it some 100 psi there, right?
These are the pressure increases. This current will increase because more number of
molecules, so there is a more number of ionized gas, ionized gas molecules. So more,
if the more number of ions, the more current, right? If the pressure falls, the less
number of molecules, less number of chances of the gas to be ionized; less number of
molecules less the, lesser chance of gas to be, molecules to be ionized, right? So, the
current i i also will be less, right, where S is called the sensitivity of the gauge, right?
A typical value of the S for nitrogen is S equal to 20 torr to the power minus, which is
2.67 kilo Pascal to the power minus 1, right? The exact value of S must be determined
by calibration of the particular gauge, right? So, for each gauge it is not gas
independent, I mean like McLeod gauge and all those things. So, you have to measure
30
it, you have to calibrate it separately for a particular gauge, particular gas, particular
gas and particular gauge also. The value of S is a function of the tube geometry and
the type of gas. So, it depends on the, both the tube geometry as well as of the gas. So,
you have to calibrate it separately to find the value of S. The current output is usually
is linear. So, it is a linear instrument. So, p is directly proportional to ion current, low
pressure is directly proportional to the, proportional to the ion current.
Now, what is it? There is some disadvantage of this particular gauge, ionization
gauge.
You see, the filament may burn out if exposed to air while hot, decomposition of
some gases by hot filament and contaminations of the measured gas by the gases
forced out of the hot filament. So there is a, this chance is there. So, because at the hot
filament what will happen? It will decompose. It will, I mean some decomposition of
the gases hot filament and contamination of the measured gas, but the gas is forced
out of the hot filament, so that will create, contaminate the gas actually where, which I
am measuring. In some situation this is not allowed. So, hot cathode ionization gauges
covers the range of 10 to the power minus 10 to 1 torr.
31
Now, Knudsen gauge is another instrument. Even though it has very scientific
applications, but it is, do not have any electrical output. You have seen the pirani
gauge, thermocouple gauge, ionization gauge, has the direct electrical output, whereas
we will find in the case of Knudsen gauge, like your McLeod gauge it is basically
monitoring instruments. There is no, actually they are electrical output. Knudsen
gauge you will find, it is rather difficult, I mean you will find there is a, in the case of
McLeod gauge I am getting a manometric reading, whereas in the case of Knudsen
gauge a spot of light is moving on a scale which is calibrated in terms of low pressure.
Let us look at Knudsen gauge. It is relatively, that is another advantage which is not
there in the other gauge like pirani and the ionization. It is relatively insensitive to gas
composition and thus gives promise to the development into a standard for pressure
too low for McLeod gauge. McLeod gauge is not, as I discussed even though we are
using as a pressure of standard, dimensions, pressures can be expressed in the
dimensions of the gauge, but it is not suitable for the pressure below 1 torr. So, in that
type of situation, this Knudsen gauge can be used as a pressure standard.
32
You see, this is the diagrams of the Knudsen gauge. We have, you see there is a gas
chamber here. The gas which we are measuring p i, it is inside, is going inside. There
is two fixed plates. You see, this is one fixed plate. You see, this is one fixed plate and
this one fixed plate. In between two fixed plates we have a movable plate. This is our
movable plate, right and mirror is installed here. A light spot is moving inside this,
falling on the mirror and this is getting deflected and coming to this and it is putting
on a scale.
This scale is calibrated in terms of, now what will happen? It is like this one. There is,
there is two plates, heated plate is there. Two heated plates like this one and in
between this heated plates there is a movable plate, right? Movable plate is inside,
right, like this one. Now see, what will happen? The, now the distance between this
heated plate and this should be less than the mean free path of the gas molecule. This
is the one condition we have in the case of Knudsen gauge, right? Let us look at.
33
The fixed plates are heated to absolute temperature T f which temperature must be
measured and a spring restrained movable vane where temperature T v also must be
known. The space between the fixed and the movable plates must be less than the
mean free path of the gas whose pressure is being measured.
Now, kinetic theory of the gases shows that the gas molecules rebound from the, it is
not from, I am sorry, this is from, from the heated plates with greater momentum than
34
from the cooler movable vane thus gives a, giving a net force on the movable vane
which is measured by the deflection of the spring suspension. What is that actually?
You see, see that there is a gas molecule inside, is not it? It is everywhere. So, this is a
fixed heated plate. With a coil we are heating this and the gas molecules are getting
rebounced from this one. So, we are assuming that the gas which is, gas molecule
which is rebounced from this heated plate to a greater momentum than those which
are reflected from the movable plates which is colder one. So, what will happen?
There is a moment, a couple will act on this movable plate. So, it will rotate. So, the
light is there, so it will fall on the mirror. Mirror will also rotate. The light spot will
rotate on this scale. This is the basic principle.
35
Now, kinetic theory of gas molecules rebound from the heated plates with greater
momentum than from the cooler vane, movable vane thus giving a net force on the
movable vane which is measured by the deflection of the spring suspension.
The analysis shows that the force is directly proportional to the pressure for a given T
f and T v of the form, p i equal to K F upon under the square root T f upon T v minus
1, right, where F is a force and K is a constant. The range of the Knudsen gauge is 10
36
to the power minus 8 to 10 to the power minus 2 torr. This is a typical range. So, we
can see that pressure can be measured, low pressure can be measured by this
particular type of instrument, right?
Now, let us solve a problem on a diaphragm gauge. In the following diagram, four
strain gauges are placed over a diaphragm. We have seen this one already; four strain
we have seen this already, rosettes this types of things are there. So, two strain gauges
2 and 4 which will actually sense the actual strain and two strain gauges at the centre
which will sense the tangential strain, right?
37
The problem is, for the above arrangement draw the signal conditioning circuit, first
of all. Find the sensitivity of the circuit in millivolt per Pascal. Calculate the natural
frequency in vacuum. Find the maximum allowable pressure for a non linearity of
2%. Find the full scale output, right?
Data given are that r t, the radius r t, where is r t, please note.
38
This r t is just the radius, where the, our circle on which this is the tangential strain
gauges are placed, where r r is the radius of the circle on which, on which these two
strain gauges which will sense the radial strain, this will sense the tangential strain,
this will sense the radial strain, are installed, right, right? Let us go back again;
already discussed.
39
So, this is our data given. r t is the point, zero point zero, 0.015 meter. r r equal to
0.064 meter, diameter of the diaphragm is 15 meter and thickness of the diaphragm is
t, 1.28 millimeter. In the case of rosette we will find this is quite valid, because
rosettes even though it is large in size, because the diameter will be, the circle on
which it is placed, the tangential and the radial will be much less than the diameter of
the, you see, almost you see it is 0.15 that is 0.075, 0,075, it is 0.064, correct. Now
you see, thickness of the diaphragm t is 1.28 millimeter. Poisson’s ratio nu equal to
0.26, battery voltage E ex of 5 volt, gauge factor is advance we are using, it is 2 and
gauge resistance we are saying 120 ohm.
Density of the diaphragm material row is 6 into 10 to the power 3 kg per meter cube
and the modulus of elasticity for the diaphragm material is 2.1 into 10 to the power 11
Pascal, right? So, let us look at the solution.
40
This is gauge 1, this is gauge 2, this is gauge 3, this is gauge 4. We have output
voltage here, right, E ex, right? Now, the tangential stress are given by, we take a new
page.
S t equal to 3 PR square nu 8 t square 1 by nu plus 1 minus 3 1 upon nu small r by

capital R whole square, like this one and radial stress will be given by S r equal to 3
PR square 8 t square, square and biaxial stresses we know that also; expressions we
41
know, we will give after some time. Now, let us represent this we assume the 3 PR
square nu by 8 t square equal to, we are writing capital A. Now this, say this is A and
R equal to 0.075 meter, right?
So I can say, sorry, S t at r equal to r t is equal to 4.572 A, right? S r r equal to r t

equal to 4.344 A.
42
Now, again S r at r equal to r r equal to minus 4.285 A and S t at r equal to r r minus
0.139 A. Now, the diaphragm is in the state of biaxial stress and both the radial and
tangential stress continue to the radial and tangential strain.
So, epsilon t we know S t minus nu S r by E. So, epsilon t at r equal to r t, I should say

3.4426 A by E, right?
43
Radial strain epsilon r equal to S r minus nu S t by E. So, epsilon r at r equal to r r
minus 4.249 A by E. Let E represented by, suppose this is represented by B, then what
will happen?
You know that epsilon t will be given by 3.4426 B and epsilon r equal to minus 4.249
B. Now, the gauge factor is equal to 2 and with the application of the pressure, so I
can say that the change of resistance R 1 dash will be R 3 dash equal to R naught,
initial value of the resistance 1 plus lambda epsilon t and R 2 dash R 4 dash R naught
1 plus lambda into epsilon r, right?
44
So, the unbalanced voltage will be given by e naught equal to R 2 dash by R 1 dash
plus R 2 dash minus R 3 dash upon R 4 dash plus R 3 dash, excitation E ex. So, this
will lead to, substituting all these values of R 1 dash, R 2 dash, R 3 dash, R 4 dash, I
will get, e naught by E ex equal to minus 1 upon 1 minus 0.8064 B 7.6916 B. Now,
substituting all the values, we get that B equal to, if I substitute all the values of B,
B equal to 1.6 into 10 to the power of minus 9 into p, unknown pressure, right?
45
So, sorry, so I will get, e naught by E ex equal to e naught by E ex equal to or I should
directly write instead of e naught by E ex, I can directly write that e naught by p equal
to minus 7.6916 into 1.6 10 to the power minus 9 E ex upon 1. So, neglecting the term
that means this small value of, it is almost I should say, almost equal to this. So,
sensitivity, I will find sensitivity that is in millivolt per Pascal will be given by e
naught by p 7.6916 into 1.6 into 5 into 10 to the power minus 6. So, it will be 0.615
10 to the power minus 4 millivolt per Pascal.
Now, for the second part you see that we know the expressions that f is natural
frequency 10.21 pi R square, radius of the diaphragm E t square upon 12 rho 1 minus
nu square. This is in Hertz. So, if I put all the values, I will get 1308 Hertz, right?
46
Now, third problem what we said that the, you know the p can be related as 16 Et
power 4, the third part of your problem 3R to the power 4 1 minus nu square y c by t
plus 0.48 y c by t cube. This is our non-linear term, is not it? So, non-linearity, non-
linearity is introduced due to second terms. So, it is given 0.488 y c by t whole cube 2
by 100, right? So, this will give me y c by t equal to 0.345.
47
So, the maximum available pressure from this we can write is that p is, if we can
compute this thing, we will get p equal to p equal to 371.96 kilo Pascal, right?
Sensitivity, S will be given by 615 into 10 to the power minus 4 millivolt per Pascal.
So, the maximum available pressure is this one and the full scale output will be given
by e r e naught 615 10 to the power minus 4 into 371. 96 into 10 to the power 3 equal
to 22.87 millivolt, right? This is your answer. This ends the lesson 19 of Industrial
Instrumentation.
48
Prof. A. Barua
Indian institute of technology - Kharagpur
Lecture - 20
pH and Viscosity Measurement
Welcome to the lesson 20 of Industrial Instrumentation. In this lesson, we will discuss

basically the analytical instrumentation. We will find that the two most important
physical, I mean process parameters are pH and viscosity and as you know, the pH is
very much necessary for many industrial application as well as bio-reaction and
biomedical applications. Because, in some applications we will find that we have to
monitor the pH of the blood and in some bio-reactions, bio-reactions, we have to
purposefully make a particular pH value to have a good growth and there are numerous
applications of measurements of pH in chemical industries. So, keeping all this in mind,
in this particular lesson, we will discuss pH and viscosity measurements and different
techniques of the viscosity measurements, pH particular probe and pH amplifier what are
the, should be the typical characteristics of the pH amplifier, the, all these things will be
discussed in this particular lesson.
1
Contents are: the definition of pH, why is it necessary to measure pH, pH probe,
characteristics of pH amplifier. Now, viscosity measurements - basically we will discuss
two types of viscosity measurements. One is capillary method, another is rotating drum
method and some industrial viscosity meter or viscosimeter. In industry it is called the
viscosimeter. So, rotating drum viscosity measurements, then we will discuss the
capillary flow method. Also we discuss industrial viscosimeter; it is called Saybolt
viscosimeter, right?
So, definition of pH you see, pH is the measurement of the number of grams of hydrogen
ions per liter of solution, right? Mathematically it is expressed as negative logarithmic of
hydrogen ion concentration, right? Mathematically can be expressed like this - pH equal
to minus log base 10 of hydrogen ion concentration. This is equation number 1, where H
plus is the hydrogen ion concentration in the solution.
2
The value of pH can range from 0 to 14, where 0 describes the extreme acidity and 14
describes the extreme alkalinity, right? So, 0 describes the extreme acidity and 14
describes the extreme alkalinity. This is the range of pH. Either it will be 0 to 14 and if it
is neutral, neither acidic nor alkaline, then in that case the pH will be 7 like distilled water
has a pH, supposed to have a pH of 7. Now, pure water or distilled water has a pH of 7.
As pH values change by 1 unit, then the acidity or alkalinity values change by a factor of
10. The pH scale is shown in Figure 1, you see.
3
Next figure we will find, this is the pH scale we have seen. You see here, so this is, you
see zero strong acid and this is a strong base. Base and alkali, as you know it is same
thing and pH 7 is neutral, as I told you the distilled water. Strong acid like your, our acids
are, I mean inorganic acids are very, very strong acid. In that case pH will be very close
to 0, if it is not zero, right? In, similarly alkaline like, I mean any alkalis like sodium
hydroxide and other types of solutions will give you a very strong base and as I told you
distilled water should have a pH of 7.
Here in this scale you see here, in this scale we will show you the hydrogen ions. It starts
from 1 into 10 to the power minus 4 and in this end we are showing the -OH ions. When
the -OH ions are strong, we will call it strong acid. When the hydrogen ions are strong we
were calling it, sorry and it is other way. When the ions are becoming weak, then it is
becoming the strong acid. When H ions are strong in this way, because it is decreasing in
this directions, it is say strong acid, right? So, it is 1 here, so it is 10 to the power minus
14 here.
4
Now, why is the pH measurement is important? That is in the initial stage I told you once
and let us, let me discuss this in details. Measurement of pH of a solution is necessary to
judge whether the liquid is acidic, alkaline or neutral, right? It is necessary in many
applications, because in some environments we will find that the, we have to make those
environments neutral. So that tap water, suppose the drinking water it should be neutral;
so, in that case it should be neither acidic nor alkaline, right? In some reactions like,
suppose in the bio-reaction it will be slightly alkaline, so there also the measurement of
pH is necessary.
In biomedical applications we will find that the measurement of pH is necessary, because

we, we can monitor, by monitoring the pH we can tell some of the particular diseases
also. The application areas are food, beverage industry, aquarium, agriculture and
gardening and colour photography, etc. There are numerous applications, even more
applications like the industrial application as I told you earlier.
5
The measurement of pH of liquid is necessary in many condensate systems as a check on

corrosion, in pharmaceuticals and drug production for maximum yield, as I told you in
the bio-reactors we always make, I mean suppose the yeast culture, say always make the
pH slightly alkaline, right? So, that type of situations we call it, I mean the measurement
of pH is also necessary. In food manufacture, because we have seen that if you make it
slightly alkaline, so we will find the growth will be better, right? So, that reason
measurement of pH, continuous monitoring of pH is necessary for any bio-reactions.
In food manufacture to reduce the spoilage and improves the taste and in innumerable
chemical processes, you will find the applications of measurement of pH. The continuous
monitoring of blood pH is essential for the proper treatment of the patients suffering from
metabolic and respiratory problems, because you know, the blood has a pH. It is, I mean
varies 7.35 to 7.4 that is 7.35 to 7.5. So, if this is for a normal healthy person, now if this,
if this goes out of this range that means the patient has some problems. We can monitor
those and appropriate actions can be taken, appropriate drugs to be given to the particular
patient, so that the pH value comes to the actual, normal healthy person’s range of 7.35 to
7.4 or 7.5.
6
The pH values of some common substances are given in Table 1; let us look at. You see,
this is the pH, the values of some common substances. Battery acid which is supposed to
be a very strong acid, sulphuric acid, you see it is almost close to 0, .3, very strong acid.
Lemon juice – 2.3, because these are all organic acid, is not it? Vinegar – 2.9, orange
juice, organic acid – 4.3. So, more and more, you see it is increasing in value of pH,
right?
7
Then we have boric acid - 5, corn - 6.2, milk – 6.7, slightly acidic that means, distilled
water 7.0. Now, it is, if it goes, crosses zero, then you will find, crosses 7 that means it
has become acidic, I mean alkaline. Blood – 7.5; as I told you, its range lies from 7.35 to
7.4 or 7.5. So, we have given the value 7.5, right? Sea water is alkaline; it is 8, baking
soda – 8.4, milk of magnesia – 10.3, ammonia – 11.4. It is quite highly alkaline, you can
say. Bleach is even higher; it is 12.6, right? This is the range of the pH values for
different commonly used substances.
Now, pH probe – the, because pH is, is usually you see that is the, you need a probe and
details of the probe needs to be, I mean to be discussed. The traditional method of
indicating hydrogen ion concentration is the use of litmus paper. This we have seen in the
chemistry laboratory. The people use the litmus paper to know whether the solution is
acidic or alkaline, but this, this method is very much qualitative method of measurement.
It will not give you how much, how much strong the, it is acidic or how much strong is, I
mean it is alkaline.
So, some measurement is necessary, but litmus paper, obviously people are using over
the years, over the decades and it gives actually some indications whether the pH value
8
is, either the liquid is, either I mean acidic or alkaline. So, traditional method of
indicating hydrogen ion concentration is the use of litmus paper. That does not give exact
value. It is qualitative measurement, as I told you earlier. So, we need some quantitative
measurements, which will give you the exact value of the pH.
Functions of the electrode - what is the function of this electrode? The pH measurement
method requires that an electrode be immersed in a solution, obviously. An electrolytic
potential is produced or developed at the electrode which forms an electrolyte,
electrolytic half cell. This is called the measuring cell. There are two cells, we will find.
One is reference cell, another is measuring cell, right? Now, each of the cells we are
calling it half cell. There are combined pH probe also that will be discussed later on.
Now, we will first discuss the measuring cells and reference cells. What is the content of
this? A second electrode is required to provide a standard potential and to complete the
cell. It is the reference cell, right?
9
The algebraic sum of the potentials of the two half cells is proportional to the
concentration of the hydrogen ions in the solution; actually we want that. The two
voltages will be algebraically, I mean added and we will get the value of the pH, right?
The measuring glass electrode is shown in Figure 2. So, we will show you the figure.
You see, the glass electrode, we will first discuss about the glass electrode, then we will
go to the figure. It operates on the principle that a potential is observed between two
solutions of different hydrogen ion concentrations when they are separated by a thin glass
valve, wall this should be, I am sorry, it should be a wall, wall.
This potential is function of two concentrations. A buffer solution is contained in the

permanently sealed glass electrode which is surrounded by the solution whose pH is
being measured, right? A buffer solution has a constant hydrogen ion concentration and
the potential at the electrode therefore depends on the hydrogen ion concentration of the
measured solution. You see this is our measuring glass electrode.
10
You see here, everything is very much clear. You can see it is the electrode lead. This is
the glass envelope. We have a platinum wire and we have a buffer solution here, right?
We have a buffer solution here for pH measurements.
Now, the liquid junction is provided by a small hole in the electrode over which a ground
glass cap is placed. Thus the potassium chloride slowly escapes from the electrode into
11
the measured liquid and the potential at the reference electrode is also a constant. pH is a
temperature sensitive process, I mean process parameter. So it, you must compensate for
any changes, because if the temperature changes, pH value also will change. So, the
output voltage of the meter also will change, so that you must make some compensation
for this temperature change.
Consequently, a temperature error must be compensated, any temperature error must be

compensated. A simple RTD is incorporated to make this compensation. The pH
electrode can be placed in the vessel or in the liquid pipe line. Either it can be placed in
the vessel in which we have to take the sample of the liquid or in the liquid pipe line
where the liquid is flowing that can be also be used to make the measurement, right? The
glass calomel electrode can be used as it, at liquid temperature between 1 degree
centigrade and 100 degree centigrade.
12
You see, the reference electrode; the calomel electrode is in common used as a reference
electrode. The calomel is mercury and mercurous chloride is contained in the inner tube
and covers a platinum wire. A saturated solution of potassium chloride is in contact with
the measured solution that surrounds the reference electrode.
13
You see, this is the calomel reference electrode. We have a ground glass joint here. This
is put on some solution, so that the, it should remain, I mean it should not dry up. We
have a, you see Hg and HgCl solution and this is the glass envelope and this electrode
lead is coming and this is the KCl solution, right?
Now combined pH probe, because you see, using two probes it is, obviously it is
problematic in many cases. So, we want something very handy. So that type of electrode
is the combined electrode. So, we will find most of the application nowadays, I mean in
fact in our country Phillips is making that type of probe that which have, but other
companies are also making. Those are separate electrode, two separate electrodes we
have to put, otherwise you can have a one combined electrode that means both reference
electrodes and the measuring electrodes are in the same system. It is very handy, it is very
useful and ease of measurement will be, also be there. The most common forms of pH
measurement is the combined pH probe which is shown in Figure 4.
14
You see, we will show the figure. This is you see the combined electrode, right? You see
it is a connecting cable and you see here it has pH sensitive glass membranes. This is not
responsive to glass body. You see, this is the non-responsive glass body. We have written
here non-responsive glass, glass body, right? So there will, there is no interactions with
the, if we, even if we have a liquid up to this, suppose we have a liquid up to this in
which, suppose in a beaker we put this, in a beaker we put this electrode, so obviously
what will happen?
You will find that this will not act, but this is the pH sensitive, you see this glass is pH
sensitive, right? It will react with the, so liquid, it will react with the liquid outside, right,
so that, because it is filled with the reference solution, so I will get the measurements and
here you see in this case, the reference cell, this is the reference electrode and it is put on
the reference solution, right? This is the, this is totally separate. This unit is totally closed.
It is not, so only the electrodes are coming out. You see the, here, here the electrodes are
that, this lines is going out. Similarly these lines also.
This is the reference electrode line. This will be going out. This is the combined
electrode. We will see mostly this is the thing people are now using. This is the same
15
thing, excuse me, we will find here that there is also reference electrode. The reference
solutions like KCl solutions are there, potassium chloride solution and you see, this is all
measuring electrode. The same thing is there that platinum bottom or you have, we have
seen in the separated electrode, it is already there but it is combining too, combining bond
system.
The combined glass probe consists of a glass probe containing two electrodes, a
measuring one and a reference one, separated by a solid glass partition. The reference
electrode is a screened electrode, immersed in a buffer solution which provides a stable
reference emf and it is usually zero volt, right? Reference electrode is a screened
electrode, immersed in a buffer solution which provides a stable reference emf and it is
usually zero volt. Let us look at again.
16
You see here, there is a buffer solution here. You can put this is a buffer solution. This is
filled up with a solution, sorry, you see there, here we have a solution, right that is a
buffer, I mean solution, right? So, there is a reference electrode, fine. So, reference
electrode is a screened electrode and immersed in a buffer solution which provides a
stable reference emf and it is usually zero volt, right? No problem, fine.
17
The tip of the measuring electrode is surrounded by a pH sensitive glass. Again let us go
back. We can see, it will be more clear.
You see, this is the tip of the electrode, this is the measuring electrode. This is the
reference electrode. Let me again go back again like this one. So, this is our measuring
electrode. It is coming here. This is all those platinum business. Here, here you can see
here this is the reference electrode. This you see, is surrounded by a pH sensitive glass
membrane, clear?
18
The tip of the measuring electrode is surrounded by a pH sensitive glass membrane at the
lower most end of the probe which permits the diffusion of ions according to the
hydrogen ion concentration in the fluid outside the probe, clear? Outside the probe we
have some liquids, right?
19
Right, you see, outside the liquids. So, what will happen you see here? So, we have a
liquid here. Obviously, as I told you here I have a liquid, but there will be no reactions.
Liquid, this liquid cannot go inside. Here you will see there is a diffusion of the hydrogen
ions. So, it will react with the, these probes, we will get some output, clear?
So, tip of the measuring electrode is surrounded by the pH sensitive glass membrane at
the lower most end of the probe and which permits the diffusions of ions according to the
hydrogen ion concentrations in the fluid outside the probe. Actually there should be a full
stop here, anyway. The measuring electrode therefore generates an emf proportional to
pH, because other thing is constant. So, because it is electrical circuit I need at least two
lead that is the reason reference electrode is also necessary, right? So, measuring
electrode therefore generates an emf which is proportional to the pH of the liquid in
which that measuring electrode has or measuring tip has been inserted, clear? Because,
the measuring tip is to be inserted there, so that I will get some emf which is proportional
to the pH of that solution and the reference electrode, so we got two terminals, we will
get a voltage that is measured and that will be converted to the current later on.
20
The characteristics of glass electrode are dependent on the ambient temperatures with
both zero drift and sensitivity drift. This already we have discussed. Because it is
sensitive to the temperature, so we have to take proper care by RTD or any other
temperature sensitive devices to make the correction. Thus the temperate compensation is
needed. It is normally achieved through calibrating the system output before use by
immersing the probe in solution at a reference pH values, right?
That is the very interesting thing, because I can have a reference pH value. Suppose if, I
mean solutions of known pH I will put, insert, so I can make necessary corrections or I
can have a temperature compensation scheme also by which I can make the correction,
right? This is some analytical, I mean that means we have some because normal, I know
that it is, pH is suppose 4, now it is coming 4.2. Accordingly I can make the corrections,
because this 4.2 into the some temperature, temperate change, so that can be corrected for
any unknown measuring value of the pH, right?
21
Now, range of pH measurements you see, the typical range, I mean for measurement is 0
to 12, because it cannot read, it is very difficult to measure beyond 12. It will become so
much of alkaline, it is very difficult. The electrode contaminations at high alkaline liquid
become a serious problem. Electrode can be contaminated at high alkaline liquid and
becomes a serious problem. The glass starts to dissolve in acid solution containing
fluoride and this represents a further limitation of the …. use. That is for the lower range.
Even though I am saying 0, but we cannot measure the pH value of zero, which is very
difficult, right? It is highly acidic.
22
The voltage output of the pH probe you see, because we will get a voltage, the net
potential in the pH probe is given by this expressions, right? V naught equal to minus
2.30 RT by F log of C H by C R, equation number 2, where R is the universal gas
constant which is 8314 joule per kg per mole per Kelvin, T is the absolute temperature in
Kelvin, F is the Faraday’s constant; it is 9.67, 9.647 into 10 to the power 7 coulomb per
kg mole and C H is the hydrogen ion concentration in solution and C R is the
concentration in glass electrode. So, this will cancel, I mean these two units will be
cancelled out and that is not, that is the reason it is not mentioned, right? So, we will get
that voltage output. If I plug in all the values I will get, which is 1.0 for 1 normal HCL.
23
Inserting all the values, the output voltage can be expressed as V naught equal to minus
1.98 into 10 to the power minus 4 T in pH units. So, pH units we can see is very small,
right? It is sensitive to temperature, so that is the reason T unit is there. Now pH
amplifier, you see the pH amplifier should have very typical characteristics. The pH,
since we have glass electrodes, we can consider as a voltage source. It is like this one you
see that we can consider as a voltage source.
24
That means I have a voltage source and with a high source impedance, right? So, any
current, if I try to draw some current through this circuit, because when we measure with
this resistance by some meter, it has some input impedance, is not it? So, a current will
start to flow if you connect this one. Suppose this is a meter, then what will happen?
There will be a large drop across this. So, I won’t get the proper measurements. So, the
pH amplifier should have the basic characteristics of the pH amplifier that it should, first
of all it is a DC amplifier and amplifier means actually it is amplifying and also the,
measuring the voltage output of the pH probe which is coming from the reference and the
measuring electrode. Because measuring electrode is one point measuring, I mean two
electrodes are there. If I have to take separately, two electrodes are there, so this voltage
is to be measured between this point and this point, right?
If it is combined also it is same thing, because this is also two separate and two, as I have
seen the two electrode leads are coming out even in the combined electrodes. So I have,
we have to measure these two, this voltage between these two electrodes and since we
have a glass, so it is, I mean source impedance of this, this we can consider as a voltage
source and because of it, the glass, it has a very high internal source resistance, right?
That is the reason we cannot get any voltage. You see, this is very common. You see
those, we are most familiar with the our dry cell batteries, because I am giving the
examples of dry cell batteries because it is almost all of us we have seen that when the
dry cell is almost dead, you can see almost, suppose it is, it is supposed to be for normal
healthy dry cell it is 1.5.
Now after use over the, suppose days and months we will find the dry cell voltage drop,
but we will look at, very interestingly that is the voltage does not drop much and
supposedly if you want to draw some current, you will find that you are not getting any
current. The reason is due to chemical reactions inside, the dry cells internal impedance
has increased so much that you cannot draw any current from that source, right, that is the
voltage source, is not it? A dry cell is a, is a voltage source of 1.5 volt.
25
Now, initially this resistance is so small that is you cannot measure a current. You cannot
measure a short circuit current. It will be very high, right? But, as the time goes there is a
chemical reaction and we will find this internal impedance initially it is very, very small
and initially, after use you will find this voltage, this internal resistance is getting
increased. At some level we will find that you cannot take out any current from that
voltage source, right, because internal impedance is so high. Similar the case, in the case
I mean in the case of pH probe, because of its internal, I mean because of the glass
probes, internal impedance is in the order of several mega ohms.
So, I will need some amplifier. First of all I have to amplify that signal which is very
small, DC signal and I have to amplify the signal as well as the main characteristics of
this amplifier should have a very, very high input impedance. That means looking from
this terminal, if I look at, looking from this terminal the impedance the, the amplifier, the
impedance the meter on the amplifier will offer should be extremely high and atleast 100
times higher than the source impedance of the, 100 times higher than the source
impedance of the pH probe, clear? So, let us look at the details.
26
The pH probe has source impedance of the order of 100 mega ohm, because the electrode
is made of glass, right, fine? A very high input impedance DC amplifier is a, is a good
choice. The input impedance of the amplifier should be of the order of 10 to the power
10, 10 to the power 12 ohm, you can think of. It is, I mean 10 giga ohm to 1000 giga ohm
I want, right? This much of higher input impedance is necessary to measure the pH.
How will I achieve that high value of the, because you know that we have studied so
many amplifiers and this that, we know that I mean if we use a, suppose if is, you can
simply use, those who are from electrolysis, you must be knowing that if I use a common
collector amplifier, a single thyristor common collector amplifier, I will get a higher
impedance than the common emitter or common base amplifiers. We can have instant, I
mean we can have the op amp also. Suppose in the unity gain feedback mode which is
obviously used, I mean in many applications we have seen in the instrumentation it is
used as a buffer amplifier, right or I can use cascade amplifier also which will give you
the higher input impedance.
But this input impedance is not that high that I can measure the pH voltage, right? So, we
need some other applications like MOS devices, FET device at least or MOS device. In
FET, in the case of FET, as you know the input impedance is very high. In the case of
MOSFET input impedance is even higher. So, we should have some devices of some
circuit by which I can have, achieve this much of input impedance, right? Let us look at
details of that.
27
An instrumentation amplifier should be made of op amp which has a high input

impedance differential stage, right? So, instrumentation amplifier which is to be made of
op amp, so instrumentation amplifier is know, you know there is input impedance is
higher than the op amp and its common mode ejection is also better. So, keeping all, this
means you can have instrumentation, but first of all I need op amp, very high quality op
amp, which is high gain as well as high input impedance, extremely high input
impedance, otherwise you cannot measure, please note.
Op amp should be, preferably be made of MOSFET input differential stage and it is
implemented in BiCMOS. BiCMOS, as you know, CMOS is complementary metal oxide
semiconductor. It is a bipolar complementary metal oxide semiconductor process by
which we have, in the same wafer we have, should have both bipolar devices as well as
MOS devices and this type of, I mean amplifier will give you a proper impedance which
we can, I mean utilized, I mean that amplifier can be utilized to measure the pH voltage.
This will lead to an input impedance of 10 to the power 12 ohm. You can see 10 to the
power 12 means almost 1000 giga ohm, right? Now, analog devices AD 5449 op amp is
such a device which will, can be utilized, safely utilized for measurements of pH value,
right?
28
Now this all, we talked about this pH value. Now, let us come to the, another analytical
measurement, which is called the viscosity. Viscosity is also very important, because in
many applications we will find that I mean the quality of the food product also depends, I
mean viscous, quality of the chocolates depends and depends on the raw materials. So,
people try to measure the viscosity of that suppose the ….. in which we are producing in
the industry they also measure the quality, because basically it is for the quality control,
in the process industry we will find basically for the quality control.
In food industry, this actually gives you the good taste. In some there is restriction of the
pipe whether if it is very viscous then there will be a large pressure drop. I need a large
pumping cost will, pumping cost will be very, very high. So, in that type of situation,
continuous monitoring of viscosity is very much necessary, right? Let us look at.
Now, absolute viscosity is defined as tau equal to mu dv by dy. This is equation number 3
in continuation of the p this that is the reason I have given the equation number 3, where
tau is the shear stress between the fluid layers in laminar flow, it is in Newton per meter
square. Mu is the dynamic viscosity in Newton second per meter square and dv by dy is a
normal velocity gradient as indicated in Figure 5. I will show the Figure 5.
29
You see this is the relation of the viscosity gradient and the fluid shear. So, this is the tau
equal to mu dv by dy, right? This is, sorry, this is v I think and this is y. So, this will be v
I think. So, this will be v and this will be y, right?
Let me see, this is the v y, so it will be, I am sorry, so if I go back, so, so this will be v
and this will be y as it is no problem in that, right, right?
30
So, mu k or kinematic viscosity we defined as mu by rho. Another term we defined also

that is the absolute viscosity or dynamic viscosity. The absolute viscosity, this is absolute
viscosity, people sometimes this is a, I am sorry, so units for kinematic viscosity would
become Newton second meter per kg, right? That means we have to, I mean divide the
absolute viscosity or dynamic viscosity. Mu is the absolute viscosity or dynamic viscosity
that is to be divided by the density of the liquid, you will get the kinematic viscosity,
right? These units obviously will be different.
The, some of the more common units are given below along with the appropriate
conversion factors. What are the different units? Even though, as I told you that those SI
unit is not very popular in industry for expressing the viscosity, let us look at all the
different units.
31
Dynamic viscosity, you see dynamic viscosity or absolute viscosity is given as 1 Newton
second per meter square equal to 10 Pascal, sorry 10 poise equal to 1000 centipoises. It is
defined as cP then equal to 1 kg per meter second, meter second is in the denominator.
1 poise is equal to 100 centipoises equal to 1 dyn second per centimeter square which is
equal to 0.1 Newton second per meter square, right, which is 0.1 kg per meter second and
kinematic viscosity is given by 1 meter square per second equal to 10 to the power
stokes. This is also 1 stoke equal to 1 centimeter square per second equal to 100
centistokes. Most commonly used, the unit you will find in the process industry is called
the, see the industries we will find the cP or centipoises, right which is equal to 10 to the
power minus 4 meter square by second.
32
Now, velocity distribution between large parallel plates, we are showing that what is the
velocity distribution? Suppose this, the two plates, two parallel plates are there and plates
are supposed to be stationary. Now, suddenly suppose if these plates, they start to move
in this direction, these plates start to move in this directions, plate is moving in this
direction, you see this plate is moving in this direction, then what will happen? You see,
the liquid which is in contact with this one will move in a faster direction and liquid
which is contact will have no velocity. So, all the liquids have slowly, you will find that
the, the velocity distributions will look like this one. This is highest, this is highest, this is
low, this is low and which is liquid in contact with this plate, this stationary plate, this is a
stationary plate will be stationary, right and the liquids which is in, corresponds to this,
this, all this velocity we are measuring with respect to the stationary plates, right? At the
top this will be the highest.
33
The two most common methods of viscosity measurement are the rotating concentric
cylinder method and the capillary flow method. These are the two most important method
of measurement. This is the way direct electrical output you will get. Now, rotating
concentric cylinder method, let us first discuss this. Then we will come, come to the
capillary flow method. Consider the parallel plates as shown in the Figure 6. We have
shown in the parallel plates, as shown in the Figure 6. One plate is stationary. That means
lower plate is stationary and the other moves with the constant velocity v.
34
The velocity profile for the fluid between the two plates is a straight line and the velocity
gradient is dv by dy equal to v by b.
Let us go back. You see, this is the separation between the two plates. This is b. This is
stationary plate, this is the moving plate. You see, this is the straight line, right?
35
A velocity profile for the fluid between the two plates is a straight line and the velocity
gradient is dv by dy equal to v by b, equation number 4. The principle could be used to
measure the viscosity by measuring the force required to maintain the moving plate at the
constant velocity v. The principle could be used to measure the viscosity by measuring
the force required to maintain the moving plate at the constant velocity v.
36
The parallel plate system is difficult to construct. However the, because you see this type
of parallel plates are very difficult. Two plates are moving parallel. It is, just because I
can have some liquid inside moving in between two parallel plates and the plates are
moving, one plate is having it is very difficult to realize. Instead we can make two
concentric cylinders, right? One cylinder, I mean, I mean stationary, other cylinder is
moving, so that we can realize the parallel plates phenomena there also. That will be
easier for us to make, mechanically to make the construction, right?
The parallel plate system is difficult to construct. However, the parallel plate, parallel flat
plate situations will be realized with rotating concentric cylinders as shown in figure. The
inner cylinder is stationary and attached to the torque measuring devices, while the outer
cylinder is driven by a motor at a constant angular velocity omega. Let us look at.
You see this is our, right you see the inner cylinder is stationary and this outer cylinder
with a shaft is connected to motor. It is rotating at a constant, constant angular velocity of
omega, right? This is shaft. Now this, this torque, this is the torque on this shaft will be
measured, right in a, in a stationary it has a length of L. This is, this is a separation
between the, bottom separation between the two cylinders, right? This is the separation
37
between the, that means I mean the inner diameter of the outer cylinder, difference of the
inner diameter outer cylinder and inner diameter of the, in a, I mean diameter of the inner
cylinders that is in the b, right and L is the length of the cylinder.
If we have then we can write, if the annular spaces b is sufficiently thin in comparison
with the, you see actually, I am sorry, let me go back.
38
Actually this will be, no, no, no; so, this R 2 actually if you look at, this R 1 will come to
this position, right? Now R 2, from the center R 2 will, should come to the, this one. That
means I should say by R 2 is equal to R 1 plus b. b is the separation between the two
cylinders, separation between the two cylinders or I should say R 2 minus R 1 is equal to
b, R 2 minus R 1 equal to b. That means the inner radius of the outer cylinder minus the
radius of the inner cylinder is equal to b.
If the annular space b is sufficiently thin in comparison with the radius of the inner
cylinder, then the rotating cylinder arrangement approximates the parallel plate situation
and the velocity profile in the gap space may be assumed to be linear. Then, dv by dy
equal to R 2 omega by b, equation number 5.
39
The distance y is taken in the radial directions, right; distance y is taken in the radial
directions and it is assumed that that b we have assumed here that the b is, b is much,
much less than R 1. It is, did not come, so b is, should be much, much less than R 1. If the
torque T is measured and the fluid shear stress will be expressed as tau equal to T upon 2
pi R 1 square 2 pi upon, sorry T equal to tau equal to T upon 2 pi R 1 square into L,
right? This is equation number 6, where L is the length of the inner cylinder and the
viscosity is determined by the, combining equation 3, 5 and 6, what are the those 3, 5, 6
let us look at.
40
Very beginning we have the equation number of, yes this equation of the absolute
viscosity. So, this, this is the equation, then 5, 5 and 6.
So where the, this I combine, so where L is the length of the cylinder, inner cylinder and
the viscosity is determined by combining the equations 3, 5, 6 equal to mu equal to T into
41
b 2 pi R 1 square R 2 square ,T b upon 2 pi R 1 square R 2 square L into omega. This is
equation number 7, right?
If the bottom gap space a is small, because then the bottom disk will also contribute to the
torque and the influence the calculations of the viscosity. Usually it is small, so the torque
on the bottom disk will be Td equal to pi mu pi omega by 2 a multiplied by R 1 to the
power 4. This is equation number 8.
42
So, the total torque due to the bottom and the annular space actually that will be
measured. The total torque by the some torque cell or torque meter, total torque due to the
bottom and the annular space will be giving by mu pi omega R 1 square multiplied by R
1 square by 2a plus 2LR 2 by b. This is equation number 9, right, right? If the torque,
now angular velocity, these are all constant, so if the torque is measured, angular velocity
is measured, dimension of the cylinder that means R 1, R 2, L, a, b, these are all known to
us, then the viscosity may be calculated from the equation 9, is not it?
So, because omega is known, because what is the rotation that we can measure by some, I
mean our tachometer we know this is the dimensions R 1, R 2, L, a, b. These are the
dimension of the, all the cylinders; that also it is known to us. Now we can measure tau
T, because by some torque cell, so obviously only unknown is mu, right? So, the
viscosity of the liquid, so please note that liquid is to be filled up. That is the most
important thing.
43
That is I have not, you see, sorry, so this gap, this gap is to be filled up of the, by the
liquid. This gap, please look at, this gap is to be filled up with the liquid of which we are
interested to measure the viscosity. So this gap, in, that means in between the two
cylinders, in a cylinder, in a cylinder …. there is a gap. So, this gap is to be filled up by
the liquid by which you will measure, in which we are interested to measure, of which we
are interested to measure the viscosity, right? This is the, so the, if I can measure, so
obviously from these equations I can measure.
Now, there are many other methods of, very conventional methods we will find in the
laboratory, physics laboratory people use. What they use? They use a cylinder, right, a
small, a long cylinder and they filled up the liquid, they filed up that cylinder with the,
the particular liquid in which we want to measure the viscosity, right? I need, we need a
large quantity of the, that liquid, obviously around a liter or something like that. Then,
they drop a ball from the top, a steel ball, right, of known diameter or radius, spherical
ball, so they are fallen. So, they between two particular points of that cylinder they
measure the, when they, when they crosses that particular line, top line they turn on
the stop watch. When it crosses the bottom line in the liquid they stop the stop watch.
44
Then what will happen? So, the total time of fall by, they measure and from that also if
we know the dimensions of our, that ball we can find the viscosity. It is a very common
experiment, also you will find in many physics laboratory people uses to measure the
viscosity of the liquid. But here the problem is that that, the time of fall ….. to measure
very precisely. If there is an error, so you will even find the viscosity of any known liquid
also which we are, you are getting.
In, usually in a, in a, in the physics lab you will find that they are asked to make the
measurement of viscosity of glycerin right? So with some, glycerin is the known,
viscosity is known. Obviously you will find that is very difficult to measure, because if
the time is short you will find there is a large error in measurement, right? When the time
will be short, when the balls will be large, right, when the ball will be large, so it will
have a large gravity, I mean gravitational force. So, it will have a large mass, so it will
quickly, the time of fall between two points within the liquid is very, very short.
So, in that type of situation it is very difficult to, as I close like this, suppose I have a, I
mean cylinder like this one, like this one, so they measure. It is a transparent body, so it
should be made of glass. So there is a, there is a mark here. There is a mark here, so we
put a steel ball. So, whenever it crosses this mark, I turn on this, our meter, stop watch.
When it crosses these marks we stop the meter. We note the distance between these two,
right; distance between these two marks. So, obviously if this time is large, so there is a,
if this is only large when there is a, I mean the ball size is small. If it is, if it is, if it is
large, yes, the time will be large when the ball size is small and time will be very short,
when the ball size will be large, right, so there is the problem in two way.
But, the problem of the time measurement is more crucial we found and most of the time
it is very erroneous to measure the time, right? I mean turning on the stop watch and
stopping the stop watch is very difficult. Rather, if you take a small ball and with a, if, of
which we know precisely the radius or diameter, we can find that it is very easy, most
accurately, almost accurately you can find the viscosity of the liquid, right? But, there is a
very ….. experiments as you can see, it is not, you cannot say it is very industrial sort of
45
devices, whereas rotating drum method or capillary method is, can be utilized in industry
for the measurement of viscosity.
The capillary flow method - in this method the pressure drop is measured across a
capillary tube in which the liquid is flowing in the laminar fashion. Consider the tube
cross, cross section shown in Figure 8. Reynolds number is defined as, you know, rho, I
mean R e rho v m by d is the mean flow velocity, d by mu. Density of the liquid, velocity
of the liquid, mean velocity of the liquid through the capillary, d is the diameter and mu
is the viscosity of the liquid.
46
If the laminar flow exists in the tube, the familiar parabolic-velocity profile will be
experienced as shown in Figure 8 and if the fluid is incompressible, incompressible and
the flow is steady, the volume rate of flow Q will be given by Q equal to pi R to the
power 4 p 1 minus p 2 by 8 mu L. It is meter cube per second, right?
47
You see here the velocity distributions of laminar flow. The liquid is in contact with the,
this pipe diameter, inside pipe diameter there is no motions and here you will find the
velocity is maximum at the central point and as we, again it goes to the, this position,
what will happen that there will be no movement of the velocity of the liquid which is in
contact with the surface of the, surface of the, inner surface of the pipe. It is similarly,
so this is the velocity profile of a laminar flow of a liquid inside a pipe, right? We are
talking of v m that is the average velocity including ….. points measuring and taking the
average.
A viscosity measurement is possible by measuring the, this, I mean the very familiar
equations all of us know.
48
This p 1 is, p 2 is the two points between which we are measuring the Q. It is the pressure
at p 1 at terminal 1 and terminal 2. L is the length between the two tapings of the
pressures, right?
This is, you see the, here we have pressure tap. We have a pressure tap here also, right
and this is at length between the two pressure taps.
49
Viscosity measurement is possible by measuring the volume rate of flow and the pressure
drop for flow in such a tube. To ensure the laminar flow will exist, a small diameter
capillary tube is used and the small diameter reduces the Reynolds number as calculated
from equation 10. In equation 10, the product rho, v m may be calculated from rho, v
equal to m dot by pi r square, where m is the mass flow rate, right? v m is the mean
velocity, mean velocity of the fluid inside the capillary tube multiplied by rho ….. will
give you the mass m dot. You see, the m dot is the mass flow rate divided by pi r square.
50
When viscosity measurement is made on the gas, the, the compressibility of the gas must
be taken into account and resulting expression for the mass flow of the gas under laminar
conditions in the capillaries m dot equal to pi R to the power 4 16 mu R naught T p 1
square minus p 2 square. R naught is the gas constant. We want to differentiate from this
radius that is the reason this R, R. So the, we have used the different notations. R is
actually radius of the pipe of the capillary through which the gas is in the moving.
51
Care must be taken to ensure that the flow in the capillary is fully developed that is that
the parabolic velocity profile has been established. This means that the measure, pressure
measurement should be taken for a far enough downstream from the entrance of the tube
to ensure that the developed flow condition presents.
It may be expressed that the flow will be fully developed when capital L by d equal to
greater than Reynolds number by 8, where L is the distance from the entrance of the tube.
52
In that industrial viscosimeter you see, the Saybolt viscosimeter is a very commonly used
viscosimeter. Saybolt viscosimeter is an industrial device that uses the capillary tube
principle for measurement of viscosities of liquid. A schematic of the device is shown in
Figure 9. A cylinder is filled with the top, with the liquid and the liquid is then allowed to
drain from the bottom through the short capillary tube.
53
Now, let us go back first to the figure. This is our industrial viscosimeter. You see, this is
to be placed in a constant temperature bath. Please note, these are all the dimensions of
the same we have given. So, this is to be placed in a constant temperature bath, because
viscosity, if the temperature changes also the viscosity changes and this dimension, this is
basically if you look at this dimensions is around 0.1765 centimeter just above 1
millimeter, right? This is 0.9 centimeter, this is 0.1765 centimeter, this is 2.46 centimeter.
So this, all the dimension are given, right? Let us go back.
The time required for a given quantity of liquid to drain is recorded and this time is taken
as the indicative of the viscosity of the liquid. Since the capillary tube is short, a fully
developed laminar viscosity profile is not established and it is necessary to apply a
correction to account for the actual profile, right?
54
The velocity profile were fully developed. If the velocity profile fully developed, the
kinetic, kinematic viscosity would vary directly with the time for drainage. That is mu k
is kinematic viscosity which is absolute viscosity upon rho equal to c 1 t, right? This
already we have shown, right? All the dimensions we have shown here, we can see here,
right? See, industrial viscosimeter, you see a cylinder is filled with the top with the liquid
and the liquid is then allowed to drain from the bottom, right, right?
This is all about our viscosity measurements and viscosity measurements as you know
also is very much necessary for industry also, I mean for quality and the taste of the food
also and I mean both in the food and beverage industry also this measurement is
necessary and this is very, paper and pulp industry is very much, viscosity measurement
is very much important. We will find that the measurement of viscosity is very much
necessary, because as I told you there is, it reduces the pumping costs of the liquid and
the quality, I mean control also it by because this is one of the parameters by which you
can make the quality control of your liquid or product. So, for all these reason viscosity
measurements are equally important as pH, right? So with this, see I come to the end of
lesson 20 of Industrial Instrumentation.
55
Welcome to the lesson 21 of Industrial Instrumentation. In this particular lesson, we will

do some exercises. That means I will give you some problems and also we will provide
the solutions. The best way to solve the problem you first do not look at the solution.
First try to solve the problem and then see whether the, whatever the answer you got it,
that is getting, I mean tally it with the results which we have given, right?
So, we will particularly solve in this lesson the problems on the LVDT and then pH probe
and McLeod Gauge. As you know, the McLeod Gauge is used for the low pressure
measurements, pH is used for the, pH probe is used for the pH measurements and so
LVDT, LVDT already we have covered. So we will have, solve different tops of,
different types of problems of LVDT, right? So, let us look at the contents of this lesson.
So, these are problems and solutions on industrial instrumentation.
The contents are the problems and solutions – LVDT, linear variable differential
transformer, pH meter as well as McLeod Gauge, right?
56
Problem number 21.4: the problem is in order to calculate the pH, the following
arrangement is made. A pH electrode is connected through a shielded cable to a non-
inverting amplifier as shown in the figure, we will show the figure.
The input resistance or impedance of the non-inverting amplifier is given by R equal to R

i 1 plus A naught R f by R 1. Everything will be clear in the circuit, so let us go back.
57
Let us look at the circuit once again, R 1, R f. Now R f, R equal to R i 1 plus A naught R
f by R 1. A naught is the open loop gain and R i is the input resistance of the op-amp.
Find the output voltage V naught of the following circuit when 225 millivolt signal is
generated at the electrode. Shielded cable is, looks like this pH electrode and this looks
like this way. So you see, this is buffer amplifiers we have used, supposed to be very high
input impedance, so we have used shielded cable.
So, with this I come to the end of lesson 21, where we have solved all the problems.
58
Prof. A. Barua
Lecture - 21
Problems and solutions on Industrial instrumentation
Welcome to the lesson 21 of Industrial Instrumentation. In this particular lesson, we

will do some exercises. That means I will give you some problems and also we will
provide the solutions. The best way to solve the problem you first do not look at the
solution. First try to solve the problem and then see whether the, whatever the answer
you got it, that is getting, I mean tally it with the results which we have given, right?
So, we will particularly solve in this lesson the problems on the LVDT and then pH
probe and McLeod Gauge. As you know, the McLeod Gauge is used for the low
pressure measurements, pH is used for the, pH probe is used for the pH measurements
and so LVDT, LVDT already we have covered. So we will have, solve different tops
of, different types of problems of LVDT, right?
So, let us look at the contents of this lesson. So, these are problems and solutions on
industrial instrumentation.
1
The contents are the problems and solutions – LVDT, linear variable differential
transformer, pH meter as well as McLeod Gauge, right? So, let us look at the
problem.
The problem 21.1, you see it is telling, in the figure shown below which will come in
the slide 4, let x i be a periodic motion with significant frequency that you can see that
is no problem.
2
You can see here the, this is our LVDT, right, with a phase sensitivity demodulations.
You see the, we have used here four diodes here, four diodes here and we have
discussed this in details, right? So, we make the algebraics, I mean summation of the
two voltage to get the voltage output and this is the input to the system and this is the
excitation. E ex is excitations of our LVDT and x i is the input motions or the input of
the system and e y is the output of the system of the LVDT, right? Let us go back.
In the figure shown below, let x i be periodic motion. So, this time we are measuring
the periodic voltage, I mean periodic signals, right? So, in the figure shown below, let
x i be a periodic motion with a significant frequency content up to 500 Hertz and the
excitation frequency is 5000 Hertz. The output signal obtained is then passed through
a low-pass filter and then to an oscilloscope with an input impedance of 10 to the
power 6 ohm. So, just you can see, just for the sake of problem solving you have to
take an, usually the input impedance are slightly more or even 10, 10 mega ohm or
more. So it is, in this case we have taken 1 mega ohm. We desire the ripple due to
higher frequencies to be no more than 5% of the unfiltered value, right?
3
Find, number a find the frequency range after the modulation process, excuse me,
design a low pass filter for the above application, also calculate the value of R in the
figure. The resistance values which we have shown in the diagram that resistance you
have to calculate and c, calculate the time lag introduced due to the low pass filter,
right? This is our, all the problems. So, these we will solve one by one, right? So, let
us see, let us look at how we can solve this problem, right?
Now, let me take a blank page, right? The diode bridge shown in the figure, as you
know, act as a phase sensitivity demodulator. Now, frequency of the motion is 500
Hertz, right?
4
The frequency of motion, frequency of motion is 500 Hertz. This is actually is f m,

right? The carrier frequency, the carrier frequency f c, this is 5000 Hertz, right and
after the modulation process, other than the motion frequency we will have
frequencies, two frequencies we will get. The two frequencies are two f c plus m plus
f m equal to, this will become 10,500 Hertz and 2 f c minus f m, this is 9,500 Hertz,
right, excuse me. Thus, we can say that the frequency range will lie within these
particular signals which we have given. Now, this is a problem, I mean a, we have
solved. Now let us look at the part b.
In part b, we are telling that the design a low pass filter for the above application.
Also, calculate the value of the resistance R of the phase sensitivity demodulator
circuit as shown in the figure, right? You see, let us go to figure, so which will be
more clear. So, let us take a new, first go back, right? So, you can see here.
5
So this is our figure, so this R we have to calculate and the low pass filter, because we
need a low pass filter to get the actual signals that we say movement of this one that
correspond to that what is the voltage output that we will get? So, we will find that
thing. So, let us take a blank page again, right? Now, in order to design the low pass
filter, we first need to choose, excuse me, we first need to choose the order of the
filter that will give us our desired result. Now, let us first consider a first order filter,
right?
6
The first order filter will look like, this input is coming from the phase sensitive
demodulator output. This output we are designated as e 1 and this resistance is R f and
this resistance is C f, right? Let us choose that. Now, as you know we will have e 1 by
e o j omega will be given by 1 upon j omega R f C f plus 1. Now, it is given in the
problem that the resistance due to higher frequencies to be less than 5% of the
unfiltered value. So, for higher frequencies we consider the 9500 Hertz and above. So,
if it is 9500 Hertz, then obviously I can write the equation which looks like this.
That 0.05 equal to 1 upon root over omega square tau f square plus 1, where tau f
equal to R f into C f that is the time constant of the filter circuits just we have drawn
and omega, obviously is equal to 2 pi into 500, right? Sorry this is, if I take, so omega
is equal to 2 pi into 9500, so or tau f, the time constant of the circuit will be given by
0.0003348 second, right? For this T f we have the amplitude ratio, you know. So, for
this T f, the amplitude ratio we will get that means e 1 by e o you take mod of that. It
will become 1 upon root over 2 pi f m tau f whole square plus 1. So, this will give you
0.69.
Thus we see that the, that the high frequency portions of the x i will be distorted
considerably, right? So, high frequency component of this x i that means the
movement x i is the movement to the, our LVDT core. So, that will be distorted
7
considerably. So this, it, inference is that this type of first order section will not be
suitable for this type of application. So, let us take a second order section, right? So, if
you take a second order section, it will be like this.
The filter will look like, these are all passive filters. So, most of the cases it will
suffice, because if you use the active filter there are some advantages, as well some
there are disadvantages also. This is e 2, right? This is R f as before, this is e o, this is
C f. So, this is 10 R f we have taken and this is we have taken point C f, right? So, e 2
by, I can write e 2 by e o j omega is equal to 1 upon j omega R f C f plus 1 whole
square, right?
Now again let us check whether this, I mean is good for my work. So, .05 that we
have done before also for single order filter, omega tau f square plus 1, right, so tau f
will come up as, tau f will come up as 0.00731 second, right? So, tau f will come up
as, so I can write like there. So, the tau f will come up as 0.000731 second.
8
So, if I look at the amplitude ratio for the, for the second order case, so the amplitude
ratio for the second order case will be, amplitude ratio for the second order case is
given by e 2 by e 0 equal to 1 by 2 pi f m into tau f square plus 1. So, this will give
you point, 0.95, right, clear? So, this will give you this value. I think there will be one
square root here, is not it? So, this will give you the, thus you see that the amplitude
ratio from omega naught will be, so the amplitude ratio 1 by omega naught, I am
sorry, so actually this will be, let me write down again.
9
So, the amplitude ratios will be e 2 by e 0, e 2 by e 0 equal to 1 upon 2 pi f m tau f
square plus 1, which will give you 0.95, right, clear? So, if this is the case, so I can
say that the amplitude ratio from omega 0 to e is nearly flat. That means the amplitude
ratio from omega equal to 0 to omega equal to 3146 radians per second is nearly flat.
That is from 1 it will be 1 and it will fall to 0.95, right. So, it is reasonably good. Now,
in order to determine the, the values of R f and C f, we use the fact that each stage in
the process must have an impedance 10 times more than that of the previous stage.
Now, you see the oscilloscope has an input impedance of 1 meg ohm, is not it?
If the oscilloscope has an input impedance of 1 meg ohm, I can write 10 R f equal to
10 to the power 5 ohm. So, R f we can take equal to 10 to the power 4 ohm, right and
now C f into R f equal to 0.0000731. So, C f will be equal to 73 into 10 to the power
minus 6 by 10 to the power 4. So, it is coming up as 0.0073 micro farad, right? So,
ultimately we get the circuit which will look like the filter circuit. Let me take a new
page.
10
Sorry, this will be again, right, so let us go to pen. So, this will be again resistance.
So, let me take the eraser again, so this is gone. So, we take the pen. So, this will be a
capacitor, right? So, between this we will take the output, right? Here, I will get the
input or I can say or between this and this the output from the phase sensitive
demodulator will come and here the output from the, output which will go to the
oscilloscope, to oscilloscope, right? The values are 10 kilo ohm, 100 kilo ohm. Then,
0.0073 micro farad, 0.00073 micro farad, right?
Now, the third part, the part c we have to find the phase angle. So, let us do that
problem. Let me take a new page.
11
So the problem is, solution to the part c of the problem, problem 21.1, to determine
the phase angle, a phase angle e 2 by e naught j omega minus of tan inverse 2 omega
tau f upon 1 minus omega square by tau f square, so this will give you minus 25.8
degree, right? So, the phase angle will introduce a delay which can be computed as, so
the delay it will produce will be given by 25.8 upon 57.3 into 3140, which is equal to
144 micro second. These are the, all solution to the problem number 21.1.
So, now we will go to problem 21.2 that is also an LVDT. So, let us look at that,
right?
12
Problem 21.2, in the following figure a voltmeter with a finite input impedance R m is
connected. Find the expression for the frequency at which the phase shift will be zero.
In many situations this is very good, I mean, I mean problem is to know if the
manufacturers also will give you some frequency, manufacturer of LVDT will give
you the frequency, where the phase shift will be zero, right? So, it will be utilized, that
frequency can utilized to make the input output phase shift zero. Because you see the,
we have seen when we discussed LVDT to kill this phase shift we have to use lead lag
network.
In the case of leading phase angle we have to use a lag network, in the case of lagging
phase angle we need a lead network. So, that type of additional circuitry we can
dispense off, if I use a particular frequency of signal when there is no input output
phase shift, right? So, it is important in that case. So, how will I know, how will I
know what is the frequency, where the phase shift will be, input output phase shift
will be zero? Once you solve this particular problem 21.2, so you will find that the,
we can find the frequency where the phase shift is zero. Typically the manufactures
also supply this particular frequency, right? Let us go now.
13
This is our circuit, so let us go back.
You see, this is our, in the following figure, a voltmeter with a finite input impedance
R m is connected. Find the expression for the frequency at which the phase shift will
be zero.
14
This is our circuit. You can see here, so we have connected a voltmeter here actually,
see, we have connected a voltmeter here. We have connected a voltmeter here, right,
to get a phase shift, I mean to measure this voltage, right? So, let us first compute,
right, all the values are given. L p is the inductance of the primary side, in the
resistance of the primary side. L s is the inductance of the one secondary, L s is the
inductance of another secondary. So, we should have one core here which we have
not shown, right, which has a moment x i, right? So, this is our problem. So, let us
solve the frequency, let us find the frequency when the phase shift will be zero, right?
Let us take a blank page, right? You see, what will happen here? Anyway, let us first
go back to the, let us take a blank page.
15
So, this is a problem number, solution to the, sorry, solution to the problem 21.2, right
and analyzing the, our circuit we have seen that we can write the circuit equations like
this. i p R p plus L p Di p, D is the delta operator, we have seen that thing many times
before, M 1 minus M 2 D that means D means as you know d by dt, right? So, in this
case you see what is that you see here? This is D means d by dt, right? So, it will be,
in this case it will be di s by dt. In this case it will be di p by dt, right, right? Let us
take the pen again, fine. Di s minus excitation voltage equal to zero, right and M 1
minus M 2 in the secondary side also we can write into Di p that colour, the pen
colour changed, it does not matter, R s plus R m into i s plus 2 L s Di s is equal to
zero. So, this will give you the output which will look like, let us take this pen, e
naught by excitations D equal to R m, meter impedance M 2 minus M 1 D upon M 1
minus M 2 whole square plus 2 L p L s D square plus L p 2 R s plus R m plus 2 L s R
p D plus 2 R s plus R m R p, right? Let us take new page.
16
So, the e naught by e x I can write in s domain equal to R m M 2 minus M 1 s upon M

1 minus M 2 whole square plus 2 L p L s multiplied by s square, second order system
plus L p into 2R s plus R m plus 2L s into R p s plus 2R s plus R m into R p, right?
This equations if I write in the j omega domain, so it will look like ex equal to j
omega R m M 2 minus M 1 upon j omega L p 2 R s plus R m plus 2 L s R p plus 2 R s
R m … L p L s, like this one. See, now if I want to make the phase shift, input output
phase shift zero, then you see that this term that means if I take a different pen, this
term, this term, this term and this term will be zero. Then, j omega, j, j will cancel out.
So, it will be only the imaginary parts. So, there is no phase shift between input and
output. Otherwise there will be phase shift, right?
So, this entire term that means 2 R s R m multiplied by R p minus omega square M 1
minus M 2 whole square plus 2 L p L s should be zero, right? So that, obviously then
I can make that I can say that to make the phase shift zero that to make the input
output phase, phase shift zero, the last term of the denominator is to be zero. So, let us
do that. Take a new page, fine.
17
So that means to make phase shift zero I will make 2, sorry take another new page, so
that to make phase shift zero, 2R s plus R m R p minus omega square M 1 minus M 2
whole square plus 2L p, 2Lp, let me take eraser that means 2L p L s will be equal to
zero. Therefore, omega will be equal to root over 2R s plus R m R p by M 1 minus M
2 whole square plus 2L p L s, right? So, obviously at that frequency when there is no
phase shift between input output, our input output relations will be e naught by ex j
omega will be equal to R m M 2 minus M 1 upon L p 2R s plus R m plus 2L s R p,
right? So, this is our final expressions, clear?
Now, let us go to the problem number 22.3, right?
18
Now 21.3, I am sorry, design a linear variable differential transformer LVDT based
on the following details. What are the details let us look at? Supply frequency is 10
kilo Hertz, length of the core L a should be 20 centimeter, maximum distance of the
core from null position that is that means maximum movement of the core from the
null position is 6 centimeter.
19
Error introduced due to non-linearity is 10%, maximum emf induced in the secondary
coils e 0 that is .5 volt, I am sorry. Current in the primary for maximum emf in the
secondary is fixed also, 20 milliampere. We assume that number of turns in the
secondary to be the, which is the N s, 4 times the number of turns in the primary.
As you know, the number of turns in the secondary is always higher, right? So, you
have to design this LVDT, right? So, let us proceed in this particular design, Let us
take a blank page.
So, the net induced emf e j omega of the secondary coil is given by e j omega. So, this
is the problem, solution to problem 20.3, right? So, the net induced emf, so the net
induced emf, emf e j omega of the secondary coil is given by j omega I p 4 pi N p N s
mu naught p x upon 3 s natural log r o by r i multiplied by 1 minus x square by 2 p
square, right? Now, we have taken the standard notations for this and we can write
obviously that that e naught if I take the mod of this, so this will be, we have used all
the standard notations. As you know that to solve this problem we must refer to the,
our, our lessons on the LVDT, right?
So, because I do not like to repeat all these notations which I have used, so if you,
once the person attended to that particular lesson, he can also solve this problem. So,
same notations we have used in solving the problem also, right? Now, e naught, I
20
have, obviously I can write omega I p equal to 4 pi N p N s mu naught p x 3 s natural
log r o by r i 1 minus x square by 2 p square, like this. We know from the text that the
ratio r i by L a is about right is a, the ratio r i by L a, that means r i by L a typically is
0.05, right? Both are in meters or in centimeters.
Then it is, therefore the inner radius of the LVDT assembly, so the inner radius of
LVDT assembly will be that means r i will be equal to .05 into 20, because 20 is the
length, already given length of the, I think primary, right? Length of the primary is 20
centimeter. So, it is coming as 1 centimeter, right? The ratio r o by r i varies between
2 to 8. So, let us take that means ratio r o by r i it is also given, varies between 2 to 8.
Let us take r o by r i 4, then the outer radius of the LVDT assembly is equal to 4
centimeter, right? Now, length of the primary winding, so the outer radius let me
write r o. So, since this is 1 centimeter, so 4 into 1, so it is 4 centimeter, right? Now,
length of the primary winding will be given by, take a new page.
21
Length of the primary winding, primary winding will be equal to p equal to x max
upon 2 epsilon. So, this will come up as 6 upon 2 into 0.1, right, right 2 into 0.1,
where epsilon is a non-linearity, so which is we are saying 10%, right, 10% non-
linearity. So, .1; so, it will give you the value of and since it is, 6 is the maximum
displacement of your LVDT core, so it is coming up as 13.4 centimeter. So, let us take
it 14 centimeter, right, say right?
Now, length of the secondary windings will be, so the length of the secondary
winding, secondary winding S will be equal to p plus x max is equal to 20 centimeter,
right, 14 plus 6, it is 20 centimeter, right?
22
Now, in equation 1, if I go back, the equation 1 means that the, which is a mod 1,
right, see if I go back whether you can see, yes so this is the equation 1, right? So, so
the equation 1, let me take a new page.
Now from the equation 1, when e naught, mod of e naught equal to 5 volt, I p equal to
20 milliampere, putting the values of p, s, r o by r i and omega that means putting the
value of p, s, r o by r i, omega in the equation 1, equation 1 and noting that the
23
maximum emf is induced at the maximum distance from the null positions, we find
that taking permeability of the free space as mu naught as 4 pi into 10 to the power
minus 7 Henry per meter, so we will get the, the value of, we can write down that
means in the left hand side if e naught is 5 volts, so 5 volt it will come up as six point,
after plugging in all these values p, s, r o by r i, omega in equation 1 and putting mu
naught equal to 4 pi 10 to the power minus 7 Henry per meter, we will get, on the left
hand side we will get 5, because e naught is 5 volt.
That will be equal to 6.28 into 12070 into 10 to the power minus 8 upon 4.2 N p N s,
right? Now, N p N s we will get 27427, right or because N s we have given that N s is
equal to 4 times of N p it is given already, right? So, I can right that N p square I will
get equal to 6856, right?
So, Np will be equal to, so N p almost equal to 83, right? We take N p, now we take N
p equal to 85, say and N s we have taken equal to 4 times, 340, right? So, if I write
down all the design parameters of the LVDT which have we have designed, so it will
look like this.
24
So, complete design of LVDT L a equal to 20 centimeter, right? r i equal to 1

centimeter, then r o equal to 4 centimeter. Then, p primary length 14 centimeter, s is
20 centimeter, number of turns in the primary is 85, number of turns in the secondary
is 340, right? So, this is a complete solution to the problem number 12.3, right?
Now, let us do one problem on the pH meters. We have solved the problem on the pH
meter, we have already done pH sensors in the lesson, I think 20. So, we will solve
some problems on the pH also. Let us go back.
25
Problem number 21.4: the problem is in order to calculate the pH, the following
arrangement is made. A pH electrode is connected through a shielded cable to a non-
inverting amplifier as shown in the figure, we will show the figure.
The input resistance or impedance of the non-inverting amplifier is given by R equal

to R i 1 plus A naught R f by R 1. Everything will be clear in the circuit, so let us go
back.
26
Let us look at the circuit once again, R 1, R f.
Now R f, R equal to R i 1 plus A naught R f by R 1. A naught is the open loop gain

and R i is the input resistance of the op-amp. Find the output voltage V naught of the
following circuit when 225 millivolt signal is generated at the electrode.
27
Shielded cable is, looks like this pH electrode and this looks like this way. So you see,
this is buffer amplifiers we have used, supposed to be very high input impedance, so
we have used shielded cable.
Now, data given, let me see whether we have, yes; find the output voltage V naught of
the following circuit when 22.5 millivolt signal generated at the electrode, right, very
small signal were generated.
28
The data given are: the resistance of the electrode is 10 to the power 8 ohm, right?
The resistance leakage of the shielded cable is 2 into 10 to the power 8 ohm, right? A
naught, open loop gain of the amplifier is 10 to the power 5, R i is, the input
impedance is 10 to the power 6 ohm and R f is 2 kilo ohm and R 1 is 1kilo ohm. So,
this is our… Now, let us look at the solution.
The solution looks like this, solution to problem 21.4, right? The op-amp given in the
problem can be represented like below. You see that is we can, we can draw it like
this. So, plus minus, so it has a very high input resistance. This is R i, right, R i and
this signal is coming here and this signal is coming here, connected here, connected
here this is R f, this is R 1 1 kilo ohm and this is 2 kilo ohm, right? Now, obviously I
can write you see that R if, R if is looking from this terminal, R if. R if equal to R i 1
plus A naught R f by R 1 and so, R if equal to 10 to the power 6 1 plus 10 to the
power 5 by 1. So, this will give you almost 2 into it is 10 to the power 11 ohm, quite
high input impedance, right?
So, now if draw the equivalent circuit for, for measurement systems it will look like
this.
29
Let me draw the first equivalent circuit of the entire system. This is the pH voltage
which we are getting, source impedance, then resistance, this is the impedance of the
meter, then we have an amplifier, amplifier has a gain of 3, because if you look at the
…. value of R f and R 3, you will find the gain as, is of 3. So, this is the output
voltage, so this looks like this, right? So, this R s 10 to the power 8 ohm, source
impedance, this is our pH voltage and this is a length R L actually the, the cable
resistance, right? So, this cable resistance or leakage resistance of the cable, leakage
resistance of the cable is 2 into 10 to the power 8 ohm; cable resistance to the 8 ohm
and R if, input impedance becomes 2 into 10 to the power 11 ohm that we have just
computed, right?
So, that leakage resistance 2 into 10 to the power 8 ohm, I mean input impedance 2
into 10 to the power 11 ohm, please note it is 11 if I take, so it is 11 ohm. So, this is
11 ohm. Then, the source impedance is 10 to the power 8 ohm and gain of the
amplifier is 10, 3 that is we can find because the closed loop gain is 3. So, the R s is
the resistance of the probe or the resistance of the sensor and R L, R L is the leakage
resistance. Now, the closed loop gain, obviously what is the closed loop gain of the
circuit? That is I already put it, it looks like this - 1 plus R f by R 1.
30
So, this will give you 3 and analyzing the above circuit we can write, analyzing the
above circuit we can write that V o will be equal to K V i R L parallel R if upon R s
plus leakage resistance R L parallel R if, right, so which becomes, almost equal to 3
into 225, because this is voltage generated by the pH probe into R L upon R s plus R
L. Please note this is almost equal to …, right, like this. Why? Because you see that R
if is equal to 2 into 10 to the power 11 ohm, right, which is much, much greater than 2
into that means R L which is equal to 2 into 10 to the power 8 ohm, right, quite
obviously. So, this we can write ultimately, this can we can write 3 into 225 into 2
into 10 to the power 8 upon 10 to the power 8 plus 2 into 10 to the power 8, right? So,
this will give you 3 into 225 into 2 upon, let me take eraser, upon 3. So, it is coming
up as, since this is in millivolt, 225, so this also will be in millivolt, 450 millivolt. This
is the answer, clear?
Now, let us one solve, let us solve one problem on the McLeod gauge. In low pressure
measurements we have seen that there is a, McLeod gauge has a, it is typically used
for measurements of the low pressure. Now, McLeod gauge has non-linearity. So, if
you discard that nonlinearities we can sense as a, the linear sensor, but nonlinearity
introduce some amount of error. So, let us look at that. The problem will be on that,
let us look at that, right?
31
This is problem on the McLeod gauge. This is 21.5 I think, yes, 21.5. The McLeod
gauge has a bulb of volume, you must refer to the lectures or to the video lectures of
the low pressure measurements of the McLeod gauge, right? The McLeod gauge has a
bulb volume of 110 centimeter cube, right? The capillary diameter is 1.2 millimeter.
Initially the reading was found to be 3 centimeter. Later it was found that the observed
reading was wrong and find the error in the, find the error in the measured pressure if
the true reading is 2.5 centimeter. So, there some error has been introduced, let us
look at that, right?
32
So, solution to problem 21.5, right? The volume of the bulb is given by, volume of the
bulb will be given by 11 into 10 to the power 4 millimeter cube, right? The volume of
the capillary, of the capillary for the initial reading is given by V c1, let us take that
volume, 2 into 1.2 to the power square upon 4 into 30, so which is giving you 33.93
millimeter cube. So, the pressure p 1 in that case, I will take a new page.
33
The pressure p 1 is equal to 33.93 into 30 upon 11 10 to the power 4 minus 33.93,
which is coming up as 009256 torr, right? So, it is coming up as 1.2332 Pascal, clear?
However the exact capillary volume is, what is the exact capillary volume?
The exact capillary volume is V c2 equal to 2 into 1.2 whole square by 4 into 25. So,
this will give you 28.27 millimeter cube, right? So, the corresponding pressure,
corresponding pressure will be p 2 if I take that is 28.27 into 25 upon 11 point10 to
the power 4 minus 28.27 is equal to 0. 00642 torr, right? So, it is coming up as 0.857
Pascal. So, this is the true pressure.
34
So, therefore the error is, so error in the measurement is 857. Let me take, it is 857
minus 1.2332 upon .857 into 100%, which is giving a error of 43.9%, right? So, this is
the answer to the system, right? So, with this I come to the end of lesson 21, where we
have solved all the problems.
Welcome to the lesson 22 of Industrial Instrumentation. Actually in this lesson and

subsequent lesson, we will find that, we will discuss some of the basic signal
conditioning circuits. As you know, the signal conditioning circuits are very much
necessary in various phases of the sensors, because we need the, whenever the signals
are electrical, we need, we need to process, we have to process that about signals and
we need some signal conditioning circuits. So, in this lesson and the next lesson we
will discuss some of the signal conditioning circuits commonly used in
instrumentations. This is lesson 22. Now, this is the signal conditioning circuits I.
35
Contents of this lesson - positive and negative feedback topology we will discuss, we
will discuss the active filters, we will single amplifier structure.
So, it does not matter if the sensitivity parameters are high, because I have exactly
designed the resistance value and the capacitance value. So, in that case even though
sensitivities are high, I will get the desired value of the, desired value of the filter
parameters, omega p, Q p, omega z, Q z and capital K, right? So, it is very cheap. It is
very small, noises also, because if you increase the number of amplifiers your noise
problem will also, I mean will be predominant. So, these are the typical problem in
the higher amplifier structures. However, we will see that, we will in a subsequent, I
mean lessons that the, we will go for three amplifier structures where we can achieve
this orthogonal tuning. Also, the sensitivity figures will also, will be less.
With this I come to the end of the lesson 22.Thank you!
36
Prof. A. Barua
Indian Institute of Technology, Kharagpur
Lecture - 22
Signal Conditioning Circuits - I
Welcome to the lesson 22 of Industrial Instrumentation. Actually in this lesson and

subsequent lesson, we will find that, we will discuss some of the basic signal
conditioning circuits. As you know, the signal conditioning circuits are very much
necessary in various phases of the sensors, because we need the, whenever the signals
are electrical, we need, we need to process, we have to process that about signals and
we need some signal conditioning circuits. So, in this lesson and the next lesson we
will discuss some of the signal conditioning circuits commonly used in
instrumentations. This is lesson 22. Now, this is the signal conditioning circuits I.
Contents of this lesson - positive and negative feedback topology we will discuss, we
will discuss the active filters, we will single amplifier structure.
1
At the end of this lesson, the viewer will know single amplifier based active filter, the
design equations, advantages and disadvantages of single op-amp based active filters.
So, basically we will discuss the active filters in, because as you know active filters,
amplifiers, ……. circuit, these are most commonly used analog signal conditioning
circuits. So, we must go into, now initially in this particular lesson we will see that the
single amplifier based active filters. In the subsequent lesson we will find that we will
discuss some other more complex filters. Even though those need single amplifier, I
mean the second order structure, but it has some advantages, so people go for the
multiple amplifier structure or multiple amplifier topology, right?
2
Now filter type - the filters are frequency selective networks and they are categorized
as low pass, band pass, high pass, band reject and low pass notch, right and high pass
notch. So, these are the typical, I mean class of filters we will get. Now, also we have
a, I mean we can say the delay equalizers or all pass filters that is also a class of filters
that will be discussed later on. So, we see the all pass filters, because in all pass filters
we will not consider the, the frequency functions, we will, basically there we will
discuss about the or consider the delay of the networks, right? So, that is the reason it
is called all pass, it is not part of with the game, but in the time domain how the, its
delay occurs that we will see.
3
Now, a general biquadratic filter function is expressed by T s equal to capital K s

square plus omega z by Q z s plus omega z square s square plus omega p by Q p s
plus omega p square, where this is a generalized second order structure. So, let us first
look at second order structure, because if you have second a order structure, we have
a, suppose I have a low pass filter, suppose second order structures I can have a
characteristics like this. So, if I take this page, so low pass characteristics will look
like this. So, this slope of this will be, if this is the frequency omega and this is gain,
so you will find that this will be minus 40 db per decade, right?
4
Now, the legends are in this particular expressions, you see the omega z is zero
frequency or Q z is zero selectivity, omega p is the pole frequency, Q p is the pole
selectivity and capital K is the gain constant of the circuit, right? So, it is like a, these
are the filter parameters. Suppose if have a resistance, you know that resistance if you
say just resistance value it does not define, it does not have any meaning. So, we have
to define everything that means the resistance value, its tolerances, its wattage,
everything is to be defined, right? So, in that sense only you will get the actual filter.
So, similarly in the case of filter also you see, you know that you will have some
parameters which is to be given. Only that, that case, in that case your filter will be
characterized, right? That is called filter specifications.
5
You see, the general biquadratic filter function can also be written as, in last equations
we have seen that it is written in terms of pole frequency and pole selectivity, right
and pole frequency, pole selectivity, zero frequency, zero selectivity. Now, in this
case we are writing with some coefficients, where T s equal to K ms square plus cs
plus d upon s square plus as plus b, where m, c, d, a, b, are, these are all, I mean
coefficients, right? So, in this case please note that all this filter function is a complex
conjugate poles, right and complex conjugate zeros, if it is, if the zeros are there.
Here poles are always, you see this denominator function will define the poles. So,
this is always complex conjugate poles, right? Now, K is a gain constant. Now, this
all m, c, d, a, b, all will be realized by some resistors and capacitors, is not it?
6
A second order low pass filter function is given by the following transfer function - d
equal to s square plus as plus b, right? Now, even though in this case that the, our pole
zero pattern will look like this, suppose sigma j omega plane, this is sigma, this is j
omega, we have a complex conjugate poles and no zeros, right? So, this is called the
pole zero pattern of low pass filter, clear?
7
A second order high pass filter function is given by this - T s equal to ms square s
upon s square plus as plus b.
So, you can see that in this case if I make m and c zero, then I can get the, because if
we look at the generalized functions that it see, you see, if I make m and c zero I will
get, coefficients m and c zero, I will get a low pass function. If I make c and d zero,
then I will get a high pass function, is not it?
8
You see, this is high pass function. Now, K is absorbed in m, because we have a gain,
constant K. Here also d is, K is absorbed in d.
So, we have a, this high pass function and the pole zero pattern will look like, if I take
a white page it will look like sigma j omega plane, so it has a complex conjugate poles
and two zeros at the origin, right, double zeros at the origin, fine.
9
So, this is our high pass function, second order high pass filter function. Always you
please note that this m, a, d, everything is to be realized by some resistance and
capacitance, right and op-amp, obviously will be there.
A second order band pass filter function is given by the following transfer function. It
is given by cs upon s square plus as plus b, right? Now in this case, the pole zero
pattern will look like, pole zero pattern will look like sigma j omega plane, right? So,
we have a complex conjugate poles and one zeros at the origin, clear? This is a pole
zero pattern of a second order band pass filter, clear?
10
A second order band reject filter function will be given by the following equation - s
square plus d upon s square plus as plus b, right? In this case that d must be equal to,
that means it, the pole zero pattern will look like in the case if d is equal to b, then we
call it band reject function.
The pole zero pattern will look like, if I take a blank page, so it will look like sigma j
omega plane, right? So, you have a complex conjugate poles. This poles always lie on
11
a semicircle and two zeros, complex conjugate zeros will be there, right? So, this is
our band reject functions, clear?
Now, if it is b equal to d it is called the band reject functions. If b less than d, it is

called the low pass notch function and if b greater than d we call the high pass notch
function, right? So in this case, what will happen you know that you will have a,
suppose I have a second order all pass filter or delay equalizers is given by K equal to
Ts equal to K s square minus as plus b upon s square plus as plus b. In this case, this
pole zeros are the mirror image of each other. This poles and zeros will be mirror
image of each other. You see, this will be K equal to s square minus as plus b upon s
square plus as plus b, right? It is the mirror image of each other.
12
Now, single amplifier structure, let us look at single amplifier structure. How would
the single, because single amplifier structure is the basic structure, we can have a
double amplifier structures or also three amplifier, four amplifier structure, but single
amplifier is a, I mean, I mean the, it is the cheapest one. So, let us exclude that what is
the advantage, disadvantage of this type of circuits? Now, single amplifier structure or
single amplifier topology can be connected in two different, I mean way. One is that
you put the RC network in the negative feedback path, then we will call it negative
feedback topology. If we put the RC networks in the positive feedback path, we will
call it positive feedback topology, right? We will discuss both. Let us discuss one by
one.
Single amplifier filters can be classified into three major categories - positive
feedback topology, negative feedback topology, we have also an enhanced negative
feedback topology, right? We will discuss the, it is a slight, it is a negative feedback
topology with certain amount of positive feedback that also we will discuss, right? So,
it is called the enhanced negative feedback topology, right?
13
Now, positive feedback topology, you see what will happen? In that case of positive
feedback topology we will have, in this case this RC network is placed in the positive
feedback path of the op-amp and however a negative feedback is also given to keep
the filter stable, right?
Because if you look at the circuit, you see that this is a positive feedback topology.
RC networks in the positive feedback path, but if I keep the circuit as it is, circuit will
14
not be stable. We need to make the circuit stable; we have to give some negative
feedback. So, this negative feedbacks are given by these two resistances r 1 and r 2,
clear?
The transfer function of the network can be expressed in terms of the feed forward
and feedback transfer functions of the positive passive RC network. Feed forward
transfer function V, T FF S equal to V 3 S upon V 1 S making V 2 S equal to zero,
right?
15
So, feedback transfer function T FB S equal to V 3 S upon V 2 S, when V 1 S equal to

zero. From the circuit we can write that V 3 S equal to V plus S that means the
positive input of the op-amp equal to T FF S, feed forward transfer function
multiplied by input V IN S plus feedback transfer function T FB S into V 0 S and so,
solving this we will, can write that, also we can write that V minus S equal to r 1 V 0
S upon r 1 plus r 2 upon V naught by S upon k, because small k equal to 1 plus r 1 r 2.
Please note that the small k and capital K are different.
This small k is the amount of negative feedback you will give in the positive feedback
circuit will be determined, whereas capital K is the gain constants of the circuits,
right? These two are different.
16
Now, where k equal to 1 plus r 2 by r 1 and V naught S equal to A. Obviously, output

voltage will be equal to the amplifier gain multiplied by the difference of the signals
which is coming in the inverting, non-inverting and inverting signal input, so which
can be written as k T FF S 1 minus k T FB S plus k by A, right, small k by A.
For an ideal op-amp, obviously I can assume that is infinite. Then, I can write k equal
to T FS 1 upon 1 minus small k T FB S, right? Now, T FF S is equal to, we can write
17
N FF S that means numerators of the N FF S and D FF S and feedback transfer
function T FB S can be written as N FB S upon D FB S, right? Now, this we will find
that these two are same. That means DFF S that means denominator of the feedback
transfer function D FB S and denominator of the feed forward transfer function DFF S
are same. Why? Let us look at.
N FF S and N FB S are zeros of the RC network observed from the different ports.
Denominator D FF S and D FB S are obtained from the nodal determinants of the RC
networks and thus we can write D S equal to D FF S equal to D FB S. Then I can, this
equations can be simplified as T S equal to small k N FF S upon D S minus k N FB S,
right?
18
So, salient features of this positive feedback topology you see the, what is the salient
features? Zeros of the transfer functions are the zeros of the feed forward RC network,
which can be complex, right? Quiet obvious, you see what I am talking about.
Zeros of the transfer function depends, it depends on the zeros of the feed forward
transfer function, right that can be complex, clear?
19
Whereas, the poles of the transfer functions are determined by the poles of the RC
network, the zeros of the feed forward, feedback transfer function and the factor k.
What is this, let us look at.
You see, the poles of this transfer functions depends on the zeros of the feedback
transfer functions and the function k, ratio that means r 2 and r 1 and DS. What is DS?
20
DS is the poles of the feedback or feed forward transfer function, depends on all these
three factors, right?
Now, negative feedback topology, if you look at the negative feedback, this RC
network is in the negative feedback path of the op-amp and hence the name negative
feedback topology, clear? As before, the transfer functions of the network can be
expressed in terms of the feed forward and feedback transfer function of the passive
network, right?
21
Now, this is a negative feedback topology. You see, this is a stable circuit. I do not
need any positive feedback, whereas in the positive feedback, I mean a circuit we
have seen that if it is positive feedback I have to give certain amount of negative
feedback, right?
Now, we can write here also as before V 3 S equal to V minus S which is zero,
obviously, T FF S, no it is not zero, because it is a negative feedback topology.
22
Let us look at, see it is not zero, V plus is zero, right?
So, V 3 S equal to V minus S equal to T FF S into V IN S plus T FB S into, into V

zero S. So, V zero S equal to A V, I mean this is the output voltage, obviously equal
to the gain of the op-amp multiplied by the difference between inputs which is
coming, right? V plus S and since V plus S is zero, in the case since it is connected to
23
ground, so it is coming minus A V S. So, I can write transfer function T S equal to V
zero S minus upon V IN S minus T FF S upon T FB S plus 1 by A.
As, for ideal op-amp A tends to infinity, then transfer function T S equal to T FF S
into T FB S. As before we know D FF equal to D FB equal to D, so obviously as
before we can write that N FF S upon D S N FB T FB S equal to N FB S by DS. So,
DS you just will cancel out. So, TS equal to minus N FF S upon N FB S, right? That
means the poles of the transfer functions will be determined by the poles of the
feedback networks or feedback transfer function. Zeros of the transfer functions
determined by the zeros of the feed forward transfer function, right?
24
So, this is salient features - the zeros of the feedback network determine the poles of
the transfer function. The zeros of the feed forward network determine the zeros of
the transfer function, right? Zeros of the feed forward network determine the poles of
the transfer function. Zeros of the feedback feed forward network determines the
zeros of the transfer function, right? The poles and zeros can be complex. However,
for a stable network the poles should remain on the left hand side of the s-plane that is
from basic control theory you know this is to be satisfied.
25
The poles of the RC network do not contribute to the transfer function, because d is
getting cancelled out. So, the poles of the RC network whether it is feedback or feed
forward transfer function is not coming into the picture, if you assume obviously the
op-amp is ideal. So, I can say that it is, does not contribute. So, let us look at the
enhanced negative feedback topology. We have seen that in the enhanced negative
feedback topology circuit is slightly different. There we have the RC networks in the
negative feedback path, but we are giving some positive feedback also in the circuit.
This is similar to the negative feedback topology with the only difference that there
also exists a positive feedback.
You see here this is the negative feedback topology, RC networks in the negative
feedback topology and positive is coming. So, so we are giving certain amount of
positive feedback, right? We will see that we will get some tremendous advantage by
this circuit.
26
From circuit analysis we get that V zero by small k S equal to V IN S into T FF S plus
V zero S plus T FB S or V zero S upon V IN S minus T FF S upon T FB S minus 1 by
k or V zero S upon V IN S equal to k N FF S minus upon D minus k N FB S, as
before.
So, design equations let us look at. Among positive feedback single amplifier filter
topologies, we will this, discuss this, we are not discussing this in details. We will
27
discuss actually when we will consider the circuits we will find that in the enhanced
negative feedback topology we will get some tremendous advantages which we will
not get in the positive, neither positive topology or simple negative feedback
topology, right?
Now, let us come to the practical circuits which we will realize a based on single
amplifier that means a practical second order filter circuit which is built around a op-
amp, right? Design equations - among the positive feedback topology single amplifier
filter topologies the Sallen and Key filter is the very popular. For the Sallen and Key
filters, we generally have the widely used choices which are as follows: we have three
choices - equal R equal C design, unity gain amplifier design
In equal R equal C design you will find the, all the resistance and capacitance are of
same values. Unity gain amplifier design you will find that the amplifier gain is unity,
so the bandwidth of the circuits will be higher and Saraga design. Saraga design you
will find that in equal R equal C design you will find that the sensitivities are
extremely high. Even though the element spread is very less, all the resistance are of
equal value, all the capacitance are of same value, but the sensitivity is very high.
In the unity gain amplifier design you will find that the, the element spread especially
if it is in low Q that means Q p is less, the element spread we can, the elements spread
is high, obviously. It is extremely high in the case of unity amplifier design, but we
will find the sensitivity, some of the sensitivity is identically zero, right?
That is the tremendous advantage of the, and this circuit unity gain amplifier design is
very popular when I am using the circuits at the low pole selectivity region, because if
the pole selectivity is less, obviously we will find that the element spread will be less,
right? But, if the elements, the pole selectivity is high, the element spread will be
extremely high though the sensitivities in both the cases will be very, very low, right?
In some cases it is zero. Saraga design as the, both is name of the scientist who has
developed this one. He says that if you can make a particular choice of the elements
you will find that the, these are the neither it is a very large element spread nor the, so
it is a good compromise between the sensitivity and the element spread, right? Let us
look at one by one.
28
Equal R equal C design: in this type of, in this type all the resistors and capacitors are
made of equal value. Major drawback is that the passive sensitivities are very high.
Unity gain amplifier design: in this case, the gain of the amplifier is chosen as 1. As
you know, the unity gain buffer concept or unity gain buffer its bandwidth will be
higher, right and it is found that in this case the sensitivities are quite low and some of
the sensitivities are even zero. The main disadvantage of this is that the capacitor
29
spread of this type of design is very high and it is, can be, however if the, if the, if the
Q p is less or low, then we can, we can use. In that case the capacitors spread will not
be that high.
Saraga design: so, we have k is always, small k is always designated. It is assigned

value of 4 by 3. In doing so, a good compromise between the sensitivity and the
element spread can be achieved, right? So, we can have a good compromise between
the two.
30
So, this is the Sallen and Key, famous Sallen and Key low pass filters using single
amplifier structure, where we need 4 resistors. These two resistors values are
different. This is small r 1 and r 2. This will determine the negative feedback which
you are giving the positive feedback circuit. This is your RC network, so it is positive
feedback topology, because RC network is in the positive feedback path.
31
The transfer function, from the circuit so I can write V naught S by V IN S equal to
small k upon R 1 R 2 C 1 C 2 upon s square plus s 1 upon R 1 C 1 plus 1 upon R 2 C
2 plus 1 upon 1 minus k upon R 2 C 2 plus 1 upon R 1 R 2 C 1 C 2, where this
frequency or pole frequency will be given by omega p equal to root over 1 upon R 1 R
2 C 1 C 2.
Q p equal to root over 1 upon R 1 R 2 C 1 C 2 1 upon R 1 C 1 plus 1 upon R 2 C 1

plus 1 minus k by R 2 C 2 and gain constant capital K equal to small k upon R 1 R 2
C 1 C 2, right?
32
Now, equal R equal C design is, basically is actually you will find that in this equal R
equal C design for the Sallen and Key low pass filter C 1 C 2 equal to 1 or R 1 R 2
equal to R equal to 1 upon omega p. So, obviously k, small k equal to 3 upon 1 by Q
p.
Now, unity gain amplifier design is that design equations are C 2 equal to C and C 1
equal to 4 Q p square C. Now you see, the Q p is slightly if it is 1 the capacitor spread
33
is not that much, but the Q p if it is 10, so it is, but not very uncommon. In that case,
capacitance spread will be extremely high. R 1 equal to R 2 equal to R and CR equal
to 1 upon 2 omega p Q p.
For simplicity we take C 2 equal to 1, C 1 equal to 4 Q p square. R 1 R 2 equal to 1

upon 2 omega p by Q p.
34
Saraga design, you see here the design equations are C 2 equal to 1, C 1 equal to root
3 Q p and R 1 equal to 1 upon Q p omega p, R 2 equal to 1 upon root 3 omega p.
So, small k equal to always 4 by 3 upon 1 plus r 2 by r 1, right? So, it is fixed.
Now, this is a high pass filter. You can see we have interchanged the resistance and
capacitances, right? So, if you change the resistance and capacitance, I will get high
pass filter. You see, this is a very handy circuits and if I have a high pass filters, I
mean if we know the design equations, I can make my own filter. It does not cost
much. It costs more than 5, 6 rupees. Only op-amp is costly, this resistance
capacitance can be ordinary, right? But we need certain amount of tuning. You will
find the, whatever the prescribed center frequency, pole frequency we have asked for,
that I may, we may not achieve that particular frequency. We need little tuning,
because these are, sensitivity of this type of structure is quiet high.
35
Now, from the circuits we can write this high pass transfer function equal to V
naught, you see the small k into s square V naught s, V naught by V IN s ks square
upon s square plus s 1 upon R 2 C 2 plus 1 upon R 2 C 1 plus 1 minus k upon R 1 C 1
plus 1 upon R 1 R 2 C 1 C 2.
So, omega p equal to root over 1 upon R 1 R 2 C 1 C 2, Q p equal to root over 1 upon
R 1 R 2 C 1 C 2 upon 1 upon R 2 C 2 plus 1 upon R 2 C 1 plus 1 minus k R 1 C 1. So,
36
it is, small k equal to this capital K, obviously k equal to 1 plus r 2 by r 1 in this case,
right and yes, capital K equal to small k upon equal to 1 plus r 2 by r 1.
Equal R, here also we have equal R equal C design. The design equations are C 1
equal to C 2 equal to 1, R 1 equal to R 2 equal to 1 by 1 upon omega p. k, small k
equal to 3 minus 1 upon Q p.
37
Same as the low pass, unity gain amplifier design also you have done for the low pass,
where k equal to 1 that means unity gain amplifier design, so obviously k will be
equal to 1 and C 1 equal to C 2 equal to C and R 1 equal to R and R 2 equal to 4 Q p
square into R. CR equal to 1 upon 2 omega p Q p.
For simplicity we take C 1 equal to C 2 equal to 1 and R 1 equal to R equal to 1 upon

2 omega p Q p. So, R 2 equal to 2 omega p 2 Q p by omega p.
(Refer Slide time: 29:46)
38
Saraga design for the high, I mean this structures are high pass structures. C 1 equal to
Q p, C 2 equal to root 3. So, it is R 1 equal to 1 upon root 3 omega p Q p and R 2
equal to 1 upon omega p. Small k equal to 1 plus r 2 by r 1 equal to 4 by 3.
So this is, you see, see this is band pass filters. In the band pass case we need one
more resistor R 3. All other resistors are same as before. We have, but the structure is
slightly different, right? So, in a positive feedback, to make circuit stable I have to
give certain amount of negative feedback, right? So, this is our band pass case. Let us
see the transfer function.
39
ks as it happens 1 0 at the origin upon R 1 C 1 whole upon s square plus s 1 upon R 1

C 1 plus 1 upon R 3 C 2 plus 1 upon R 3 C 1 plus 1 minus k upon R 2 C 1 plus R 1
plus R 2 upon R 1 R 2 C 1 C 2, right?
So, as before we can find this pole frequency, pole selectivity and the gain from the
expressions which looks like omega p equal to R 1 under the square root R 1 plus R 2
upon R 1 R 2 C 1 C 2 Q p equal to under the square root R 1 R 2 R 1 plus R 2 upon R
40
1 R 2 C 1 C 2 under the square root are whole upon 1 upon R 1 C 1 plus R 3 C 2 plus
R 3 C 1 plus 1 upon 1 minus k by R 2 C 1 and capital K equal to small k upon R 1 C
1.
Equal R equal C design - C 1 equal to C 2 equal to 1 and R 1 R 2 equal to R 3 equal to

root 2 by Q p, in the case of band pass; small k equal to 4 by root 2 Q p, right?
41
Now, unity gain amplifier design - when small k equal to 1, the design equations are
C 1 equal to C 2 equal to C, R 1 equal to R 3 equal to 9 Q p square minus 1 into R and
R 2 equal to R and CR is a constant 3 Q p upon 9 Q p square minus 1 into omega p,
right?
Now, for simplicity we can take C 1 equal to C 2 equal to 1, R 2 equal to 3 Q p. Now,

you see these are all, actually we are getting in normalized values, because 1 Farads
you will know what you will use. I may use .01 micro Farad. Accordingly, I have to
scale down, I have to scale up the value of R 2, so that the time constant will remain
same. That is very simplified way of using, I will have 1. So, I can take this .01 micro
Farad. That means I am multiplying it by 10 to the power, well minus 6 and at the
same time, I have to multiply this at 10 to the power, 10 to the power plus 8, right?
So, I have to scale down, scale up resistance, scale down capacitance. So, the total
time constants will remain same. R 1 equal to R 3 equal to 3 Q p by omega p.
42
Now, negative feedback circuits you see that the, these type of circuits have multiple
feedback. So far we have discussed only about the positive feedback circuit that
means a single amplifier or second order filter structure. We have discussed the band
pass, high pass, low pass that means all these three different structure, I mean filter
functions, the circuits which will realize this filter functions and based on the single
amplifier structure, but all the, in all the cases RC network in the positive feedback
path, so we are calling it positive feedback topology.
Now, let us look at negative feedback topology. Even though we have discussed the
negative feedback topology, but actual circuits we have, we have not discussed. So,
this we will discuss now. They may have two types of structures. One is without
positive feedback and another is with positive feedback, right? There are, two types of
structures are there, with positive feedback.
43
Now, band pass circuit with negative feedback, without positive feedback it looks like
this. V input R 1 C 1 C 2 R 2, you see, is, this type of circuits also called the multiple
feedback circuits. If you look at, there are multiple feedback. There is multiple
feedback path. One is this path, another is this, through this path it is coming, right?
So, this is called multiple feedback circuits and there is no positive feedback. The
positive feedback is also not necessary to make the circuits stable. But, we have seen
that in the positive feedback, I mean if you give certain amount of positive feedback,
some advantage we will get, right that we will discuss after sometime.
44
Now, from the circuits we can write that there is minus s upon R 1 C 2 upon s square
plus s upon 1 upon R 2 C 1 plus 1 upon R 2 C 2 plus 1 upon R 1 R 2 C 1 C 2.
We have omega p equal to 1 upon root over R 1 R 2 C 1 and C 2. Q p equal to under

the square root 1 upon R 1 R 2 C 1 and C 2 and whole upon 1 upon R 2 C 1 plus 1
upon R 2 C 2, right and gain constant K equal to 1 by R 1 C 2. That is quiet obvious,
because there is no positive feedback. So, it is a negative feedback topology that
45
means RC network in the negative feedback path. So, this is a, a filter. These are
called the filter parameter, as you know, this is called the filter parameters. Like, like
at the beginning as I said, if you buy a resistance you have to tell its value, right? If I
say value that does not make any sense, you have to tell the tolerances, you have to
tell the wattage. Once you define all these things, more or less the resistance is
defined. More precisely you have to tell that the, what type of resistors are, these are,
whether these are the wire wound resistors or this is a carbon film resistors or any
other type of resistors.
Similarly, here also in the filters also, these are called filter parameters. These, that
means these five parameters that means pole selectivity, pole frequency, zero
selectivity, zero frequency and the gain constants are the, they are called the filter
parameters. Once you define these things, these five parameters or filters are defined,
right?
Now, design equations here in the case of a negative feedback topology, simple
negative feedback topology or negative feedback without positive feedback circuits,
looks like this. Let us take C 1 equal to C 2 equal to 1 and R 2 equal to 2 Q p by
omega p and R 1 equal to 1 upon 2, 1 upon 2 omega p Q p.
46
So, this is our positive, without positive. So, this is a circuit, is with positive feedback.
You see that to make the circuit stable I do not need this positive. This we have
discussed when we have discussed in general the enhanced negative feedback
topology. This is called the negative feedback with positive feedback or there is
another name we call it enhanced negative feedback topology. Here, you see that we
are giving some certain amount of positive feedback. Some great advantage you will
get if you give certain amount of positive feedback in a negative feedback topology
that will be discussed now.
47
You see, from the circuits I can write that V naught s by V IN s minus s R 1 C 1
minus 1 minus 1 upon k s square plus s 1 upon R 2 C 1 plus 1 upon R 2 C 2 minus 1
upon k minus 1 1 minus 1 upon R 1 C 2 plus 1 upon R 1 R 2 C 1 C 2. Here, k equal to
1 plus R B R A. These two resistances are different. I want to distinguish it from the
positive feedback topology. That is the reason we have given the name R B and R A,
right?
48
Now, pole frequency and pole selectivity of the and gain constants will be given by
omega p equal to 1 upon R 1 R 2 C 1 C 2 Q p equal to 1 upon R 1 R 2 C 1 C 2 and
upon 1 upon R 2 C 1 plus 1 by R 2 C 2 minus 1 upon k minus 1 1 upon R 2 C 2,
right? You see that we will get some advantage if I get, you see, look here this
functions. This is a function, this is a function. There is a subtractive term at the, at
the denominator. Now, why it is, let us, let us look at, sorry, if I can go back, so it is
actually if I, now why I am telling you see what will happen here? Let us take a blank
page, it will be more clear.
You see, in the enhanced negative feedback topology what we got? We got a
subtractive term, is not it? Q p is a, I can write like this one x 1 x 2 minus x 3, is not
it? This is in the enhanced negative feedback topology, right? Now, you see that in all
the previous cases, we will find in all the cases that means if I whatever the, we have
so far discussed
49
That means the single amplifier we have discussed, the positive feedback topology
that means we have discussed the Sallen and Key circuit, is not it? Low pass, high
pass, band pass, is not it? In all the cases we will find, in all the cases we will find that
Q p or the pole selectivity is given by, is given by some x 1 by x 2. Now, what is this
x 1, x 2? Actually this x 1 x 2, actually you will find there is a ratio of some
capacitance or ratio of capacitance or resistance, is not it? So, in all the cases we will
find if I want to make this high, this I mean if I make, want to make this pole
selectivity high, my, this ratio is to be high.
50
This ratio is only high, you will find, if I use a large element spread to make that
means all the single amplifier structure that means in positive feedback and as well as
in the simple negative feedback topology that means Sallen and Key, all the cases that
means low pass, high pass, band pass, as well as the multiple feedback negative, the
multiple feedback circuits or the negative feedback circuits without any positive
feedback, we will find that the, that the Q p is directly proportional to the element
spread. I have to make element spread very high to make the Q p high. This is very
undesirable. That means I have to use large capacitance ratios that means one, if the
one capacitor is 1 pico Farad, other capacitor might be several thousand pico Farad
which is not acceptable or if one resistance is 1 kilo ohm, if I, the Q p is high, the
other resistance might be, some mega ohm, so that is not desirable, right?
51
Whereas, you will find that in the, in the enhanced negative feedback topology that
means the negative feedback topology where you give certain amount of positive
feedback, my expression of sensitivity or selectivity, I am sorry, is coming like x 1
upon x 2 minus x 3. Look at this subtractive term. This is very important, right? I
should take some other, look at this subtractive term. If I make this difference very,
very small, my Q p will be very, very high. If I make x 2 that means x 2 almost equal
to x 3, however x 2 not equal to x 3 and x 2 greater than x 3, then what will happen? I
will get large pole selectivity.
So, in the case of band pass filters, this large pole selectivity is great advantage. So,
that is the reason we talk about this band pass filters with the, because the circuits
which we have discussed, is basically band pass filters with a negative feedback band
pass, negative feedback circuit band pass filter or RC networks in the negative
feedback path and we have given some positive feedback topology. Let us go back to
the, actually so this is our 59 I think, yeah this is 59, I am sorry, yes, this you see, yes
you see, this is a basic band pass circuits we have discussed.
52
This is band pass circuits we have discussed, is not it? So, this is an enhanced
negative feedback band, band pass circuit. That means negative feedback circuits with
positive feedback. So, in band pass circuits, always it is desirable that the selectivity
should be higher and higher. That is a good. Why? Quiet obvious, if I see there, here
so that means what will happen? If I say, why I am telling?
53
That means if the frequency plot if I make omega and gain here, larger the Q more
and more will be the selective networks, is not it? So, the Q increases, right, so Q
increases here, is not it? Now, what is Q? You see, the Q in the, in the terms of
frequency, so what will be the Q? Q will be like this one. So, we will take this
frequency. Suppose this is say, omega naught minus 3 dB point I mean from, coming
from this minus 3 dB. So, I will take this one, this one. So, it will be, suppose this is
omega 1, this is omega 2, so the Q will be omega naught upon omega 2 minus omega
1.
When I take colour different that means Q equal to omega naught upon omega 2
minus omega 1 or omega naught by delta omega or f naught by delta f. So, the larger
and larger the Q, higher and higher, you find the delta f will be smaller and smaller.
That is our requirement. How, but how will you achieve that in actual circuits? I can
achieve only if I have a structure like that, right?
If I have a structure like that, you see here, you see this term minus 1 upon k minus 1
by R 2 R 1 C 2, let us look at the expressions of Q. You see, this is the subtractive
term. So, obviously this I can write, obviously this I can write like this one. So this is,
so let me, let me take, I can write, so I can write this one as x 1, whole this as x 2 and
whole this as x 3. So, x 1 I can write upon x 2 minus x 3, is not it? So, I can do it by
54
controlling the k. You see, I can make x 2 very nearly equal to x 3 by controlling k
only, because these elements already determined; R 1, R 2, C 1, C 2 already
determined by this omega p. So, by controlling k, I can make this. So, I can make the
selectivity very, very high. So, that is the great advantage of the single amplifier
enhanced negative feedback topology, right?
So, capital K which is gain constant upon 1 minus R 1 C 2 minus 1 minus 1 by k.

Now, disadvantage - what is the disadvantage of this single amplifier structure? One
of the problem of the single amplifier structure you will find the sensitivities are
rather high. If you compare with the state variable structures and all these,
sensitivities will be high. Also, the tuning is very difficult in this type of filters. What
is tuning? Let us go back, it will be more clear.
55
You see that once you design the circuits, we will find that I, I am designing circuits
with or synthesizing the circuit with some, I have some intention to get some
particular value of omega p, right with R 1, R 2, C 1, C 2, everything will be there,
right? But the problem will arise. You will find that I may need little bit of tuning the
circuits after fabrications. Whether you are doing in discrete circuits or you are doing
in some integrated circuits, some little bit of tuning is necessary.
Suppose if it is, omega p is prescribed is 1 kilowatt, I got 900 Hertz. So, I need some
tuning. That means I have to choose some value of the resistance which I have to vary
to get the desired value of omega p. The problem in this type of single amplifier
structure is we will find that you see this if I tune, if I, to get suppose after fabrications
I got 900 Hertz to, now to get this 1000 Hertz, the problem will be or 1 kilo Hertz, the
problem will be, if I tune the Q p also will change, because this R 1 is common in
both. So, there is no element you will find which is, which is there in omega p but not
in Q p or which is in Q p, but not in omega p.
So, the orthogonal tuning, similarly for k also; if we assume that the I have to tune or I
have to tune all the five parameters, omega p, Q p, omega z, Q z and K, it is, are just
impossible to make the orthogonal tuning by the single amplifier structure. That is the
great drawback.
56
Now, I talked about the sensitivity, but sensitivity is, look like, you see here now
logarithmic sensitivity we should define like this. Logarithmic sensitivity is, is given
by you see in many cases we found, we have found that omega p, sorry, omega p
equal to 1 upon root over R 1 R 2 C 1 C 2, is not it?
Now, logarithmic sensitivity is something like this that means sensitivity of omega p
with respect to R 1, R 2, C 1, C 2 or I can write in generalized form that sensitivity
57
suppose, I am sorry that means S y, y is a parameter, filter parameter with respect to
the passive element x I want to find the sensitivity, will be delta y by y by delta x by x
or I can write del y by y by del x by x.
Best thing to find the sensitivities that suppose if I have the expressions like this one,
omega p equal to 1 upon root over R 1 R 2 C 1 C 2, you take the logarithm of both
side, natural log omega p equal to minus of half natural log R 1 R 2 C 1 C 2. So, this
will be minus half, obviously natural log R 1 plus natural log R 2 plus natural log C 1
plus natural log C 2. Since it is partial derivative what will happen that you see that
you will, once you vary R 1 all other things will …..
Now, if you take the derivative of this thing, what will happen? I will get, on the left
hand side you see if I do it here I will get on the left hand side, if I take the derivative,
delta omega p by omega p equal to minus half suppose if I take delta R 1, I am
varying the delta R 1, so R 1, sorry so it will be by R 1, right? So the, what is the
sensitivity now?
58
If I write that means S omega p R 1 will be delta omega p by omega p delta R 1 by R

1. In this case it is minus half, right and interestingly you see that in this case that
since omega p is equal to root over 1 upon R 1 R 2 C 1 C 2, so S omega p R 1 is equal
to minus half, I can write S omega p R 2 S omega p C 1 S omega p C 2.
Or in combined way we can write like this that S omega p R 1 R 2 C 1 C 2 equal to

minus half, right? So, this is a sensitive filter. So, similarly I can find the sensitivities
59
of Q p, right? So, typically what we will find I mean if you have all the element,
suppose in a transfer function, most of the transfer function which we have written
that its, its, the parameters, filter parameters we are more interested are omega p,
sorry, let me take a new page.
Omega p, Q p, capital K and if possible omega z and Q z, clear? So, this sensitivities
we have to find with respect to the passive elements. If that particular passive element
are not in the expressions of omega p, so with respect to that, so obviously the
sensitivities will be zero. What is that? You see, look at, if the omega p in my
expression is 1 upon root over R 1 R 2 C 1 C 2, right? So, if may, I may ask you what
is the sensitivity with omega p with respect to small r 1? That is also there in the case
of positive feedback topology. Since it is not there, so it will be, because if I vary
small r 1, how much our pole selectivity, pole frequency will change? Nothing, is not
it, right?
So, it is obviously zero, because it is not equal to obviously again S omega p small r
2, clear? This is our sensitivities figure, right? So this way, obviously I can find also
the sensitivity with respect to Q p. That means Q p with respect to all the elements R
1, R 2, let me take a new page.
60
So, it is Q p with respect to R 1, R 2, C 1, C 2, small r 1, small r 2, in the case of

positive feedback topology. Also S K, gain constant R 1, R 2, C 1, C 2, small r 1,
small r 2, this may not be same S. Q p R 1 may not be same with the S Q p R 2. In the
case of pole frequency it is same, because all …… denominators under the square
roots are becoming same, but this may not be the same, right?
Once I talked about, that you see in the case of unity gain amplifier design, some of
the sensitivities are zero. You will find exactly that that some of the sensitivities
especially the Q p sensitivities we will find S Q p sensitivities for the unity gain
amplifier design is zero for some of the elements, not for all the elements, right? So,
this way you will find the sensitivity.
61
So, the problem with the single amplifier structure thus we have discussed is basically
we will find that disadvantage of the single amplifier, the sensitivities are extremely
high. Number 2, component spread is also high, right? Most of the cases we will find,
except if you take, because you, if you do not say the sensitivities are, I say that I will
take some other structures where the sensitivities are not high, suppose if I take
Saraga design or if I take unity gain amplifier design, right, in that case we will find
that the sensitivities are zero, but those element spreads are very, very high, which is
also not acceptable, if I want to make your filters in a very, in integrated circuits,
right? So, component spread is also high. So, that is not also acceptable.
That is a problem with the single amplifier structure and quiet obviously you will say
that I will make you Q p less. But, Q p is not in your hand. Some, somebody ask you
suppose in some signal conditioning circuits I need a very large Q p, so in that case
your component spread also will be very high, if you use even unity gain amplifier
structure, right and orthogonal tuning that I discussed, this is also not possible.
Orthogonal tuning is just impossible in the case of single amplifier structure. So, these
are the typical drawback of the, these type, I mean single amplifier structure, but
obviously you see that if I can, suppose sensitivities we are talking about, suppose
tomorrow some technology is available, where I can make the, our desired value of
the resistance and capacitance precisely, so it does not matter if the sensitivity
62
parameters are high, because I have exactly designed the resistance value and the
capacitance value.
So, in that case even though sensitivities are high, I will get the desired value of the,
desired value of the filter parameters, omega p, Q p, omega z, Q z and capital K,
right? So, it is very cheap. It is very small, noises also, because if you increase the
number of amplifiers your noise problem will also, I mean will be predominant. So,
these are the typical problem in the higher amplifier structures. However, we will see
that, we will in a subsequent, I mean lessons that the, we will go for three amplifier
structures where we can achieve this orthogonal tuning. Also, the sensitivity figures
will also, will be less.
With this I come to the end of the lesson 22.Thank you!
63
Prof. Alok. Barua
Lecture - 23
Signal Conditioning Circuits - II
Welcome to the lesson 23 of Industrial Instrumentation. In this lesson, we will continue

with the signal conditioning circuits. So, this is the signal conditioning circuit II. In
lesson 22, we have discussed some signal conditioning circuits. We have discussed the
filters and basically we concentrate on the signal amplifier filter. But, as you know that
filter is one of the, one of the most vital part for analog signal processing, so we have
devoted at least two, one and half lessons for filters. Now, these analog filters are off the
shelf devices, I mean you can make the filters with a very low cost and it will give you
good performance also if you can design it properly.
Now, the signal amplifier filter suffers from the drawback that the tuning of the filters is
very difficult. As you know the filters has some typical parameters. This is omega p, Q p
and K which we have discussed in the lesson 22 in details and these basically depends on
some resistance and capacitance, right? Now the, in the case of filters, whatever the, you
will design, whatever the filters you will design, obviously you need you know that you
need perfections of the components, I mean if you say that the you have designed
components, I mean actual design equation gives you a component value of 923 ohm, so
instead of 923 ohm it is very difficult to get that value you may get it, little deviate, even
if you get that exact value there will be some tolerances of that resistance which will give
you, which will lead you to, ultimately once you design the circuits some sort of tuning of
the I mean these filter parameters omega p, Q p, k, omega z, Q z.
Now, in the problem and the problem of the signal amplifier structure is that it is a low
cost, its signal and dynamic range is also good. Its signal to noise ratio is also good.
Unfortunately if you do not, if your technology does not permit very tight component
tolerances, so the tuning of the filter parameter is very difficult in this case of filters and
1
in this lesson, lesson 23, 23, we actually discuss these two amplifiers and the general
immittance converters, two amplifiers based general immittance converters, then two
amplifier based filters and state variable filters and then sample and hold circuits and then
logarithmic and antilogarithmic amplifier.
This is all, I mean this all is very common sort of, I mean, I mean circuits which is
frequently used in signal processing of signal conditionings.
Let us now lesson 23 you see, the contents of this lessons are two amplifier filters and
general immittance converter and state variable filter, then we have sample and hold
circuits.
2
We will also, now let us look at the general immittance converter. You see, the general
immittance converter, the circuit looks like this. You can see here that that in case
suppose I am interested to measure the output, the input impedance of the circuits, so
from this point what is the input impedance circuit? So, I assume that this voltage is V
and obviously we can see here that you can see here that this resistance, I mean this is
connected to ground and this general immittance converter is basically a device or circuit
which will give you the, you can simulate inductance from the capacitance.
As you know the, it is very difficult to implement the inductance in integrated circuits.
So, people tried over the years to simulate this inductance with the help of capacitance.
Now, the general immittance converter, you can simulate inductance with the help of a
capacitance; you can simulate capacitance with the help of inductance. But, nobody
actually simulate capacitance with the help of inductance, because that is not necessary.
In interior circuits we can always fabricate capacitance of the wide value, so widely
different values in the chip itself, which is not possible in the case of, in a in the, if I want
to do it, the inductor in the, in the chip. Even though people tried for last 2, 3 years to
make a on chip inductance, but it is still in the research state, right?
3
Now, you see that I want to measure the input impedance or I want to look at the input
impedance of the circuit, looking from this terminal, right? So that in this connector I
have connected a voltage source V and I measure the current which is going inside, right?
So, you see, two op-amps are connected. So that is the reason we call two amplifier
structures and I mean, let us make the analysis of the circuits. We will apply the
Kirchhoff’s voltage. Now, it is, one thing is very common, you see. This op-amp has very
high input impedance, right, so that this potential at the node 1 and potential at the node 3
is same.
Again, since the, this op-amp, this potential 3 is, node 3 is common, node 3 is common
here, you see here, so this is connected to the, again the second amplifier. So, this
potential, node 3 potential and node 5 potential is same, clear?
So if I now, I apply the Kirchhoff voltage law in the node 5 we will get, if you apply the
Kirchhoff for the circuit analysis, applying the KCL, Kirchhoff’s current law I will get
V 2 minus V s C equal to V by R 4. Now, let us look at what is, so this is our V 2. You
see here this is our V 2. So, this is our V 2, this is our V 1. Now, I want to apply, this is
the, in this loop these are Kirchhoff’s current law. Let us look at what will we get. This,
4
so applying Kirchhoff’s current law also at the node 3, we will get V 1 minus V by R 2 V
minus V 2 R 3.
So, or V minus V 1 by R 2 equal to V 2 minus V by R 3. So, if this is the equation, so

obviously substituting V 2 minus by V from equation 1, we get V minus V 1 by R 2 equal
to V by s R 3 R 4 C, right?
5
So, applying KCL at node 1, I can write I IN, that means current which is flowing inside
just at the input node, V minus V 1 by R 1 equal to V R 2 by s R 4 R 3 R 1 C. So,
obviously so V by I IN or the input impedance of the circuit will look like s R 4 R 3 R 1
C by R 2, right? So, Z impedance, so I can write Z IN is equal to, the input impedance of
the circuit at the node 1 can be written as s R 4 R 3 R 1 C by R 2. Now you see, these
entire things I can write as an equivalent inductance, you see this entire thing, is not it?
So, I can write it sL eq. So, we can see, interestingly using one single capacitor I can
simulate inductance in the circuits. So, I have used one capacitor and some resistors. So, I
am uniting actually the inductance in the circuits, right? So, it has a tremendous impact in
the signal processing, in instrumentation at least.
So, thus I can write the L equivalent or inductance equivalent equal to R 4 R 3 R 1 C by

R 2. You see there is a various combinations of these; you can have any one of this one.
Suppose you can choose R 4 also as in, in, I mean capacitance by which you can, you can
get the equivalent inductance, right? Now, thus the above two amplifier structure
simulates an inductance ….. Now you see, one thing is very important. If you look at this
circuit, you see this particular circuit will generate actually the grounded inductance.
6
Now, if I want to generate, if I want to simulate the floating inductance, then these two
circuits should be connected back to back, right? Only that time you will get the
inductance, the floating inductance. Now, the inductance which we have generated here
is the grounded inductance. Now, with the help of these inductances I can obviously
simulate low pass, high pass, band pass filters can be realized using these two amplifier
structures. I can realize low pass structures, I can realize high pass structure, I can realize
band pass filters using the two amplifier structures.
Now you see, if you compare this single amplifier structures and two amplifier structure,
you won’t get much advantage in these two cases, except the, except in the form that that
you will find that in the case if you use the band pass filter it will give you very high
input impedance and resonance, which we will not get in the single amplifier structure.
Otherwise all the problems what you have in the case of single amplifier structure that
means the tuning difficulty which is most obvious in the case, in this case also will
remain there. So, I am not getting an advantage compared to the single amplifier
structure, so far the tuning of the filter is concerned, right?
7
Now low pass filter, let us look at this low pass structures that means between input, this,
between this input and this input, output I will get a low pass structure. That means
V O by V IN will give you a value like, will give a value like K divided by s square
omega p by Q p s plus omega p square and this type of structures it will get, right?
Now, if I want to, try to analyze this circuits, from circuit analysis I can find out that the
V naught by V IN s 1 by 1 upon R 1 R 2 multiplied by 1 upon R 4 plus 1 upon R 5
multiplied by R 5 C 1 C 2 and upon s square plus s by R 3 C 2 plus R 5 R 1 R 2 R 4 C 1
C 2, clear? Now see that, I mean I can, as, as it happened in the case of general
immittance, I mean making the circuit analysis for the general immittance converter that
same thing we can do it here also to find the overall, I mean transfer functions …..
domain of this circuit, right?
Now, also I can write the nodal ….. matrix of this network and find the transfer functions.
I am not going to details of that. Now, let us go back again. So, this is my circuit. So it is
a very equal I mean I can, obviously this I can write, what I can write? This I can write K
equal to as I have just written s square plus omega p by Q p s plus omega p square, is not
it?
8
So, this will give you omega p square. So, square root of this will give you the, square
root of this will give you omega p. See, if I make the square root of this, I will get the
omega p and this will give you omega p by Q p, right and this will give you capital K.
So, obviously what is, you can see also here it is very difficult, you see R 3 is present,
obviously you can choose R 3 to tune the Q p. So, it is not that difficult. You can choose
R 3 to tune Q p, but in, in tuning the K that means gain constant is also difficult, because
you see whatever there R 1, R 2, R 4, R 5, everything is present there. It is very difficult
to tune the omega p and capital K independently, though I can tune Q p and omega p
independently in this particular type of circuits, right?
So, design equations are R 1, R 2, R 4, R 5 equal to R, I have taken. R 3 equal to …. we

have taken Q p and C 1 C 2 all equal C. You can, all you can take 1 Farad, accordingly
the resistance will come and C multiplied by R equal to 1 by omega p. So, omega p equal
to 1 by CR.
9
Now you see, the high pass filter slightly structure, the basic two amplifier structure
remains the same. May be I am making little modification. You see this is basic two
amplifier structure remains same. Again I am giving the input here, I am taking output
from this position, right? You see, the two amplifier structures I am giving the input here,
I am taking the output here. See here, as, as it happens in the case of general immittance
converter, in this cases also I can, I can take the, transfer, find the transformation between
an input and output and I can find, I can find the transfer functions as well as I can find
the design equations for this type of high pass filters, right?
10
You see, from circuit analysis obviously I can write s square into 1 plus R 4 by R 5 s
square plus s upon R 1 C 1 plus R 4 upon R 2 R 3 R 5 C 1 C 2.
Now design equations are R 1 equal to Q p R; R 2, R 3, R 4, R 5 equal to R. C 1, C 2

equal to C and CR equal to 1 by omega p, right?
11
Now band pass cases you see here that R 1, similar structures only there is some
difference at the input side. Again, I am giving the input here at this positions, so I am
taking the output from this position. You see, I am giving the input in this position, I am
taking the output from this position; input at this position, output at this position, fine.
So, again I am finding the transfer function V naught upon V IN in s domain.
12
So, I will get a transfer function which looks like this. So, obviously it is a, you can see
here there is a single zero at the origin and two complex conjugate poles. So, V naught by
V IN s equal to s 1 plus R 4 by R 5 upon C 1 R 1 s square upon s square plus s upon R 1
C 1 plus R 4 C 1 C 2 R 2 R 3 R 5, right?
The design equations looks like this R 1 equal to Q p R; R 2, R 3, R 4, R 5 equal to R,

C 1 equal to C 2 equal to C and CR equal to 1 by omega p.
13
Now, state variable feed forward structure looks like this. Now you see I, we have
discussed about the signal amplifier, signal amplifier structure using two amplifiers. I
have just discussed only low pass, band pass, high pass, but you can use, you can
simulate, you can generate other, I mean you can make the other topology like the high
pass notch, low pass notch or band reject or low pass also with the signal amplifier
structures, right? I am not going to details of that, so I am just giving the two amplifier
structure, where we can make low pass, high pass, band pass. Please remember, but
please remember, you can generate other filter functions like high pass notch, band pass
notch, band reject, as well as all pass structure by this two amplifier topology.
Now, as it happen, as I told you that in the case of signal amplifier structure, it is just
impossible to tune orthogonally all the filter parameters. In the case of two amplifier
structures we have seen that we can independently tune the omega p and Q p, quite
obviously but I cannot tune Q p, omega p and K independently. So to, people need some
other structure, where we can make total orthogonal tuning of the filters, right? So, that
type of structure is basically state variable feed forward structure, let us look at.
14
The three amplifier structure looks, you see that we need three amplifiers. Why it is
calling feed forward? You see, there is a feedback. There is no doubt this is a state
variable structure. I have a lossy integrator. You see what is this actually? This is a lossy
integrator, right? This is an inverter, simple inverter and this is another integrator, is not
it? So the, first one is a lossy integrator or damped integrator. Then you have an inverter,
then we have again another integrator and you see here there are feed forward through R
6 and R 7, because input is coming here, it is going to R 4 to the first amplifier, then
through R 6, you see through R 6 and through R 5 we are giving to feed forward.
This type of structure can generate any filter functions. We will see that is advantage of
this type of structure, tremendous advantage of this type of I mean structure. It is, it is
available in the cheap form also, this type of filters, National semiconductor makes this
filter. It is called universal filter, where you can have, so one or two elements outside
which you can tune to get your desired value of omega p and Q p, right?
Now, design equation looks like this. From a circuit analysis we get, we have to, we can
make the circuit analysis, we can say that R, R 8, minus R 8 by R 6 s square plus s 1 by R
1 C 1 1 by R 4 C 1 R 6 R 7 R 6 by R 7 1 upon R 3 R 4 C 1 C 2 s square plus s 1 upon R 1
15
C 1 R 8 by R 7 1 upon R 2 R 3 C 1 C 2, right? One thing you see here, it is very
important, you see here that this is very similar to the structure, I can write T s equal to
minus K s square Cs plus d s square plus as plus b, is not it; very similar to this type of
structure.
Now, for this reason you can see this, all this value, R 8, these components, you can see
this component, this one, so we can realize by some resistance and capacitances. So, if I
purposefully make that portion zero, obviously I can integrate it. Suppose in this case,
suppose if I make c zero and d zero so it will be high pass structure, right? Similarly if I
make, suppose R 6 infinite, in that case it will be a, your, I can make this as a low pass or,
or band pass structure; like that it goes, right? So, with this type of structure, obviously I
can generate any filter function.
Only problem, if you look at the numerator you see that here we have a subtractive term
at the, as the coefficient of s. So, the problem with this subtractive term is that it will, we
have to make a signal, we are getting the, we are making it zero, this coefficient zero by
signal subtractions, right? So, in that type of situation what will happen you know, if
there is a, if there is, suppose in that case, if suppose during cancelation, if this coefficient
is not exactly zero, this coefficient is not exactly zero, then what will, this coefficient is
not exactly zero, then we will find that even though I tried to realize the low pass
structure, so the, in the case of low pass structure as you know the pole zero pattern will
be like this one.
So, two complex conjugate poles and no zeros, so it will find that it will not exactly the
same pole zero pattern, right? So, this is the only difficulties in that this type of state
variable structures. Otherwise it is a tremendous advantage. Its, its sensitivity is also quite
low. Its tuning is a great advantage. You can see in the next slide I can show you what are
the different elements we can choose to tune omega p, Q p, omega z, Q z independently,
right?
16
Because as you know, in that, I mean in the filter parameters we, always we know that we
have to tune omega p. I can take another pen. Omega p, Q p, capital K, omega z and Q z,
these are the things, I mean which we are going to simulate, right? So, I mean this, the,
the parameters which you have to tune it with some resistance elements, capacitance, we
will not use for tuning. Now, how I can make, now orthogonal tuning means that if I use
some resistance to tune or some capacitance I will consider the resistance, if I consider
some resistance, suppose this omega p I am, want to change or modify or tune, so other,
by changing some resistance if I tune omega p, other filter parameters should remain
intact. That is called the orthogonal tuning.
17
Now, if you go back, you see here, you see here what will happen? In this case I can
choose some of the resistance. Suppose if I choose some of the resistance to tune omega
p by that time what will happen? You see that if I choose some resistance to tune omega
p is, there is omega p, what will happen that omega z, Q z may change. It does not matter,
let it go wherever it goes. Then we will use some other resistance or capacitance to bring
it back to the original position, right? So, let us look at that.
18
Now above equations, as you can, can also be written as Ts equal to K s square plus cs
plus d s square plus as plus b and the design equations are C 1 C 2 equal to 1, R 2 R 3 R 7
R 8 equal to R equal to 1 by root b and R 1 equal to 1 by a, R 4 equal to 1 upon K a
minus c, R 5 equal to root over b by Kd and R 6 equal to 1 upon K root b.
The design equations mentioned above yield positive element values for a greater than c.
This, most of the cases this will be satisfied, a will be greater than c. This equation,
design equations mentioned above yield positive element values for a greater than c and
the above equations can also be written as T K, Ts equal to K s square plus omega z by Q
z s plus omega z square upon s square omega p by Q p s plus omega p square.
19
Thus, for orthogonal tuning we can use the following elements as shown in the table
below and they need to be done sequentially as given. You cannot haphazard fashions.
These are the filter parameters and which particular element, passive elements we will
choose to tune it, let us look at. For capital K, gain constant I am tuning R 8, right? For
omega z, I am using R 3. For omega p, I am choosing R 2. Omega p Q p equal to R 1, I
am choosing R 1 to tune and Q z equal to R 4, this completes the tuning of the, all the
filter elements. You have to go in particular this sequence that otherwise there will be a
problem.
You see, because if you choose, tune R 8, let us look at I am taking examples of tuning R
8 or R 3; omega z K for R 8 and omega z equal to R 3, let us look at.
20
Now you see here, so K we are using R 8 and omega z R 3. Let us look at, you see that K,
R 8 is changing. I am getting the proper value of R 8 by tuning the, this proper, proper
value of capital K. This is the, because this K is equal to R 8 by R 6, is not it? So, I am
tuning R 8 to getting proper value of K, but you look at that R 8 is also present in the
expressions of omega p. So, omega p will change, right? It does not matter, let it change.
So, I will bring it back by using some other elements to its desired value.
Similarly, you see that when I am, after K, once the K is tuned, now I am choosing R 3 to
tune omega z. What is omega z here? You see it is omega z. Now, this will also vary. By
that, during that time it does not matter, I will choose some other elements later on, so to
tune the, get the omega p exact desired value. So, this sequence, that is the reason I am
telling that particular sequence you have to maintain, otherwise there will be a problem.
Now you see, so once I have chosen, I have chosen R 3 to tune omega z, I am not using R
3 anywhere to choose any other further elements. I will choose some other elements to
tune the omega p or Q p like that, right? So, that is the case that I want to explain to you.
You see it is here, sorry, here you see omega z R 3 omega p R 2 Q p R 1, so if I vary
omega z, then what will happen? You see if I vary R 3 omega, omega p will vary. It does
21
not matter, I will use R 2 to bring it back to the original value, right or desired values.
Similarly, I can the tune Q p and Q z by R 1 and R 4 respectively, right? This ends the
particular, the only, the filters. That means we have discussed in the lesson 22, the signal
amplifier structures. Then in the lesson 23 we have discussed the two amplifier
structures, we have discussed the three amplifier state variable structures.
State variable structure has another advantage. It is, I mean it is the frequencies, it is the
sensitivities, logarithmic sensitivity is very, very low. Though the one single problem
with the signal amplifier structures is the power dissipation and also the dynamic range of
the filter, because as you know, if you are increasing the number of op-amps you are
making the, your signal to noise ration poorer and poorer, right? But, I mean since it has
the advantage of the tuning and it has the advantage of the low sensitivity, this structure
is preferred over the other signal amplifier to the two amplifier structures.
Now, let us go back to sample and hold circuits. Let us look at that.
You see, this is a simple sample and hold circuit. In sample and hold circuits it is very
much necessary in many signal condition. One of the good examples of the sample hold
22
circuit is a DC. Now, analog to digital convertors we have, you know there are various
types of analog to digital converters. We have successive approximations analog to
digital converters, we have the counter ….., we have dual slope adc, we have a flash adc,
there are so many adc’s are there, enhanced dual slope adc.
Now, all these cases I assume that during the conversion whatever small it may be, all
the, all the adc has some conversion time, we are assuming that the, during the
conversion the signal remains stationary of value. Its value is not changing that is to be
assured by the sample and hold circuits, right? So, sample and this is one of the simplest
form of the schematic of the sample and hold circuit. This is a switch and we have a
capacitor here. When I close the switch, the, it is charged, the capacitor, the capacitor is
charged by the input voltage and when I open it, it will hold for that for some period,
right, whatever small it may be. So, during that times I can make my conversion …., my
conversion complete; let us look at, sorry.
Now, voltage across the capacitor tracks the input signal during the time Tg, when the
logic control gate S is closed. This is a logic control gate. I can use some MOS gates also,
right, for this purpose. The capacitor holds the instantaneous value attained at the end of
23
the interval Tg when the current gate opens S, right, fine. So, during Tg, so it holds the
charge. The switch may be a bipolar transistors or a MOSFET controlled by V gating the
signal voltage or a CMOS transmission gates.
Now as you see, this is an example. So, I am using one, I mean MOS gate. Here I am
using one buffer amplifier at the input side, another buffer amplifier at the output side,
right? So, this gates, this analog signals, so whenever this gate voltage if it is NPN,
suppose if it is a MOSFET, I mean MOSFET, if it is the signal is like this, what will
happen? So, whenever this voltage is positive it is open. So, whatever the charge it has,
voltage has, it will be, come to this and capacitor will be charged to that voltage.
Once this gate voltage is removed, so that voltage is supposed to stay here. So, during
that time if the V 1 change and that will not affect this output voltage or the voltage on
the capacitor C. That is the basic theory of the sample and hold circuit.
24
Now, positive pulse at the gate of NMOS if I assume it is NMOS, so in the gate I have to
give positive pulse, positive signals or control voltage. If I use a PMOS I have to give
negative voltage or negative signal at the gate of the NMOS or of the PMOS. The hold
capacitor, holding capacitor C will charge to the instantaneous value of the input voltage
with a time constant of R 0 plus r DS ON multiplied by C. You know the time constant is
basically R into C, so where R 0 is very small output resistance of the input op-amp A 1.
Assume that the output obviously this op-amp has a very low output impedance.
Now r DS ON is ON resistance of the MOSFET; I am writing FET is the MOSFET, ON

resistance of the MOSFET, right?
25
In the absence of the positive pulses, the switch is turned off and the capacitor is isolated
from any load through the op-amp A 2 and thus will hold the voltage impressed on it. So,
because there is an output capacitance, obviously it will isolate the, our utput circuits, this
sample and hold circuits will be totally isolated from the, I mean subsequent circuits
which follows after the, which follows the amplifier A 2, right? So, this charge will be
hold, because there is no path to discharge, so that charge should hold, if there is no
leakage in the, obviously in the capacitor. So, the capacitor is the crux or the heart of this
sample and hold circuits. We will discuss what are the different types of capacitors
typically used for making a sample and hold circuits?
It is recommended in discrete component or hybrid circuits a capacitor with a

polycarbonate, polyethylene, polystyrene, Mylar or Teflon dielectrics be used. This is
because all these names actually came for the capacitors for the typical dielectric. This is,
dielectric switch is used for the, which is placed inside the plates or a parallel plate
capacitors. So, we are talking of the, not necessarily it will be parallel but it can be, I
mean foil type also that as it happened in the case of electrolytic capacitors.
26
In most other capacitor than mentioned above, there, there arises two major problems.
They are polarization phenomenon which causes the time constant to decay with a time
constant of the several seconds. In effect this is the leakage resistance of the capacitor.
Dielectric absorption which causes a capacitor to remember a fraction of its previous
charge that also creates the problem.
27
Even if the polarizations and absorption do not occur, the OFF current or the, I mean
current of the switch is less than 1 micro ampere less than and, and the bias current of the
op-amp will flow through the capacitor C, right? That is the problem. Since maximum
input voltage, input bias current is less than 1 micro ampere it follows that with a
capacitor .5 micro Farad the drift rate during the hold period will be less than 2 millivolt
per second.
There are two additional factors which influence the operations of the circuit. They are
the aperture time typically less than 100 micro second which is delay between the time
when the pulse is applied to the switch and the actual time the switch closes. Aperture
time typically less than 100 micro second which is the delay between the time when the
pulse is applied to the switch and the actual time the switch closes. Acquisition time
which is a time that the capacitor takes to change from one level to holding voltage to the
new value of the input voltage after the switch is closed, is not it?
28
Because you see what will happens that is capacitor, suppose I have some voltage, I have
some voltage on the capacitors, so we should charge it. Suppose its voltage is changing,
so it will from this place for during that it remains constant, so it will go again to that
value, we are talking about that. Acquisition time which is the time that the capacitor
takes to change from one level of holding voltage to the new value of the input voltage
after the switch is closed.
29
Now, next we will discuss the logarithmic amplifier. Logarithmic amplifier is basically
an amplifier which will give the signal the output, this one. So, the output should be some
constant multiplied by logarithmic of the voltage V s that is the function of the
logarithmic amplifier, right? So, the output is a logarithmic of the voltage V s. This is the
basic principle of logarithmic amplifiers. We can use two transistors and two amplifiers
to make the logarithmic amplifier. The circuit analysis shows that is basically used, these,
the diode characteristics of the, because you know diode characteristics is exponential
characteristics of the voltage current relations of the diodes that means we will utilize to
make these logarithmic amplifiers.
If I make the circuit analysis you see here assuming that i B1 is less than, than less, much,
much less than i C1 that is quite obvious in all the cases of the transistors and in the
transistor 2 also i B2 base current is much, much less than the collector current, in the
transistor 1 the base current is much, much less than collector current, in base two,
transistor 2, base current is much, much less than collector current, we get V naught equal
to minus V t R 3 plus R 4 by R 3 natural log of V s by R 1 into R 2 by V R. So, obviously
you can see that if I assume this R 1 R 2 V R are all constant, so only variable here is the
30
V s, because V t is also constant, but for a particular room temperature this V t will also
remain constant.
So, you can see that V naught if I assume this R 1 R 2 V R reference voltage all are
constant, so the, this output voltage is, is a logarithmic of the voltage V s which we have
impressed, right, which we have applied at the input of the circuit. So, this is basically the
logarithmic amplifier. Similarly, since we have a logarithmic amplifier we can have
antilogarithmic amplifier also, let us look at.
So, an anti-logarithmic amplifier circuit looks like, this is basically same. You can see
two transistors and two amplifiers A 1 and A 2 and two resistance R 4 and R 3 at the
initial side. So, V s output will be, output will be antilogarithmic of the voltage which
you are giving here.
31
So, let us if you make the circuit analysis I can write that V naught equal to R 1 into V R
by R 2 exponential of minus V s V t into R 3 R 3 plus R 4, right? So we can see this one,
this is a negative logarithmic of the, so this is a, obviously this is, you can see this is a
antilogarithmic amplifier.
32
Now, analog switch is extremely used in signal conditioning circuits. Analog switches are
the two switches one NMOS and one PMOS. I think this figures I have to modify.
Actually it looks like this and this is, this, this is NMOS, this is PMOS or I can use the
symbol like this as it happened in the case. This is NMOS and in the case of PMOS there
is no arrow or you can write like this one, right? This is in the case of PMOS. I am not
showing the substrate, so here bulk and now the, if you look at this NMOS PMOS these
all basically called a CMOS switch. It is extensively used in the signal conditioning
circuitry in the multiplexing, in demultiplexing.
Now what is this PMOS? Why should I use two MOS in parallel? Because if have to give
plus V g here I have to give minus V g here to turn on these two transistors.
There are some advantages. You see this is a complementary MOSFET switch or CMOS
switch, a combination of PMOS and NMOS in parallel, right? Now, ON resistance of the
MOS is a non-linear function of the voltage drop across FET. That means V i minus V 0,
so the non-linear function of the FET. So, this can be linearized by using in one case it is
increasing with the voltage, in another case it is decreasing with the voltage. If I use two
MOS op-amp in parallel, so this will be nullified and the voltage relations, voltage and or
33
turn ON resistance relations will be perfectly linear or it will remain, sorry, I should, it
should remain constant. So, as V i increase resistance of T 1 increases and that of the T 2
decreases. Thus the ON resistance of the CMOS is linear and is independent of V i and is
valid for both V i positive and V i negative, right?
Now, this is the symbol of analog switch. You can see this is a bidirectional analog
switch. We have V i, we have V o and there is a control signal. This, when this signal is
ON that means this is the signal MOSFET is turned ON. So, whatever the signal, analog
signal that will appear at the output. This is, please note this is not an inverter switch or
anything; it is analog switch. So, whatever the signal voltage here, whatever is the level
of the voltage, it does not matter, if this signal is ON that signal will appear at the output.
If you remove this signal V c, so whatever the signal it does not matter that will not
appear at the output of the analog switches.
Now, using this analog switch I can use a multiplexor and demultiplexing. Let us look at
that.
34
Names of the, there are some commercial, excuse me, there are some commercial analog
switches. These are National semiconductor switches 4016, 4066. These are all quad
switches, I mean four switches in one chip, right and from Motorola we have 14066. This
is also the quad switches and that means one for, so it should be, it is a 14 pin DIP; in
both the cases 14 pin dual inline package, right? In one case we will find that the, in each
cases we will find there are, three pins is necessary - one for input, another for output,
another for control. So, 12 switches done, so 2 more switches one for V SS, another one
for V DD. So, it is 14 pin DIP which …… analog switches, right? So, both the National
semiconductor, Motorola specifications we have given.
35
Now, it is a bidirectional analog switch. It is a 14 pin DIP. There are four switches on one
switch. These all are the specifications, one V DD and one V SS line. ON resistance of
the chip, analog, I mean switch is around 100 ohm to 600 ohm. This value is little low, I
mean you can keep, nowadays it is even less. So, input signal lie between V DD and V
SS, where V DD plus volt and V SS will be minus 5 volt.
36
Now you see, this is a one of the typical example of the multiplexing and demultiplexing.
Now you see what is the need of multiplexing and demultiplexing? Now, basically this
multiplexing demultiplexing is necessary to reduce the number of cables and also the
number of hardwares. Now all of you know, most of you are using the calculators. In the
calculators we have, nowadays as we can have afford, you can have a 10 digit display,
you can have a 12 digit display, all those things.
Now it will be surprising to know, you always from, if you have a BCD signals, BCD
signals you have to convert to the, I mean display. So, I need ….. BCD to seven segment
decoder. Now, my BCD’s signal, BCD chip should be the same, same as the display unit.
But, I can use a single decoder BCD to seven digit segment decoder to convert the signal
to the seven segment display. This it can be very conveniently do if you can scan the
signals, if you can multiplex and due to persistence of visions if you look at the
calculator, it is we never seen that it is, it is not blinking, is not it? Due to persistence of
vision it appears that it is constant. It is not flickering. Actually it is flickering that means
one digit at a time they are displaying on the, on the display board, right?
Similarly in the case of, suppose in the case of ADC we have also multiplexing.
Demultiplexing, one of the good examples of this, this type of ADC is 0809. In 0809
ADC you will find it is a basically eight bit ADC. Now it has an input. I can, I can
connect eight different input to this ADC. Even though ADC is more, only one ADC,
inside the chip, but I can take the signals from eight different sensors. I can, from eight
different thermocouples or eight different RTD I can take the signals, because it is
multiplexing that signals one at a time and every time it is only one A to D converter
converting the signals. Similarly I need one demultiplexor at the output also. All these
thing is necessary to reduce the number of cables.
This is very important because the cable cost is a lot, because especially if you use a low
capacitance cable I mean sorry, I mean, I mean where the capacitance capacity loss is less
we will find that ……. it is very expensive, right? So, to avoid that I always used
multiplexing and demultiplexing; let us look at that. You see here this is basically analog
37
switches; four analog switches are there. You can see here I have analog switches. These
are all analog switch. You can see this analogs, bidirectional analogs, so inputs are
coming V 1, V 2, V 3, V 4 and we have a control signal one. This is, these are all control
signals which is going here, right and we have a buffer amplifier. At the input we have
buffer amplifier.
We have drawn broken lines, lines might be very long it is a data lines and you see what
we are doing here. I need a two to four line decoders. What is the two to four line
decoders? Only one line will be selected at a time. When this line, this switch will be
selected all other switch will be deactivated during that time. This is very, it is not very
difficult.
You see I can have a, this type of signals. I can have a signals like this, you see here. So,
this is the switch. This is V 1, for V 1 I should say or control V 1. This is control V 2,
control V 3, control V 4. You see, when this is ON, only when, when this is high, only
the switch 1, analog switch 1 will be selected all other will be deactivated, right? So, if
we have a shift like this one, quite obviously I can achieve this type of signals.
38
What is that? You see that when I am giving V 1, control signal to V 1 which I have
shown just right now, so all the signals, all the, even the signals are present at the input
that will not come to the output, because you see the, all the outputs are connected
together; all the outputs are connected. So it is, suppose that whenever this signal is ON,
when this switch is ON all other switch will be deactivated. That means the signal will
not come. So, it will be coming through this network. It is, again it is demultiplexed the
outputs. So, V 1 will appear as a V 1 dash, V 2 will appear as V 2 dash, so and such sort
of sampling I am doing here. So, similarly when V 2 will be ON, what will happen you
see, this will be deactivated. So, this is connected at the input of the op-amp like this one.
So, we have 2 to 4 line decoder here; you see we have 2 to 4 line decoder.
So, what is the function of 2 to 4 line decoder? So, we have a 2 bits signal, because in the
2 bits signal what are the different states I can have? You know, I can have 0 1, 0 0. I can
take this one, I can have 0 0, 0 1, 1 0 sorry, I have 0 0, 0 1, 1 0, 1 1. This is the signals of
2 to 4 line decoders, because mod 4 counter have 4 different states. Since I have four
lines, I need mod four counter. So, after 1 1, again it will go to 0 0. When 0 0 will be
activated, so 2 to 4 line decoder. This is very simple. I can do with, with, with AND gate.
So, this gate will be selected, so and all other gates will be deactivated. When this is ON,
39
so this will be OFF, right and when this is, when 0 1 will be there, so V 2 will be, this
signal will be selected, when 1 0 will be there this signal will selected, when 1 1 will be
there this signal will be selected, right? Same signal should go to the output also, right?
So, this is the basically 2 to 4 line decoder
Now, one question that immediately arises then what is the use of this type of circuits?
You see that if I do not use any multiplexing demultiplexing what will happen? I have a
four signals V 1, V 2, V 3, V 4 that will be transmitted. So, four lines I need, but I need
lot of hardware. You see the, so what actually I did? I achieved that I have saved only one
line. In addition I need so many counters, decoders, all those things, right? So, counter it
is necessary, decoder is necessary. I can manage with one decoder that is fine. I can just
with, from the output of the decoder to go to the both the input and output that is not a
problem, right?
Now the question where, where is the saving? Saving will be there if the number of lines
are more. If suppose if I have 12 lines, I mean 12 different sensors, in that case how many
lines is necessary? If I have 12 different sensors, I need four lines for this one and one
lines, five lines. So, I am saving seven lines. In that case seven cables, I am in that case.
So, that is the advantage of the multiplexing and demultiplexing, right? So, people
always, but you know that we have to, one to one synchronization should be there. That
means when these lines will be selected, so simultaneous this line also should be selected
during that. It is a not very difficult, because same pulse, same mod four counter output is
coming to the both the decoder. So, when this will be selected, this will be selected and at
that time all these three analog switch will be deactivated, This analog switches also will
be deactivated. When this switch is selected this switch also will be selected, like that it
will go, right?
40
So, in the circuitry discussed above much of the cables can be saved in the, the, as the
number of switches increases. Now you see that I am giving some inputs signal V 1 of
different varieties of the signal V 1, V 2 switches, slowly varying with time, increasing
with time which is not varying with time. Now, if I sample this and some signal V 3
which is coming down and V 4 is also less value it is constant.
41
See, if the signal looks like this one what is the sampled signal? Sampled signal should
look like this one. This signal will not change.
Now this signals ….. this is coming like this, this is coming like this. Now, what will
happen actually? You see that if I pass this signal, how will you retrieve the original? My
original signal does not look like this, is not it? How does my original signal look like?
My original signal look like this, original signal look like this. So, this signal that mean V
1s, V 2s, V 3s, V 4s should pass through a low pass filter. If I pass this signal through a
low pass filter, so obviously what will happen? This, all these things will go out. Actually
I will get a good output which is very close to the input signals.
So, I can use a simple RC filter also at the, RC low pass filter at the output of the, output
of the multiplexors. That means at, at this point you see here I can use some output
signals. I can use some capacitors also, so which will reduce that high frequency
components, so I will get a signals which is very similar to the input signals I have, which
I have given here V 1, V 2, V 3, V 4, clear? So, this is the advantage of our multiplexing
and you can see here, so if I pass this signal through a low pass filter, obviously I will get
the original signal.
42
Now, on passing these through a low pass filters we will set back the original signals that
is quite obvious that I can get the original signal. So, obviously you can see the same type
of hardware is also used in the case of display, digital displays where they are, we are
saving the seven segment decoder, because that is a great saving. Even though those are
in LSI, large scale integration, so the saving is there that with a single decoder I can and
due to persistence of vision here also, see in many, in many cases I do not need sampling
that fast, right? I do not have to monitor the temperature. Suppose I have a 100
thermocouples, I do not need to, I mean monitor the thermocouple temperature every
minute, something like that.
I can sample, suppose if I want in a 5 minutes that is more than enough. So, in that case
accordingly I can make the circuit so that which will give you the, I can, I can sample the
circuit like that. So, this is the advantage that in that case what I will do that I will save
the, basically the hardware which is necessary to, I save the hardware which is necessary,
I mean which, which is the great, which will reduce the cost of the entire systems or
entire equipment. So, that is the advantage of the multiplexing and demultiplexing,
whereas in the case of filter switch you have seen that is at the very beginning of the
lessons we have discussed, you see the basically general emitter converter or inductance
43
simulators, sometimes it is called the inductance simulators, there are various types of
name.
We call inductance simulators or gyrators. That means by using a capacitance I can use a
inductance, by using an inductance I can use a capacitance. As I told you, nobody is
stimulate, I mean nobody simulate the capacitance with the help of inductance, but it is
quite obvious people always try to simulate inductance with the help of capacitance,
because having a real capacitance, real inductance in the circuit is very difficult. It is
neither it is possible so far as for today to get a large value of the inductance and also we
can have a, I mean if we want to do the simulation I can have a floating inductance as
well as a we have a grounded inductance that means two grounded inductance if we
connect back to back, I will get a floating inductance of the simulator, floating inductance
circuits. So, these are the things which we have discussed.
We have also discussed the signal amplifier structures. During the lesson 22, we have
discussed the two amplifier structures in the lesson and we have discussed also the, I
mean two amplifiers as well as three amplifier state variable. Now, state variable
structure is very attractive. Two amplifier structure has no advantage, additional
advantage except that in the band pass cases it keeps the very high, very high input
impedance. Only problem with the state variable structures we have seen that that
because you see that the high pass structures, high pass, high pass or high pass notch or
low pass notch, this type of filter functions is achieved by making the, the coefficients of
s in the numerator zero and we have seen that that is actually done by the signal
subtractions. So, if due to mismatching of the any resistance or capacitance it does not
make zero that creates the problem.
44
That means in the case of, as you know in the case of notch circuits as you know that we
have always you know what will happen the, if you look at this is sigma j omega plane.
We have a complex conjugate poles in the case and either I can have a zero here or I can
have a zero here, here or here, here. Now, if it does not cancel that means there is no real
part of the zeros, so that is only possible if the s part that means coefficient of s in the
numerator is, is absolutely is zero, otherwise there will be no real part.
Now what will happen if the real part remains due if, if due to signal subtraction it is, it is
not compressed, what will happen you will find that it will remain. So, this will be shifted
to little bit on this side, so that will create the problem. So, the pole zero location will not
be same, not exactly what is desired and also by, by that the exact notch functions may
not be achieved. That is the only drawback. But nowadays as you know, the signal
conditioning, the component tolerances are less and less going to be so that it is very easy
to get the, whatever that is the value of the or what are the, whatever the design value of
the resistance or capacitance that we can achieve very easily. So, with this we can come
to end of the, so you see that is how does the signal comes so that I can staircase wave
forms. So, with this I come to the end of the lesson 23 of Industrial Instrumentation.
45
Welcome to the lesson 24 of Industrial Instrumentation. In this lesson we will consider

piezoelectric sensors. Now, as you know piezoelectric sensors is based on piezoelectric
crystals, right? So the, it has a property that if you apply a force across the surface of this
piezoelectric crystals I will get a voltage and this process is reversible. That means if I
apply the voltage I will get the force also. So, utilizing these principles obviously I can
make the sensors which can measure force and piezoelectric crystals is extensively used
for, as I told you that is reversible that is it is extensively used for generations of the
ultrasonic waves.
As you know that the ultrasonic sensors are, are, is used, ultrasonic sensors are used
extensively in the case of flow measurements and all these things like ultrasonic flow
meters are there. So the, you have seen that actually there we are using the, to launch the
ultrasonic signals, I, we want that ultrasonics, I mean piezoelectric sensors, piezoelectric
crystals actually we have used there, so it is another use. Also, the piezoelectric crystals
as you know it is used for measurements of, for the generation of the very stabilized
frequency, because it is, piezoelectric crystal if you, we will see later on that it has, if you
draw the equivalent circuit that you will find that it has a very high selectivity.
It helps to make the, make the oscillator which is very stable in frequency. These are the
all different applications of the piezoelectric sensors. So, we will discuss one by one what
are that and piezoelectric crystals we will find that I, I need a, a special type of amplifier
to those, to amplify the charge generated across the plates of the piezoelectric crystals.
46
The equivalent circuit is like this one. You see, I have a inductance. Let us look at, I have
an inductance L. Then I have an inductance L, I have a resistance R, C and also a parallel
capacitance C dash. This value of L is very, very high in the case of piezoelectric
crystals. That makes the frequency stability excellent otherwise nobody should care for
these quartz crystals, right?
So, in this lesson we will see that at the first part of the lesson we have discussed the,
basically the piezoelectric sensors which is used for the measurements of the
displacements, accelerations, ultrasonic and which can be used as the ultrasonic
generations of the frequencies or ultrasonic waves and also it can be used in the crystal
oscillator and the second parts we have discussed a very important, a pneumatic system
which is called the flapper nozzle systems with feedback systems, right and it is very fail
safe devices as you can, if you compare with electrical devices it never fails actually, this,
except the routine maintenance of the orifice, because entire operation depends on the
orifice width also. So, diameter of the orifice also is very important.
Over the long use it should not, it should not widen, so that you have to change, you have
to make the change in entire calibration, otherwise it is a just fail set device. It never, it
47
never happens that it is not working or it fails. So, that is the great advantage of this type
of pneumatic system. For this reason it is used over the years in the industry and still it is
used in some of the industries like hydrocarbon industry, where high voltage is restricted.
That means I cannot use a voltage or any devices which has a voltage more than 40 volt,
because you know there is a, there is a large, you should have a large actuator or large
valve control bulb.
I need a voltage of at least of 220 volt to create a large torque. That is not possible
because that voltage is not allowed in the hydrocarbon industry. Though the transmitter
side of the electrical that is fine, but the basic actuator, basic sensor there are the basic
pneumatic, because it is totally hazardous free, right? Because so many supply we have
instead of electrical supply, so we have a pneumatic supply there, right? So, with this I
come to the end of the lesson 24.
48
Prof. Alok Barua
Lecture - 24
Piezoelectric Sensors

consider piezoelectric sensors. Now, as you know piezoelectric sensors is based on
piezoelectric crystals, right? So the, it has a property that if you apply a force across
the surface of this piezoelectric crystals I will get a voltage and this process is
reversible. That means if I apply the voltage I will get the force also. So, utilizing
these principles obviously I can make the sensors which can measure force and
piezoelectric crystals is extensively used for, as I told you that is reversible that is it is
extensively used for generations of the ultrasonic waves.
As you know that the ultrasonic sensors are, are, is used, ultrasonic sensors are used
extensively in the case of flow measurements and all these things like ultrasonic flow
meters are there. So the, you have seen that actually there we are using the, to launch
the ultrasonic signals, I, we want that ultrasonics, I mean piezoelectric sensors,
piezoelectric crystals actually we have used there, so it is another use. Also, the
piezoelectric crystals as you know it is used for measurements of, for the generation
of the very stabilized frequency, because it is, piezoelectric crystal if you, we will see
later on that it has, if you draw the equivalent circuit that you will find that it has a
very high selectivity.
It helps to make the, make the oscillator which is very stable in frequency. These are
the all different applications of the piezoelectric sensors. So, we will discuss one by
one what are that and piezoelectric crystals we will find that I, I need a, a special type
of amplifier to those, to amplify the charge generated across the plates of the
piezoelectric crystals. So, let us look at the piezoelectric sensors and principle of
operations.
1
So, contents is not there. Contents as I told you, we will have, we will discuss
basically the piezoelectric, piezoelectric sensors, its principle and charge amplifier
and the crystals, piezoelectric crystals as it is used in making, making the sinusoidal
oscillators, right? So, principle of operations are piezoelectric crystals, let us took at.
The piezoelectric sensors are based on the principles of electromechanical energy
conversion. The mechanical input is converted to the electrical output and that is the
basis of this transducers and this transducer shows piezoelectric effect. Actually, this
is actually piezoelectric effect. That means if I apply some force, I will get some
voltages. This particular effect is called the piezoelectric effect. Not all the material
has some, this effect we will see that there are some synthetic material which has this
property. Also, there are the, there are some natural materials which has this property.
2
Piezoelectric effect - what is the piezoelectric effect? When certain solid materials are
deformed they generate within them an electric charge. This effect is reversible and
thus if a charge is applied the material will mechanically deform in response. These
are known as piezoelectric effect. Exactly this thing we did while we are making the
ultrasonic sensors. Instead of steady voltages what we are giving? We are giving a
time varying voltages across ……. Quite obviously the force which we will get that is
also time varying signals and that can be, I mean that can be order of mega hertz
range, right, so that we can get the ultrasonic signals from the piezoelectric, I mean
sensors.
3
Applications - measurement of acceleration, force; also, it is extensively used in the

vibration analysis. That is the only sensor which is used for vibrations, for crack
detection, for thickness measurements, all those things you will find that it is used.
For generating high voltage, because momentarily it will make a very high voltage
even though, I mean, is the current capability is very small but I can generate a very
high voltage. If I apply sudden high pressures or impacts on the piezoelectric crystals,
obviously the high voltage will be generated; then ultrasonics, right? These are the
different, also we have not mentioned, it is also extensively used as I told you earlier
at the beginning of the lesson, for making the crystal oscillator. A piezoelectric
phenomenon, let us look at historic background of these particular sensors.
4
In 1880, Curie brothers discovered the direct piezoelectric effect in certain solid
crystalline dielectric substances such as quartz. Quartz is a natural substance. We will
see that quartz is not only the material which has a piezoelectric effect, there are many
synthetic materials I mean which found later on, I mean or the scientists have
developed. We will see that in those piezoelectric materials, in those materials you
have, you can get the piezoelectric effect, right?
5
Now, under stress free condition what will happen? What is under stress free? Let us
look at the diagram. Under the stress, stress free condition what will happen to these
piezoelectric sensors.
You see, this is a total, I mean this is silicon dioxide crystals. You see what will
happen here? I have silicon. This is positive and this x-axis and this, I mean this is the
silicon ion, this is the oxygen ion. So, under stress free condition it will look like this,
right? Now, what will happen? Now, if a stress is applied on these particular crystals,
let us look at? When the force along the x-axis is applied, so I will get a response like
this one. If I take this is your x-axis, this is as y-axis you see that if I apply some force
what will happen let me see.
Yes, you see, I am applying a force here, force here ……. I am applying force here,
force here also that obviously if I put on a, obviously it will, on the both sides that
means on both sides I am putting a pressure. So, what will happen you see that this
positive and negative ions will be distributed like this, right? This is along that x-axis
if I apply the pressure. Now what will happen if I apply the force in the y-axis? You
see this is the y-axis and this is the x-axis. These are two plates A and B.
6
Charge is developed on the two faces A and B, right? You can see that charges have
been developed, charges have been developed across plates A and B, right? Positive
charge will be developed here, negative charge will be developed here, right? The
above is known as longitudinal effect. This is called the longitudinal effect. We have
also transverse effect. Let us look at that.
7
When the force along the y-axis, if now, suppose if we apply the force along the y-
axis, then what will happen? If I apply the force along the y-axis let us look at, so I
am applying force in this direction, in this direction. So, this is my y-axis and this is x-
axis, right? But my plates are, I mean I am collecting the charge from the, from the
plates which is placed on the x-axis, right? So, what will happen you see the positive
charge will move in this direction and the negative charge will move in this direction.
So, this will be the positive plate and this will be the negative plate, right? So the, the
stress which I am applying this is sigma y, because the sigma x if we can look at, this
is sigma x, because this is applied here, right, applied here, applied in this x direction
and since it is in y direction we are applying, this stress is sigma y, clear?
Now, charges are developed on the two faces A and B; same the charges in the
previous case when we apply the x-axis, along the x-axis or across the x-axis the
charges will be developed is across A and B and when we apply the stress in the y-
axis, the charges developed will also be on the, across the plates A and B. The above
is known as transverse effect. One is the longitudinal effect, transverse effect. You all
familiar with this longitudinal and transverse, because we have extensively, these
points have been discussed while we have, we have, while in the, we have, we have
studied in the lesson on the stress, on the strain gauges, right?
8
Now, materials - what are the different materials which is used, I mean for the
piezoelectric crystals, because you know the, previously the, I mean this only quartz
people thought of that is only have the property. Quartz is quite expensive also.
Quartz has piezoelectric, I mean property. So, but later on we found that there are
some synthetic materials also which has piezoelectric property. Let us look at that.
The materials that exhibit or that show a significant and useful piezoelectric effect fall
under three main groups.
There are three main groups we can categorize. One is the natural quartz and Rochelle
salt are the two. Quartz is very commonly used and Rochelle salt is also used. Then
we have synthetic lithium sulphate, ammonium dihydrogen phosphate. These are the
two different synthetic materials. It has also, these, these two materials also have the
property of the piezoelectric effect. That means if I apply pressures, voltages will be
developed and if we apply the voltage, pressure will be developed or force will be
developed across the plates.
9
Also we have polarized ferroelectric crystals - these are the barium titanate and lead
zirconate-titanate. These are also coming under the synthetic, but the property is
something different that is we have put under different categories, right? Because of
their natural asymmetric structure, the crystal materials other than the ferroelectric
crystals, ferroelectric crystals, it show the effect without further processing. We do
not need any further processing. We will see that if I have natural asymmetric
structure, then the crystal material other than the ferroelectric crystals show the effect
without further processing. However for ferroelectric crystals they need to undergo
certain processing. What are those?
10
They must be artificially polarized by applying a strong electric field to the material,
while it is heated to the temperature above the Curie point of the material. They are
then slowly cooled with the field still applied. When external field is removed they
have a remnant polarization which allows them to show the piezoelectric effect or
exhibit the piezoelectric effect. This is the thing we will have.
11
Now, the piezoelectric transducer if I look at, actually is naturally available
transducer. You see that how the, its, what are different axis? We are taking one you
see these are all, I mean in this direction three, I mean this, all this, this arrow is
actually the shear stress which is developed across the different axis of the
piezoelectric crystals. These are the force typical tensile or compressive. These are the
shears. So, direction 1, 2, 3 are the compression or tension either compressive force,
can be tension or tensile force; both cases you will get the piezoelectric effect and 4,
5, 6 are the shear stress on the piezoelectric crystals, right?
Now, see this is the typical piezoelectric, I mean I should say the schematic of the
piezoelectric crystals. As it looks like, I can take the axis like this is x-axis, this is y-
axis, this is z-axis or this is 1, 2 as it happened here let us see, so 1, 2, 3. In this case
also I can show that this is 1, this is 2 and this is 3, right, like this one. So, this is 1,
this is 2 and this is 3. So, we have a shear also, right? So, the dimensions you see, the
width of the crystals is w, thickness is t and its length is l, right? So, we have two
electrodes. We have seen the two electrodes, one in this and another one this. We are
taking the voltage e 0, right? We apply the force fi across these plates, right?
12
Now, there are two families of constants - g constants and d constants which are used
to describe the piezoelectric effect, right? We will discuss two constants. These are
necessary or I should say the parameters of the piezoelectric crystals by which we can
define the piezoelectric crystals. So, that is the reason I am telling there are two
families of constants, g constants and d constants which are used to describe the
piezoelectric effect. So, g ij and d ij what are these? Where i is the direction of electric
effect and j is the direction of mechanical effect, fine.
13
So, g constant is defined as g 33 field produced in the direction 3 divided by stress
produced, stress applied for direction 3, right? So field produced that is I mean
voltage e naught divided by t thickness and stress means force applied f i across that
divided by area, is not it? You see, the force applied is f i and what is the width? w
into l that is the area of cross section. So, if I divide f i by, so I will get a stress f i
divided by w into l, I will get stress and e naught divided by t, so I will get the field, is
not it? So, e naught by t upon f i by w l.
Now, d constant is defined as d 33, charge generated in direction 3 divided by force

applied in direction 3 which can be written as Q divided by f i, right? That means
charge developed divided by the force applied.
Moreover, P is equal to d multiplied by sigma, right, where P is the polarization and

sigma is stress and d is called the sensitivity matrix of the piezoelectric crystals, right?
14
Equivalent circuits of piezoelectric, I mean for this one you see like this one, so we
have a crystals. Then we have a, you see that this R crystal, this is the charge, I mean
equivalent circuit in not that sense, I mean if I replace this, I mean piezoelectric
crystals by a signal generator, the, the circuit will look like this one. So, I have a
signal generator, voltage source of q, I mean charge source. Then we have the
resistance R crystals in parallel with the capacitance that means C crystal or C.
So, crystals we are taking the first, I mean three letters, crystal, so I am taking first
three letters cry, clear? Now, cry is the leakage resistance which is extremely high. It
is 10 to the power 12 ohm, 1000 giga ohm I should say and C crystals is equal to
epsilon into w l by t. Typically, I mean dimensions of a, of a expression of a
capacitance, right? W l is the area of cross section of the capacitance. If I consider it
as a parallel plate capacitor, t is the thickness and epsilon is the permittivity of the
medium, right or dielectric constant of the medium and q is the charge generated,
right? So, the q charge generated, R crystals - leakage resistance, which is around
1000 giga ohm and C crystal equivalent, I mean capacitance of the crystals is equal to
epsilon w l divided by t and q is the charge generated.
15
Measuring circuit loop we have like this one. We have a crystal, we have a cable, but
there must be some cable, some wire is to be connected and it is connected to the
amplifier, so that I will get the output e naught.
Now, this is the complete circuit of the piezoelectric crystals along with the, if we use,
if you think of the charge generated along with an amplifier. So, all the impedance of
the amplifier also should come in the picture of the equivalent circuit, right? So, here
16
you will see, we have q, R crystals that means leakage resistance of the crystals, C cry
capacitance, C is the cable capacitance which is coming across, R is the, is the
amplifier impedance and is the input impedance, capacitance of the, because as you
know in, any amplifier has capacitive range at which actually you will get the that
particular input impedance. So, that impedance, capacitance must be mentioned also
in the equivalent circuit.
Now, you see we have, we have simplified the circuit. What we have you see here we
have two capacitance, three capacitances and two resistances. This we have combined
to an equivalent capacitance. We are writing C and R and what is C and R let us look
at. So, this is C, all the capacitance are combined here and all the resistance are
combined here. We are getting the output voltage e 0 and charge generated q, right?
Let us look at. So, charge generated will be q. So, C equal to C crystals plus C cable
capacitance plus C amplifier, all will come in parallel. Then, since it is in parallel, all
will be added, all the capacitance value will be added and R is the, R crystal resistance
in parallel with the amplifier resistance.
17
So, e naught I can obviously write small i into R equal to dq by dt minus C into de
naught by dt whole multiplied into R or RC equal to RC multiplied by de naught by dt
plus e naught equal to R dq by dt. So, e naught by q because q is the input, I mean we
assume that q is the generated and what is the output let us look at. So, the sensitivity
will be e naught by q s in s domain R into s divided by RC s plus 1, right?
18
This, where q equal to d into f, where e naught by f s that is e naught because as you
know q is equal to d multiplied by f, where e naught is equal to e naught divided by f
that is actually, this should be, that should be the sensitivity, because actually we are
applying the force, force is generating the charge, right? So, but so actual sensitivity if
we want to find we have to find e naught by e naught by q, we have to find e naught
by f actually the output voltage divided by the force which we have applied equal to d
into R into s divided by RC s plus 1. This we can write d by c into tau s by 1 plus tau
into s, where tau equal to R into C, clear? The above represents a high pass transfer
function of this.
So it looks like, it is not possible to measure any static force with this circuit. Please
note the static is not possible. So, in dynamic whenever there is a change in the force,
then only I can measure, I will get the output. Otherwise we will not get any output.
Static sensitivity d by c will change with the change in any of the values of the C
cable capacitance, I mean cable capacitance and C amplifier. So, it is sensitive to this,
so this must be constant. For this reason we use a charge amplifier. So you see, it is
not possible to keep all these things constant. So, it is better if we use a charge
amplifier.
19
Measurement using the charge amplifier, the charge amplifier looks like this. We have
a q and we have a R and C at the input side. On the feedback side we have feedback
resistance R f and the feedback capacitance C f and we are using the operational
amplifier, right? So, this is the basic charge amplifier.
Now, from circuit analysis we can say that the dq by dt equal to minus e naught by R f
minus C f into de naught by dt. So, this will give you R f into C f into de naught by dt
20
plus e naught minus R f into dq by dt. Just algebraic manipulations we are doing or
we can write e q by, e naught by q in s domain plus e naught equal to minus R f s
upon 1 plus R f C f into s.
And e naught by f s equal to minus d by C f tau f s 1plus tau f s, where tau f equal to
R f into C f. The above circuit also cannot measure static force, but it removes the
dependencies on the cable capacitance and the resistance, right? So, what is the cable
capacitance all these things will come in the picture.
21
Now, the piezoelectric accelerometers if you look at, so the basic block diagrams of
the piezoelectric, because piezoelectric, I mean crystals are used as accelerometers, so
we can make integration of that we will get velocity, if we make another integration,
so I get the displacement, right? In vibration analysis as you know, it is extensively
used as piezoelectric sensors to pick up the vibrations. So, the basic block diagram,
because I have a seismic transducer, K stiffness of the crystal, piezoelectric transducer
this is the output voltage we are getting e naught.
22
Now, for a seismic transducer we know that that x naught by a i s equal to 1 upon S
square plus 2 xi omega n S plus omega n square, right? For piezoelectric transducers
we have e naught upon f i S into d by c tau S 1 plus tau S, where tau equal to RC and
d is the sensitivity.
Now I can write, I just make little manipulations e naught by a i S, S domain, e naught
by f i into f i by x naught into x naught by a i. Ultimately I am interested in e naught;
this will all cancel out; e naught, this will cancel out, this and this will cancel out. So,
only e naught by a i will remain. So this I, if I substitute separately e naught by f i, f i
by x naught and x naught by a i, I will get this whole expression you see; I will get
this whole expression, clear? d by c tau S 1 plus tau S into K into 1 plus S square plus
2 xi omega n S plus omega n square.
23
Now, e naught by f input S equal to 1 by M e naught by a i into S. This I can write d

by c into tau S 1 plus tau S multiplied by K by M into 1 upon S square plus 2 xi
omega n S plus omega n square equal to d by c tau S 1 plus tau S omega n square S
square plus 2 xi omega n S plus omega n square, right?
The response curve will look like, like this one. That means that the output voltage
divided by f input that means force it will look like this. This is highest point d by c,
24
we are plotting omega here. So, it is, it is almost as a band pass characteristics you can
say, with a certain hump So, after that natural frequency, because of the natural
frequency there will some hump at resonance, so after that it is falling down, right?
That 1 by tau, so I am getting the output, the mod of e naught, output voltage divided
by the force in j omega domains will be d by c, right? So, this is the response of the
piezoelectric crystals.
Now, as I told you that oscillators are different types. Now, piezoelectric crystals also
has a tremendous applications in making oscillators. We are not discussing about the
piezoelectric crystals as it is used as a generation of ultrasonic waves. When we have
discussed the ultrasonics we have seen that what are the different, how can we
generate all these things ultrasonic waves and but crystal oscillator is very one of the
common, almost all of you are using some quartz watch.
What is the quartz it is written? You see it is written quartz. It is not that just to attract
the, I mean customers and increase the price of the watch. Actually this is used for
watch or wall clock. Everywhere you will find it is written quartz. It means that they
are using some oscillators, because ultimately in all electronic, I mean watch or clock
we need some basic oscillators, right, so that will drive some stable motors, so that
will be incremented, I mean it will step change; for each signal it will change like this
25
one, so that the seconds arm will move. Subsequently it will move the minutes arm
and for the hour arm, right with the gear arrangement.
Now, but basically we see, I need an oscillator. Now if I use, what is the problem if I
use, suppose if you use a oscillator based on 555 or any other you can use two
transistors also to make an ……. oscillator or you can make an op amp to use or you
can make ……… oscillator; not necessarily I need sinusoidal oscillations there, right?
But the thing is there you know that if you make that type of oscillator, it heavily
depends on the, not only depends on the, on the type of the resistance and capacitance,
it depends on the, on the absolute value of the resistance and capacitance and as you
know that as the time goes by, so this may change, right?
This value of R and C may change, so that will change the, ultimate the clock that the
period of the clock. So, if it changes, obviously your watch or the clock will no more
remain accurate. So, for that reason people use a crystal, because as you know the
crystal has the property if you draw the equivalent circuit of the crystals you will find
this is a very high value of inductance, we know. As you know the inductance if you
have any coil, it has large inductance, obviously its q will be, I mean quotient point
will be very, very high, right?
So, if the q is high, obviously if I use some oscillator circuit, some oscillator circuit
based on this particular crystals, obviously what will be, that time you should not use
crystals. You will say the, you should, you should look at a, I mean a passive circuit
which has high value of inductance. If the inductance is very high, obviously what
will happen you know that I will get the large value of the q. So, this large value of
the q means I will get a tremendous amount of frequency stability of the circuit, right?
So, let us look at that.
You see to a piezoelectric, piezoelectric crystals with electrodes plated on the

opposite faces a potential applied between the electrodes you know forces are exerted
on the bound charges within the crystal. An electromechanical system is thus formed
which will vibrate when properly excited, right? A electromechanical system is thus
formed which will vibrate when properly excited.
26
The equivalent circuit is like this one. You see, I have a inductance. Let us look at, I
have an inductance L. Then I have an inductance L, I have a resistance R, C and also
a parallel capacitance C dash. This value of L is very, very high in the case of
piezoelectric crystals. That makes the frequency stability excellent otherwise nobody
should care for these quartz crystals, right?
27
Neglecting R, if I neglect R, the impedance of the crystal is the reactance, which can
be given as j by omega C into omega square minus omega s square upon omega
square minus omega p square, where omega s and omega p are the series and th
parallel resonance of the circuit, where omega s, 1 by LC and omega p equal to 1 by
LC plus 1 by LC dash, right? Omega s, as I just right now told, series resonance
frequency and omega p is the parallel resonance frequency. These are, one is series
resonance and parallel resonance frequency. So, what will happen actually?
Since the C dash is much, much greater than C, omega p is almost equal to omega s,
right? So, if it is omega p omega s, reactance as a function of frequency, you see, if I
plot this one, it will look like this one. This is inductive here and here you see it is
capacitive. Omega s is series resonance and omega p is parallel resonance. So, this is
the, if I plot the reactance of the piezoelectric crystals versus frequency I will get a
response like this one, right?
28
Now, the crystal reactance as well as that of the LC network must be inductive, quite
obviously. Basically this, I mean, I mean your piezoelectric crystals you can see, if
you look at from the basic point of view it is nothing but an LC oscillator. We have
studied the LC oscillators in our analog electronics. You see, simply the LC
oscillators, only thing that L is replaced by piezoelectric crystals. For the loop gain to
be greater than unity, the oscillation frequency must lie between omega s and omega
p, but closer to omega p. So, the loop gain we have seen, you see that it should lie
between omega s and omega p, but closer to omega p. If it is there, so no problem, I
can purposefully, can choose that thing.
Since omega p is almost equal to omega s, the oscillator frequency is fully dependent
on the crystals and not on the rest of the circuit. This is the most important point of
the piezoelectric crystals. See here, since omega p is equal to omega s, the oscillator
frequency is fully dependent on the crystals and not on the rest of the circuit. Because
you need rest of the circuit we will find that we have the other elements in the circuit,
but the frequency will not depend, the frequency of the crystal, the frequency of
oscillation of the oscillator will not depend on the rest of the circuit. It will depend
only on the crystal itself and it is highly stabilized. It hardly changes with time or
anything, so that is the reason I am getting a very stable frequency there. So, all these
properties happens because of, because of large inductance.
29
Now let us look at a FET crystal oscillator shown in the next, I mean we will see that
the how using a FET it is basically a Colpit oscillator.
We can see this is the typical Colpits oscillator. You can see here, so I have a crystal
here, I have a 10 mega ohm resistance and I am using a transistor 2N 2608 and this is
a tank circuit which values 61.122 micro farads with 300 Pico farad capacitances,
emitter bypass capacitor .02 micro farads, a resistance, emitter resistance 2.2 mega
ohm. So this, but please note that the piezoelectric crystal does not necessarily mean
that it will have totally independent of temperature; it has temperature dependency.
But you see the advantage of this one, this 10 mega ohm, 2.2 mega ohm capacitances,
all these things, whatever the things you see the inductance here, it is very small. One
or two turns of coil can generate this type of, this type of inductance. So, what will
happen you know that in these cases, in this entire thing what will happen? This
oscillation frequencies does not depend on the resistance, capacitance and all those
things. It will depend totally on the crystals and crystals that parameters does not
change, right?
So, obviously by this I can get a very, very stable signal from the circuit. We know, as
you know there are many other oscillators. We have sinusoidal, we have a, we have a
30
Wien bridge oscillator, because basic in all oscillator, as you know the basic mother
of the, all the function generation is a sine wave. Because, once you get the sine wave
I can …….. a Schmitt triggers. So, I can, I can make a square wave. If you
differentiate a square wave I will get a pulse. If you integrate a square wave I will get
a, I mean it looks like this.
You see that if I have a sinusoidal wave, I should take some other colour, so I will get
a square wave like this, sinusoidal wave. There are various types of oscillators. We
know Wien bridge oscillators, then phase shift oscillators, Colpits oscillators, so these
are the different types of oscillators available. Now, these signals, sinusoidal signals if
I pass through a Schmitt trigger, I will get a signal like this one, right? If I chop the
other parts I will get a signal like this one – lower part, right? If it do not have a dual
supply, so I will get … If it is zero, so it will not.
Now, if I integrate this signal how I will get the signal? I will get the signal like this,
is not it? So, I get a triangular wave. Now, if I differentiate this signal omega, what I
will get? I will get a pulse like this one, is not it? So this will, I mean this is the basic,
so basic of the, all the signals generator is basically a sinusoidal wave that we have
seen, right? But as you know that in the crystals oscillator also we are basically
generating sinusoidal wave. Then, we are passing through a Schmitt trigger, we are
31
making a square wave. So, there are many, I mean Wien bridge oscillators. Why, you
should ask, why should you go for crystal oscillator as a piezoelectric based sensors,
piezoelectric based oscillator, because of the frequency stability. In Wien bridge
oscillator or a phase shift oscillator you won’t get that type of good stability, right?
So, as you can see in this particular lesson we have discussed the, the basic
piezoelectric sensors, its structure, crystal structures, how I can apply the force, how
will you get the signals, what is the charge amplifiers that we have discussed and what
is the problem with the simple amplifiers that it is heavily dependent on the cable
capacitance, cable resistance that can be avoided if I use a charge amplifier. So, those
type of things we have discussed in this particular lesson. Also we have discussed that
piezoelectric sensors, I am not going into the vibration analysis, all those things. But,
piezoelectric crystals are also used for vibration analysis, crack detection and all those
things.
In continuation of the lesson 24, we will discuss now the flapper-nozzle systems. As
you know, flapper-nozzle is a pneumatic system. It is extensively used in the
pneumatic transmitter and all these types of things. Actually you will get a small
displacement, can be converted into a large pressure change, right and as you know
that in some industries it is still forbidden to use, use the electrical systems. Though
the transmitters are electrical, but the large torque which are necessary for controlling
32
the valve and all those things cannot be provided by the electrical system of supply of
40 volt or something and you cannot use more than 40 volt there. So that is a problem.
So, what happens there? They still use those pneumatic systems, so flapper-nozzle is a
part of that. So, I will discuss what is the flapper-nozzle systems.
A flapper-nozzle system is used for the measurement of minute displacements. A

small movement of the flapper can result in a large change in the output pressure and
hence it is treated as a displacement to pressure converter. That means displacement
can be converted to the pneumatic pressure, right? So, it is total pneumatic systems
we can see here.
Such transducers are designed for use in both hydraulic and pneumatic power
supplies. As you know, the pneumatic system is operated at an air pressure range of 3
to 15 psi, as you know that though we use SI systems in academics, still in industry
they use those terms like 3 to 15 psi. As you know that 3 to 15 psi is a standard for
pneumatic systems which corresponds to 4 to 20 milli ampere of current in the
electrical domain.
33
Now flapper nozzle system let us look at. You see, this is our flapper nozzle system.
We have a flapper. You see this is, even though we have shown it is, it is usually very
thin, right and there is a nozzle; nozzle also diameter is very, very small, right? Now,
what will happen you know that if this flapper, if this flapper goes like this, if this
flapper goes like this or this, so large pressures will be developed in this systems,
right, at here. Here we are giving some continuous supply of pressures. Now, if it
moves away what will happen? If it moves away the pressure will be less. So, this
displacement, small displacement can measure, can be measured by the pressure,
measuring the pressure at this point, right and also we can see, you can use a feedback
control systems with the flapper-nozzle, because it can have a automatic feedback
systems. That means you have, if I use a, if I, suppose if this nozzles is coming very
close to this one, automatically the pressure will increase and it will give the back
pressures here. So, it will put the nozzles in a particular position that is the feedback
systems, anyway let us look at.
34
Now you see this is our, this is the P ambient pressure, I have a flapper. This, I have
already told you this is very thin actually though it is shown, so it can move like this.
It can move like this that means flapper can move like this. So, it is very thin, so
flapper can move like this. So like this one, so it is like this one flapper, so it will
move like this. I have a hinge here, so nozzle here. It will move like this. If the flapper
comes very close to the nozzle the pressure will increase, P o will increase. If it goes
away P o will decrease. That is the basic principle of the flapper-nozzle systems and
we have also, at the end of the nozzle a fixed, I mean pressure supply or air supply,
right?
P is the ambient pressure, then x i is the displacement of the flapper, G u is the outlet
mass flow, G s is the inlet mass flow and P s, T s is inlet pressure and temperature, P
o is the pressure output at temperature T naught, obviously.
35
The output pressure of the nozzle back chamber is measured by a pressure measuring
device. As you know that you can measure by some, there are so many gauges we
have studied so far with bellow gauges or with Bordeaux gauges, by the diaphragm
gauges we can measure and that pressure can be calibrated in terms of displacement
of the flapper, right? The flapper nozzle system shown in the figure consists of a
chamber of small volume connected to a constant pressure source on one side and
vented on the other side to the atmosphere through the nozzle. Let us look at again.
36
You see, so constant pressure source is this one G s and P s, this is our constant
pressure source. Right here you see, this is our constant pressure source, this part. It is
connected to, you see this is nozzle. This is a one part of the, one part of nozzle is
free. Continuous air is leaking through this one and this is a continuous supply and
this is actually measuring the pressure which corresponds to the, this corresponds to
the position of the flapper, right? So, the output pressure of the nozzle back chamber
is measured by a pressure measuring device, as I told you. The flapper-nozzle system
shown in the figure consists of a chamber of small volume connected to a constant
pressure source on one side and vented on the other side to the atmosphere through a
nozzle.
The flapper held in front of the nozzle is used to regulate the rate of fluid flow
bleeding out through the nozzle to the atmosphere, right? There is a continuous leak
of air or bleeding of air through the nozzle, right? That is actually we are telling here
the flapper held in front of the nozzle is used to regulate the rate of fluid flow
bleeding out through the nozzle to the atmosphere. When the flapper is held tightly
against the nozzle no fluid leaks out and ultimately the output pressure P o reaches the
value of the supply pressure P s, is not it?
37
Let us look at what is that? What I am saying that when these flappers, when this
flapper, you see when this flapper comes very close to the nozzle, when this flapper
very comes close, this pressures and this P s will remain same. P s and P o will be
same at that point, right, if it is coming. But if it goes away it will not be the same.
When the flapper is held tightly against the nozzle no fluid leaks out and ultimately
the output pressure P o reaches the value of the supply pressure P s, right?
Now, when flapper is held too far away, it has no effect on the output flow rate and
the flow rate is governed by the systems size and configuration. Quite obviously at the
same time there is no control and as the system is designed for the measurement of
displacements the supply side restrictor and the nozzle are so designed that the output
pressure never falls to the atmospheric pressure even with the flapper held far away,
right? That is the way we have to do, otherwise there will be a problem.
38
Now, flapper nozzle characteristics: you see, this is our flapper nozzle characteristics.
This is the output pressure divided by the supply pressures, in the x-axis is
displacement. What is the displacement, displacement of the flapper? So, pressure
change versus displacement. This is a normalized so P o by P s. The above show a
small change in the displacement produces large output change. You see that in this
region you see for a small change in output, for small change in displacement I am
getting a large change in the ratio of the output pressures and the supply pressure,
right?
39
Now, supply side restrictor, we have a supply side restrictor also. You see here the
chocked flow it is the actually operating point of our, this is our operating point where
the slope of K sf and this is subsonic, is unchoked flow and we have a chocked flow
which is sonic flow, right? So, here we are plotting P o or output pressure.
In the neighbourhood of the operating point we can do a linear approximation,

otherwise it is non-linear, as we can see. But, at the, at the neighbourhood of the
40
operating point of our flapper-nozzle system we can consider this as a linear sensor.
So, G s equal to G s p o which can be almost approximated G so plus derivative of G
s by dp o at the point p o equal to p oo and p o minus p oo equal to G so plus K sf plus
p o into P of op.
G s o is the value of G s at equilibrium operating point, p o o is the value of p o at

equilibrium operating point and p op is the small change in p o from p oo, right and K
sf value of derivative of G s with respect to p o at operating point.
41
Load side restrictor let us look at. So, this is the load side restrictor graph. So, we
have G u here and we have plotted p o here for different values of x i, right? So, we
can also …… for a small region we can linearize, this operating point we can linearize
across this, right?
42
Here G u equal to G u p o, x i equal to G uo plus K np p op plus K nx x ip, right,
where G us is the value of G u at equilibrium operating point and x ip is the small
departure from the equilibrium operating point x i 0.
Therefore, I can write G u equal to G u naught, G u 0 plus delta G u with respect to

delta G u by delta p o at x i at x equal to i 0 p oo multiplied by p 0 minus p 0 0 plus
delta G u by delta x i x i,0 p 0,0 x i minus x i 0.
43
Now, we know that that mass in equal to mass out. Mass in minus mass out is the
additional mass stored, right? So, G s o plus K sf multiplied by p o p dt minus G uo
plus K np p op plus K nx x ip dt, which I can write dM p equal to V RT o multiplied
by dp o, since p 0 V equal to MRT 0. Let us assume that G so equal to G subscript o,
then I can write V by RT o dp 0 p by dt plus K np minus K sf p 0 p plus K nx x ip
equal to, equal to zero.
Let K equal to minus K nx upon K np minus K sf psi per inch, where tau equal to V
by RT K np minus K sf. Then I can write, tau s plus 1 p 0 p equal to K x i p or p 0 p
upon K x ip equal to capital K by tau s plus 1.
44
Thus we see that the above is a first order system. So, flapper-nozzle system is a first
order system. So, it is also a reverse kind of transducer, right? So, process is
reversible. That means if the flapper comes very close automatically the pressure will
increase or if the pressure increases the flapper will go away it is I mean from the
nozzle, sorry. It is a very fast acting amplifier with good resolution that is most
important thing. It is very fast acting. Even though it is pneumatic system, it is quite
fast. Used in precision gagging that means for a small displacement, I can use this
method.
45
Construction, let us look at. Nozzle diameter is usually 0.8 millimeter, the volume is
16 cc, supply orifice diameter is .4 millimeter and fixed side restrictions, value of cap
K equal to 8000 psi per inch.
See, equivalent circuit if I look at, if I draw the equivalent circuit it is R i parallel with
C with parallel R i in series with C parallel with R o. This is variable, this is output.
So, this is input, this is output.
46
So, R o is the variable restriction, R i is fixed restriction, C is the capacity of the

system or volume and the transfer function of the electrical system is given by V o by
V i (s) R o upon R i plus R o multiplied by 1 upon 1 plus s C R o R i upon R o plus R
i, right?
Some applications let us look at. Force balance differential pressure transducer it is
extensively used. You see, this is the force balance. We have two bellows here, you
47
see here two bellows. Pressure p 1 and pressure p 2 we are giving and this is the pivot
and there is another bellow here and this is my flapper-nozzle system here. You see
that this is flapper. So, across this pivot, across this pivot, so this can move. So, hinge
here, so this can, flapper can move in this and this direction, right? So, bringing this,
this flapper very close to the nozzle or going far off the nozzle, right?
You see, it cannot go very far. What is the, what will happen you see if it go far? So,
you see if it comes very close it cannot close it. If it come, it cannot restrict this one,
because there is a feedback bellow. So, it will push this upward, because if you put,
comes automatically the pressure here will increase. So, it will go to the feedback
bellow, so it will increase, right? Now if it, what will happen you see that if it goes far
away, it cannot go far away also because in that case it will also, this pressure will
increase. So it will, stabilization will come. So, regulated supply is coming p 0 equal
to K p 1 minus p 2. So you see, this is the p 1 and this is p 2 coming through two
differential pressures, right?
The block diagrammatic representations of the above you see like this one p 1 p 2 A
b1 A b2, like this one K b, I 1, I 2.
48
So, A b1, A b2 area of the bellows, K b is the bellows linkage compliance, K n is the
flapper nozzle gain.
In the above circuits, if p 1 greater than p 2, then the upper bellow it exerts more force
than the lower one. Thus the flapper moves away from the nozzle, right, because since
we have a right side we have a bellow. You see, it looks like this, is not it? I have a
bellow here, I have a bellow here. There is a flapper. I have a bellow here, I have
49
here, so this is p 1, this is p 2 and I have a nozzle here, sorry, is not it? So, what will
happen if this, thus the flapper moves away from the nozzle and pressure inside the
tube A falls. Suppose if this is comes then what will happen? Pressure, this, this
bellows will, because this is pivot, so this will, the bellow will, this the flapper will
move in this direction. So, the pressure will fall, inside pressure of the orifice will,
this nozzle will fall, is not it?
The feedback bellow cannot hold the flapper in such a position and brings it down.
The output pressures p naught equal to A b1 l 1 by l 2 by A b2 p 1 minus p 2.
50
Flapper type servo valve - you see, in this type of valve the sliding spool is actuated
by a pressure difference at the two ends. When a signal to the electromagnetic coil
moves the armature, the flapper moves towards one of the nozzle.
The balance of the flow is changed throughout the orifice and it causes the pressure to
increase at one end of the spool and decrease at the other. The spool moves until the
pressure difference is balanced by the tension of the spool, spool, spool springs and
51
internal feedback is provided by means of the mechanical linkage from the spool to
the flapper.
You see, this is our entire flapper nozzle systems, right? This is the flapper type servo
valve you see here. This is our orifice, right? This is our flapper valve. So, this is
electromagnetic pulse, this is armature, right? So, this is our springs on the both sides.
You can see here the springs are on the both side. This side, this side we have also
springs and this is our flapper valve, right?
So, in this lesson we will see that at the first part of the lesson we have discussed the,
basically the piezoelectric sensors which is used for the measurements of the
displacements, accelerations, ultrasonic and which can be used as the ultrasonic
generations of the frequencies or ultrasonic waves and also it can be used in the
crystal oscillator and the second parts we have discussed a very important, a
pneumatic system which is called the flapper nozzle systems with feedback systems,
right and it is very fail safe devices as you can, if you compare with electrical devices
it never fails actually, this, except the routine maintenance of the orifice, because
entire operation depends on the orifice width also. So, diameter of the orifice also is
very important.
52
Over the long use it should not, it should not widen, so that you have to change, you
have to make the change in entire calibration, otherwise it is a just fail set device. It
never, it never happens that it is not working or it fails. So, that is the great advantage
of this type of pneumatic system. For this reason it is used over the years in the
industry and still it is used in some of the industries like hydrocarbon industry, where
high voltage is restricted. That means I cannot use a voltage or any devices which has
a voltage more than 40 volt, because you know there is a, there is a large, you should
have a large actuator or large valve control bulb.
I need a voltage of at least of 220 volt to create a large torque. That is not possible
because that voltage is not allowed in the hydrocarbon industry. Though the
transmitter side of the electrical that is fine, but the basic actuator, basic sensor there
are the basic pneumatic, because it is totally hazardous free, right? Because so many
supply we have instead of electrical supply, so we have a pneumatic supply there,
right? So, with this I come to the end of the lesson 24.

consider ultrasonic sensor, right? Ultrasonic sensors, ultrasonic measurements of flow
velocity or ultrasonic based flow velocity, we have discussed sometime back. But in
this particular lesson, we will basically use the, we will discuss the basic principle of
the sensors and how the transmitters, what is the principle of transmitters, its
equivalent circuit, how the receiver works and the various applications of the
ultrasonic sensors like level measurements, crack detection, as well as the biomedical
applications, where the ultrasonic sensors plays a great role, so that part we will
discuss in this particular lesson.
Let us look at the contents of this lesson.
53
Contents looks like, ultrasonic principle, how the, what is the principle, how it works
that is actually we will discuss here. Then transmitter receiver we will discuss in
details, then equivalent circuit of the ultrasonic sensors we will discuss. Then its
applications, both the level measurements, crack detection as well as biomedical
applications we will all discuss in this particular lesson.
54
Prof. A. Barua
Lecture - 25
Ultrasonic Sensors
Welcome to the lesson 25 of industrial instrumentation. In this lesson, we will consider a

ultrasonic sensors, ultrasonic sensors; ultrasonic measurements of flow velocity or the
ultrasonic based flow velocity we have discussed sometimes back. But in this particular
lessons, we will basically use the, we will discuss the basic principle of the sensors and
the how the transmitters, what is the principle of transmitters? It is equivalent circuits.
How the receiver works and the various applications of the ultrasonic sensors like level
measurements, a crack detection as well as the biomedical applications, where the
ultrasonic sensors plays a great role. So, to that part we will discuss in this particular
lesson. Let us look at the contents of this lesson.
Contents looks like ultrasonic principle how the, what is the principle, how it works?
That is actually we will discussed here, then transmitter receiver we will discuss in
details. Then equivalent circuit of the ultrasonic sensors we will discuss then its
applications both the level measurements crack detection as well as biomedical
applications we will all discuss in this particular lesson.
Ultrasonic sensors, if I look at the ultrasonic measurement system consist of an

ultrasonic transmitter. The transmission medium and ultrasonic receiver whenever there
is ultrasonic sensor there must be a transmission medium and ultrasonic we will see that
later on that ultrasonic is a very poor transmission in air. So, medium there must be some
medium, might be water, might be some steel, might be some alloy and so on and so
forth. Especially the ultrasonic, I mean signals we will see that it is a great transmissions
and the transmission extremely good in the case of. Because ultimately we need to we
need some receiver the where the signal will be received either the same signal can be
used as a same transducer can be used as a receiver. So, in all the cases there should be a
good transmission; that means, transmission power should not be lost. Substantially,, so
that we will not it is not possible to detect either by the separate I mean receiver or by the
same receiver, which is transmitting signal some time back like that. The commonly used
ultrasonic sensors are the piezoelectric sensing element. These are most commonly used.
You see here in the figure we are showing the piezocrystals. We are applying here let us
look at, you see here, if I you see here I am applying the signals. Let me take this one
applying a voltage signal here, I am getting a force here, right. And it is reversible you
can see that if I apply the force I will get a voltage. That is the typical principles of the
ultrasonic sensors or the piezoelectric crystals I should say right. It is a typical I mean we
are using that properties of the piezoelectric crystals to generate the ultrasonic signals,
right.
As we know piezoelectric effect is reversible. The ultrasonic transmitter uses the inverse
piezoelectric effect as I just right now I told. That that is if a voltage is applied to the
transmitter, the crystal will undergo a corresponding deformation, right. If I apply a
voltage to a crystals, so there will be deformation of the crystal. So, how can I utilize?
Let us look at you see I have I have a crystal here.
I am taking a different pen, crystal here. So, I have applied the voltage, right. If I applied
the voltage; obviously, what will happen? I will get a vibrations on this side mechanical
vibrations. So, these vibrations now if I applied AC; obviously, if I applied a high
frequency AC here I will get a vibration on this side. So, these vibrations will be
transmitted on a medium. Now, if I receive if I put when on the receiver on the other
side. So, this will also will vibrate with the same frequency, because there is a movement
of the particle movement of the medium. So, obviously, this will I mean this I should say
the direct, I mean indirect method indirect way that; that means, if I am applying a
voltage and I am getting a force. And now, the force is in imposing on this crystal which
is the receiver and I am getting the output voltage.
This is basic principles of the ultrasonic transmitters and receiver though sometimes we
will see that the same crystals are utilized for both transmission, and for receiving the
signals, right. The vibration of the crystal is transmitted through the media from one end
to the other which just I have shown that the there is a vibrations and the signal
ultrasonic signals will move through that medium. Basic ideas I mean all these I am not
discussing because already we have discussed. We have also we have I mean draw the I
mean discussed the basic flow meters using ultrasonic sensors. Here actually I am going
more on the sensor side; that means, how actually what is the equivalent circuits, what is
the, what type of signals it will generate? What are the basic application?
I mean that though that type of things we will discuss in this particular lesson. The
particle displacement sets up an accompanying pressure, which is picked up by the
receiver, right. Just I shown that is it is a particle vibrations, like it will . So, this I have a
receiver this will also vibrate, because wave will the ultrasonic wave will transmit
through the medium. So, it is received by the that I mean that receiver and it will get the
output signal. It might be the; however, it might happen in the separate also. Suppose I
am giving as a transmitting a signals which is a burst of signals, a short waves I am
transmitting. So, it is getting reflected during transmission it is working as a transmitter.
So, one it is reflected and coming back here. So, I am if this I can use as a receiver, right.
The receiver use the direct piezoelectric effect, because I mean if you I mean if I use the
term that if the with the electric where with the electric voltage I am applying I am
getting the force that is indirect. So, the receiver use the direct. So, the receiver will use
the direct method, right. So, one is indirect another is direct it is nothing it is a reversible
process that I want to mean, direct piezoelectric effect can converts the force into the
corresponding voltage. For transmitter I use some notation x equal to d multiplied by V
and for receiver q equal to d multiplied by F. The performance characteristic d for both
the cases are same, right.

Moreover F we can write K multiplied by x K into x into d into v, where K is the
stiffness constant of the crystal.
This is the equivalent circuit of a transmitter. I see we have a signal generators, on the
other side we have m B 1 by K what are these regions will be shown in the next slide this
is the equivalent circuit of a I mean piezoelectric transmitter, right which is actually we
ultrasonic transmitter, I should say. So, its contrast in context with the, I mean
piezoelectric crystals.

I am discussing all this where you see the ZG the output impedance of the signal. So, let
me go back again ZG, ZG is here on the generator side, output impedance of the signal
growth, mass of the crystals m, damping coefficient B, K the spring constant. ZMIN is
the input impedance of the medium. X dot is a velocity and ideally we can assume
always ZG equal to 0 and ZMIN. That means, output impedance of the signal growth is 0
and the input impedance of the medium is also 0.
So, ideal equivalent circuit of the transmitter looks like this, right. So, taking account the
other with what we have assumed.
Now, this is the equivalent circuit, this is equivalent circuit of the transmitter with m B
and 1 by K reflected in the primary side. If I reflect in the primary side; obviously, I will
get L 1 R 1 C 1.
Where; obviously, when I reflect I mean the primary side, what is the relation between
the I mean what we tried previously in the secondary side and what it has now in the
primary side? That was been related by the, I mean I mean some coefficients which is
decay. L 1 equal to m upon dk square R 1 equal to B upon dk square and C 1 equal to d
square into k, right. Now, overall impedance I should say 1 by Hs or 1 by Hs SC plus 1
upon R 1 SL 1 plus 1 by SC 1.
Therefore, I can say that H j omega if I write in the j omega domain. So, omega R 1 C 1
minus j all inside 1 minus omega square L 1 C 1, upon omega C plus C 1 minus omega
square L 1 C C 1 plus j omega square C C 1 R 1. Thus we have 2 natural frequencies.
series natural frequency which is 1 upon root over L 1 C 1. And the parallel resonant
frequency omega p root over C plus C 1 upon L 1 C C 1 all another square root
obviously.
Now, you see the magnitude plot of this transmitter is like this one, right. It has
significant I mean there is some significance of this plot which will be discussed later on.
And the where at omega equal to omega omega n natural frequency magnitude is
minimum whereas, it is maximum at omega equal to omega p assuming R 1 equal to 0.
You see this is the phase plot of the transmitter. So, you can see within omega n omega p
it is a 90 degree plus 90 degree phase shift, and outside that range it has a minus pi by 2
phase shift what does it mean?
At omega equal to omega n omega equal to omega p the system is resistive this is most
important. And when R 1 not equal to 0, the above diagram shifts towards the right hand
side. So, this you see that is working is resistive sensors. At omega equal to omega p, the
system is resistive and when R 1 equal to 0, the above diagram shifts towards the right
hand side. The circuit behaves as an inductor between omega n and omega p, clear? In
some part it is I mean resistive in some part it is inductive.
Transmissions of ultrasound; how will you transmit this ultrasound signal if P is the
pressure or stress and X dot equal to u is the velocity, because X is the displacement. So,
we have taking the derivative of that. So, the characteristic impedance Z equal to P by u.
the power intensity W equal to P into u.

Average power intensity, is given by W equal to 1 by lambda integral omega z dz
lambda over the integration of the limit lambda to 0. It looks like this you see the
transmissions of ultrasound signal between 2 medium. You see the medium 1 and this is
medium 2. This is crystals this is the reflected wave. We will see the regions what are the
regions.
Where Z 1 is the characteristic impedance of the medium 1.What is Z 1? This is Z 1. Z 2

is the characteristic impedance of medium 2, right. And WI is the incident power
intensity and WR the reflected power intensity and WT is the transmitted power
intensity.
WI is the lesser, lesser than the power intensity generated by the crystal due to losses in
the medium 1, right. Obviously, there will be some loss in the medium 1 and alpha R is
the reflection coefficients which is given by WR by WI equal to Z 2 minus Zone the
whole square upon Z 2 plus Z 1 square. Alpha T is the transmission coefficient WT by
WI 4 Z 1 into Z 2 Z 2 upon Z 2 plus Z 1 the whole square and; obviously, alpha R plus
alpha T reflected and transmission. I mean if you take this coefficient; obviously, this
two will be one if there is no absorptions, right. If Z 2 in minus Z 1 is large than the more
of the incident power is intensity is reflected back, right.

So, the characteristics impedance of the few this you see the very important while you
will find you will see that while using this ultrasonic sensor for the level measurements,
right. If Z 2 minus Z 1 is large then more of the incident power it will be reflected back
from the mediums 2, right. Using this principle, we will make the level sensors, right. So,
this is the typical property even though sometimes it is feels like that this is a bad quality.
But this property will be utilized to make a very good sensors of the level measurements
as well as crack detections in a metal, clear. Now, characteristic impedance of the few
materials we will now drawn down some characteristic impedance of the few materials
Quartz 1.5 into 10 to the power 7 and barium titanate 2.510 to the power 7. Then
polymer PVDF we are using all sin notations please note 0.4 into 10 to the power 7, steel
4.7 into 10 to the power 7, the aluminum 1.7 into 10 to the power 7, bone 0.8 into 10 to
the power 7, water 0.5 into 10 to the power 7, air 430 only, right. So, this property of the
air will be utilized to meet the good level measurement of any liquid, most of the liquids.
So, we will define I mean the liquid is good, but in the most of the cases I mean offering
I mean transmit I mean characteristic impedance alpha R and alpha T for different
interfaces. There are 2 interfaces of the 2 material media 1 and medium 2, alpha R and
alpha T for the different mediums will be this will give you how you can interface 2
actually 2 different medium while you are transmitting the signal from medium 1 to
medium 2, right. So, that is the reason the alpha R and alpha T is necessary. So, this is
the 2 different we will take the combinations of the 2 materials, right. See, quartz steel;,
so it is 0.27 0.3 quartz water 0.67 0.733; that means, if you want to transmit, right. So,
this is reflected is the more than the, I mean transmittance and here you see the quartz the
reflectance is less than the transmittance. And quartz air; obviously, extremely poor see 1
and 1.110 to the power minus 7 into 10 to the power minus 4.
Thus we can say that the air is a poor choice for transmission of ultrasound waves, right.
As a difference of the characteristic impedance of the air with others is very large, right
PVDF is the polymer actually is used for biomedical applications.

So, measurement of ultrasound. Pulse echo technique, is a piezoelectric crystal acting as
a transmitter receiver is attached to the medium 1, we will show the diagram. The
characteristic characteristics impedance of medium 1 and medium 2 must be
substantially different, otherwise it cannot be reflected. First the crystal act as a
transmitter and it send out a pulse. It is a burst of signal I mean burst of pulse I mean it is
very short duration pulse. And it will remain silent for some time, until unless I will see
all the signals all the reflected signals, right. You will find you will get a repeated
reflected signals. We will take the first one either I mean received signals which has
highest intensity other subsequent signals reflected signals we have low intensity, right
generated by the pulse generator width Tw.
You see this is here we are measuring scheme of ultrasounds. We are showing one time
base here it is same as the oscilloscope, oscilloscope as you know we have a time base
there also. So, all together for there also we have I have 2 medium medium 1 and
medium 2. We have just discussed these I mean alpha R and alpha T for the 2 different
mediums. The 2 medium you see 1 reflected from this side, another 1 reflected for the
medium. So; obviously, this will take more than time and; obviously, intensity. So, here
you will find the intensity is much more. And if the first signal you see the signal will
come. So, it will first go sorry. So, it will signal will go and reflected back, after I mean
this signal again will come and reflected back. Obviously the first the shortest time will
be this signal, right. Because ultrasounds will transmit and will silent for some time,
right, until unless we will make all the measurements and next time again, we will
transmit the signals.
Most of the pulse energy what is this I mean the surprise what is echo amplifier echo
amplifier is nothing, it is an amplifier. It is I mean operated only for the time when the
first reflected signal will come. Because the signal you may need the amplification that is
the reason I need an amplifier here. So, that is the echo signal; that means, the reflected
back signal will be amplified and pushed here, hence we have pulse generators which
will give you in the time base. It is in the vertical signals in the echo amplifier the
signals. What is the signals which is coming at the vertical plate? That is the signal that
is reflected ultrasonic waves are coming in the vertical path. Whereas, the time base we
are giving in the horizontal path. Principle is same as they have seen in the oscilloscope,
right. Most of the pulse energy is reflected at the boundary of medium 1 and 2, clear?
What is that most of the pulse energy will be reflected here?
So, signal will come it will be reflected back and will come down here, right. The crystal
now acts as a receiver and receive a pulse, right. Initially transmitted and remain silent. It
will not transmit any signals, until unless the signal is coming back from the receiver
from the reflected surface. That means, the junctions of the 2 surfaces 2 different
medium will work as a reflector of the ultrasonic waves. The time taken by the reflected
pulse is TT equal to 2 I upon C, right. What is y 2? Because it will go and it will
reflected back. What is 2 l? If you say it is l small letter or 2 I whatever, it. So, it is the
length of the medium 1, where I is the distance of the interface of the 2 medium from the
crystals, right. What is that? Let us look at I have reflected like this 1. So, I have a 2
medium is here right. So, I am transmitting the signals. So, it is signal is coming
reflected back this is our l I, right. And C is the velocity of sound in medium 1, clear?
Now, you see the reflected means of the ultrasonic signals. There is a very nicely we
have shown you see the first outgoing pulse will look like this 1. Obviously, because a
part of the signal will be transmitted through the medium through the junction of the
medium will go to the medium 2. So, the this is the first reflected wave it will come, then
second reflected wave, then third reflected all of different magnitude. So, we will first
the we will take this 1, and we will not send any signals until unless all the signals are
died out. All the reflected signals are died out; we can calculate that part and accordingly
my burst of signals which will go to the transmitter. Because there must be some
electronic circuitry which will transmit the signal which will go the, because ultimately I
have to impose some vibrations I can use some oscillators also there, right. Electronic
oscillators which will start to make the vibrations of the ultrasonic crystals, right. So, I
can obviously, I can program in such a way that the time taken time in between
transmission of 2 signals will be such that all the reflected waves will be die out by
during within that time.
The repetition rate TR should be such that all the reflected pulses of the interest have
been observed before sending the second pulse, right. The transmit time TT should be
large compared to the pulse width Tw. To avoid interference between the outgoing pulse
and the incoming or reflected wave, right. So, we have a width. So, we have decided I
mean we are making transmitting time TT should be large compared to the pulse width
to avoid the interference. So, we are making this large width. So; obviously, using that
times we will transmit, right. What does it mean? It means you see I have a its looks like
this that means I have a width like this one, right. Because this I can program. So, during
that time this time all my signal reflected waves will be finished until unless. So, there is
a sufficient amount of elapse time by next pulse will be transmitted again. So, this is the
transmission width, right. So, this is the pulse width, Tw to avoid interference between
the outgoing pulse and the incoming or reflected pulse this is all about, right.
An application of this type of I mean ultrasonic signals let us look at. The method
discussed can be used to the to make the level measurements. Especially the level
measurements of a liquid I mean if I need a with precisions and fast measurements.
Because you know the liquid there are various types of measurement of the level
measurements. We have a steel tape and the and the ruler, that type of things are there
scale and I mean steel tape which is used mostly in the our country. You will find many
where in the municipality I mean tank, so huge tank. So, that it can be utilized for
measurements; that means, it is other way. You see the scale will be when it is empty. I
mean the tank is empty, the all the indicators will go up and when the tank is full the
indicator will come down.
So, mechanical indicator which will which is connected by a counter width. So, that I
mean it will whenever this level is up level the level is filled up. So, it will switch off
there are many other there are a capacitive method of level measurements we have seen
there are I mean method of so many other methods of I mean level measurements. We
have pneumatic level measurements; that means,; that means, I mean it looks like this.
You see I have a I mean a vessels. So, I will put a a pressure on this 1. So, there is a
float, so this will. So, if it goes up, so it will put a pressure on the values right. So, this
can be transmitted to indicate on a both the gauge or a I mean to show the pressure
because. So, this pressure gauge will be calibrated in terms of in terms of I mean in terms
of levels.
You cannot use both the cases; that means, you cannot use both the I mean both the
gauges and the I mean and the C type both the. We have to use either either you use
bellows because bellow will try to expand so that if you connect an indicator, it will
show you the level or if you connected to some, the gauges. So, there is a deflections of
the free end of the I mean tube. So, we will get the measurements and that pressure can
be calibrated in terms of level. So, there is lot of lever you see the most of the cases this
is most important that the ultrasonic method measurement you will find is very non
invasive type of technique, right. You do not have to your sensor should not be in contact
with the not necessary should be in contact with the liquids or any others I mean
materials.
See this is the typical example, I mean of a level measurements. It looks like this. You
see this is a liquid, this is our vessel in which I am interested to make the measurement,
right. So, there is a ultrasonic signal. So, I am transmitting these signals and reflected
back. As I told you the pulse width will be sufficient in large. So, that I mean I mean all
the reflected waves will be collected here before I send the next pulse, right. So, this time
will be measured. So, time will be what, so as the level goes high. So, as the level goes
high it will ultrasonic waves will take more and more time to travel the distance, right.
And will be deflected I will calculated the T as I told you. T equal to 2 I by C. So, if this
is I right, so what is C? C is the velocity of ultrasonic waves in this particular liquid,
right. So, I will get direct measurements, because how if I measure T? So; obviously, I
will be what? I will be level.
So, I will be T multiplied by C divided by 2 is not it? But there is a precision of

measurement because you see this I mean signal which go so fast the entire accuracy will
lie on how accurately you can measure the T, that is most important, right. Even though
it looks very simple I mean when you show like this one, but actual implementation is
quiet taught in that sense, clear? Now, you see why I have installed the crystal at this
side? I could have install that on this side also, I could have install the crystal on this side
is not it? Either the ultrasonic wave also will come here and it is reflected back, but this
is the liquid this is my liquid, right. So, let me take the new page. So, that will be better.
You see this is my liquid, and this is air. This is the top surface of the vessel, top surface
and this is the top surface of the liquid. If I install the ultrasonic sensor here then what
will happen? Either way it is same, because as the liquid goes up and up what will
happen that again that this it will reflected back from this side. Because you see due to
the difference of the characteristics impedance of the ultrasonic wave will be reflected.
Here also the some will be reflected, but as I told you that it is very difficult to launch
these ultrasonic waves in the air. So, that is the reasons we have put the crystal down. So,
that it can be easy to transmit the signals through the liquid other than transmitting
through the air, clear?
It is to be noted that the crystal must be placed at the bottom not at the top, just we have
what we have discussed right now, right. It placed at the top due to presence of air no
wave will be able to propagate thus giving us erroneous measurements, right. That is the
reason we have installed at the bottom of our vessel. Now, ultrasonic waves as I told you
earlier at the beginning of the lesson, it is also used for the crack detection. You see here
what will happen the ultrasonic waves will come if it there is no crack. So, what will
happen? The signal will transmitted like this one, if there is no crack. If there is a crack
then what will happen. So, again there is a difference of the characteristic impedance
here. So, what will happen? So, it will it will come it will reflected back through this one
if I know that sorry if I know the time, sorry if I take a.
So, what will happen? it will go it cannot move through air. So, it will be reflected back.
So, if I measure the time I can exactly locate the where the crack is even though if it is
from outside it might be invisible if this the crack is inside, right. This is a very excellent
technique I mean we have a different techniques of a crack detections. So, we have the
detections of the we have the method of X ray and all those thing. But it is a very you
know we have to I mean take an X ray and it is an high. We have to shield the people
who are working there then I mean you need a place where to develop all those thing.
But that is not necessary here in the case of ultrasonic signals if we use when it for crack
detection.
Here the crack or gap acts as a second medium, and thus help us to detect where the
crack has taken place. Ultrasonic can also be used in the measurement of flow. You have
already you have seen that the transit time ultrasonic flow meter Doppler shift ultrasonic
flow meter those we have already discussed. Though we have not discussed the details of
the transmitter and receiver that we actually we are discussing in the ultrasonic sensor
itself. That is actually is the part of flow measurements and there is a little scope of
discussing of all these sensor as such as a whole as a transducers flow, transducers. We
have discussed the ultrasonic measurements of flow.

Advantages; what are the different advantages? It is easy to direct and focus a beam of
ultrasound as diffraction of these waves are small, due to short due to their short
wavelength diffraction is small. Ultrasonic waves can easily pass through the metals that
is a great advantage. This help in mounting the measurement system outside the system
and it will not lead to the development. It will lead to the development of non invasive
techniques of measurement or non invasive sensors. This is a terminal logic we are
calling non invasive measurement. So, this is a non invasive measurement techniques,
we are utilizing for ultrasonic waves. I mean making the ultrasonic measurement.
Because I again install it outside I do not have to install anything inside you have seen in
all the pressure all the level measurements except capacitance obviously. But capacitance
is not a problem I am in shield shielding parasitic capacitance all these comes great
problems in a measurements. We need very accurate LCR meters or the accurate bridge
and all these this is not necessary in the case of ultrasonic measurement.
Special application; an ultrasonic application there are special applications of this

ultrasonic waves. Ultrasonic sensors have a wide applications in biomedical
instrumentation, right. Measurements of blood flow; there are I will discuss 3 different
major applications of ultrasonic waves in biomedical instrumentation. Measurements of
blood flow in the artery, determination of the blood pressure and ultrasonography. Even
though ultrasonography I have not discuss the typical ultrasonography that mean
abdomen ultrasonography. All those things, we will discuss some part of the
ultrasonography because it is a huge subjects and you can take reference choose any
books I mean text books.
Determination of blood pressures; how the blood pressures will be used? You see this is
the typical I mean typical diagram. So, ultrasonic determination of the blood pressure, so
right. So, here you will find that I mean ultrasound sensor. And we have a ultrasound this
is the blood flow and we have a ultrasound 8 megahertz detector also. This 8 megahertz
signals and we have a 8 megahertz detector also, so amplifier and detectors. This is audio
amplifier audio outputs it is going to a head phone. This is actually please note this is a
head phone, right. This is a head phone, so we can I mean by listening to the sound I can
tell what is the pressures.
This is how the and the blood is flowing through this artery. You see what will happen
that 8 megahertz incident ultrasound will come and in the blood. So, they will be
reflected. Now, what will happen? You see here I have a skin here, because we have to
put outside the skin. So, it is a non invasive technique very much. So, we will look at the
sound, we will listen to the sound during the opening and the closing of the valves. I will
get the signals of the, during opening I will get a largest amplitude signals and during
closing I will get a short I mean amplitude signals. So, by this I can detect the
ultrasounds and measure the pressures. So, these frequencies this delta f is now convert it
into the audio output. You see delta f will be different, right.
Now, measurement scheme employs a transcutaneous Doppler sensor which detects the
motion of the blood vessel walls at various states of occlusion, right. Two small
ultrasonic crystals 8 megahertz used as a transmitter as well as receiver. The transmitted
signal from the Doppler ultrasonic is focused on the vessel wall and the blood.
The reflected signal is detected and decoded, right. Decoding means; that means, I am
decoding converting into the amplitudes and all those frequency depend I mean
frequency voltage convert or that type of signal. Difference between the transmitted and
received signal is in the frequency range of 40 to 500 hertz and is proportional to the
velocity of the wall motion and the blood velocity, right.
When cuff pressure is above the diastolic, but below systolic. The vessel opens and
closes each heartbeat as the pressure in the artery also oscillates below and above the
external cuff pressure. This opening and the closing are determined by the ultrasonic
systems. That I have shown; that means, the 2 different ways during openings I will get a
signal like this 1 sorry. So, I will get a signal which looks like during opening and during
closing I will get a again during opening, so again close, right. So, this type of signal I
will get.
As pressure is increased further the time between the opening and closing decreases, and
the reading indicates the systolic pressures, right.
Conversely when we decrease the pressure in the cuff, the diastolic pressure is recorded,
clear? The advantage what is the advantage of this system? There was some
disadvantages also. Let us first look at the advantage it can be used with the infants and
hypotensive individual; that means, pressures I mean persons with a high blood pressures
as you know the medical for a healthy person the blood pressure should be within 80 to
120, right. So, but some people have very high blood pressures in for that type of patients
we can use or also it can be used for infants can be used in the high noise environment.
Because you see all this I mean this other type of measurement techniques what they are
used actually depends that the physician must locate the time when the valves are
opening And must locate the pressures sorry the manometric pressures when the valves
are opening, right. So, it is sometimes very difficult if it is a noisy environment. So, it is
quite mute noise outside noise environmental noise which is very common in hospitals
and all those things. So, it can be nicely utilized in that type of environment or that type
of in that type of I mean places.
But what is the disadvantages? The movements of the subjects body cause the changes in
the ultrasonic path, between the sensor and the blood vessels; obviously, everything will
change. So, the patient should I mean some steady condition. So, you should not nice, so
it is very difficult to; obviously, though we are saying that is very we are saying the
infant we can use. But in some cases we can utilize this type of the measurement.
Now, blood flow meters, you know this flow meters it is basic principle is same transit
time flow meters and the flow meters like Doppler shift flow meters. But the same
principle also used for measurements of blood flow on the measurements of blood in the
artery. Blood flow measurements using the ultrasonic can be done by the following.
Transit time on the measurements flow meter, doppler ultrasonic flow meters. Doppler
shift this we have discussed very details in our, I mean flow meter I mean lessons. So, I
will not go too much details of this one. In context with the, I mean biomedical
instrumentation I will discuss something nothing more than that.

You see this is ultrasonic transducers configurations we are using different
configurations. So, the ultrasonic uses I mean transducers A B C D, what is this?
Let us look at Case A; the transit time probe requires two transducer facing each other.
And the path length between them is L and are inclined to the vessel axis at an angle
theta. We know if we recall go back to the ultrasonic I mean you see the transit time
principle is something like this I have a vessel like this one. So, I have a signals always
will be inclined I have a sensor here. So, it is a this make an angle of theta right. That is
actually I am telling that means what I am telling here.
For case A, the transit time probe requires two transducers facing each other. Obviously
you see that you remember in the, I have 2 sensors liquid is flowing through this and one
sensor here and another sensor here right. So, facing each other clear?
And the transit time probe requires two transducers facing each other and the path length
between them is L and that inclined to the vessel axis at an angle theta. The shaded
region is representing the an acoustic pulse right. What is that? Let us go back shaded
region is acoustic pulse. For transcutaneous probe as in B both the transducers are placed
on the same side. They can be placed on the skin and the beam intersection region is
shaded. Let us go back this is a beam intersection region. So, two are placed. So, for
transcutaneous probe as in the B both the transducers are placed on the same side and
they can be placed on the skin. So, that we are utilizing the reflection from the blood I
mean from the end of the artery like that the intersection region is the shaded right.
For case C, we can use a lens which helps to focus and narrow the beam. Sometimes we
need to narrow the beam since it can be utilized to do there. Case D is used for pulse
operations In this case the transducer is loaded by the backing it with a mixture of
tungsten powder in epoxy and the shaded region is shown for a single time of range
gating.
Now, this is a Doppler ultrasonic flow meters we see here we have an oscillators I mean
I always encourage that the those who are I mean listening to this particular lesson they
should go back to my the lesson on the flow measurements using ultrasonic principles.
So, all the things will discussed in details there right. So, the this is the Doppler
ultrasonic blood flow measurements the principle is same. There are 2 types of frequency
will have we will get a 0 crossing detectors. We have an low pass filters and output
detectors because the audio frequency will be related on the I mean difference of
frequencies will be derived which is very small. So, that will be passed or all other
frequencies will be stopped.
So, this is actually the entire scheme of our ultrasonic flow meters using ultrasonic
measurements of blood flow using the Doppler shift method right. We can see here. So,
we have a detector, so then after that frequency will be lowered. So, we have a audio
frequency amplifier. Detector can be simply a so; obviously, what will happen that the
output signals what we will get? The low frequency and obviously, I can use a audio
frequency amplifier there. You see this is a audio frequency amplifier we have a audio
frequency output of the speaker output also. And we have a 0 crossing detector then low
pass filter then we can get the output recorder because this I will get a signal. This I am
getting audio signals. This I am getting electrical signal again here which can be
recorded, right.
For Doppler flow meter we know that fd by fo equal to u by c where fd is the Doppler
frequency shift. And fo is the source frequency which is basically ultrasonic frequency u
is the target velocity and c is the velocity of sound sorry. It is source frequency not
velocity. This f naught is the source frequency or the frequency of the ultrasonic sensor.
For blood flow meters the blood cells act as a particulate matters which form the
reflecting targets right. Because as you know in the case of Doppler shift flow meters I
mean in the case of transit time flow meters this is basically used in the case of clean
liquid. Whereas, the Doppler shift flow meter will not work with that we have discussed
in details that it will not work until unless there is some suspended particle there in the
liquid itself. The liquid is flowing through the pipe there is some suspended particle
whatever small it may be. But that particle will missing here in the case on the blood
flow measurement. But the blood cells will act as a particulate matters which form the
reflecting targets right. From that actually the ultrasonic will be reflected back it is
excellent it is not it? There is sometimes with that time we have discussed that the 2
different flow meters because in we do not know what type of liquids flow we are
measuring.
Sometimes the liquids are might be very clean suppose in the case of water or some other
liquid, but in sometimes it it may have studies. It may have suspended particles, but if the
liquid is extremely clean I cannot use the Doppler shift flow meters, but that I mean
principles is purposely we are utilizing in the case of blood flow measurement right
where the cells will act as a reflector of the signal ultrasonic signal which is transmitted
from the transmitter. In the arrangement shown in the figure frequency shift occurs
twice, once between the transmitting source and the moving cell correct. Second another
between the moving cell and the receiver, right. One is the moving between the
transmitter source and the moving cell. Moving cell means moving blood cell right. It is
moving cell means moving blood cell because; obviously, when the there is a there is a
continuous flow of blood through our artery, right. So, there is the cells are also be
continuously moving through the artery. So, another between the moving cells and the
receiver.
So, of the fd by f naught equal to two u by c as c is much much greater than u. Velocity
of the ultrasonic must be much much greater than the velocity of the blood flowing in the
artery. So, what I can say that taking angle factor theta the angle between the beam of
sound and the axis of blood vessel fd equal to twice f naught u cos theta by c equation
number 1.
And as in the figure the ultrasonic waves are transmitted to the cell which reflects the
Doppler shifted wave to the receiver, right. As in the figure the ultrasonic waves are
transmitted to the cell which reflects the Doppler shifted wave to the receiver right. So;
that means, I have a ultrasonic waves it is getting reflected and coming to the receiver.
How does it look? It looks like this one ultrasonic waves are there. So, I have a source
here sorry.
So, I have a source and I have a receiver here ultrasonic waves I am transmitting. So, it is
getting here reflected back from the blood vessel. This is our blood vessels, blood cells
rather I should say blood cell right. Reflects the Doppler shifted wave to the receiver. So,
this getting reflected I am. This is getting reflected.
The amplified radio frequency signal plus the carrier signal is detected by the produce an
audio frequency signal. It is the basic principles of a audio I mean our heterodyne
receivers. This classical method actually utilized here also AF signals given by equation
1.
Now, ultrasonography you see this is a this is a I should say the last I mean example of a
I mean this is a very I mean very wide subjects. I mean its studies very difficult to
accommodate in this particular end of the I should say of the of this particular lesson. But
it is extensively used for measurements of many to know especially when the child is in
the home of a women. So, that that time also it is utilized to know the health of that type
of the baby in the home right. Anyway there are other applications of the
ultrasonography we will discuss that part. You see the brain medicine A mode scan of
the brain midline I have a receiver transmitters. So, transmitting signals it is transmitted
and we have oscilloscopes and sweep as usual I am giving into the vertical this I am
giving in the sweep I am giving to the horizontal. I have to know I have to make it
stationary, because the signal is to be stationary. You know that is the basic principle of
the when we give to the horizontal waves or horizontal plates right. So, that is actually
utilized here. A mode scan of the brain midline.
The appropriate electronic circuit can be used to pulse the ceramic in order to transmit a
short burst of ultrasonic energy. Always please remember in all these method we are
always have transmitting a short burst of the energy. How does it look? We have seen
that already we have shown. That is your transmitting signal looks like this. Then it is
getting reflected it is lower and lower. Again I will transmit this is one width again I will
transmit. So, this is a short burst of signal. So, it is not it a continuous signals you cannot
make a if you transmit a continuous signal no where you can make the measurement
clear?
It first act as a miniature loudspeaker and then switched to act as a microphone to receive
the signals reflected from interface of various tissue types right. I can use this speakers
also and transmission and for receiver I we can use the microphone. Ultrasonic energy at
the level levels of medical imaging causes no harms to tissue that is very important,
because any other methods we will give you harm the tissues. So, ultrasonic energy will
not harm the tissues. Because whenever you are using the medical instrumentation you
have to look at the safety and it should not harm the human being. It will not damage
anything. So, those part you have to this is an A mode device which shows echo intensity
as an x y plot we have shown Y we are giving the amplitude of that reflected waves and
x plates we are giving the sweep. So, I will get a stationary waves.
This is the some information I mean B mode ultrasounds imaging intensity informations.
It looks like this.
This is known as B mode ultrasonic imaging techniques which shows 2 dimensional

shape and reflectivity of the objects by using multiple scan paths. This type of simple
device is seldom used now; however, has been replaced and more elegant systems. So,
with this I come to the end of lesson 25 which is on the ultrasonic sensors. Welcome to
the lesson 26 of industrial instrumentation. In this lesson basically we will cover a
nucleonic instrumentation.
Let us look at Nucleonic instrumentation means that we will use some radioactive
sources. And there will be corresponding detectors and there are some advantages of this
of this type of instrumentation. Because we have seen that the in the in the lesson 25, we
have covered the ultrasonic instrumentation. So, there are some typical advantage of
ultrasonic, but there are some cases like. If you want to launch that ultrasonic waves in
the air it is just impossible where the radioactive waves can I mean moves through air.
So, that is the great advantage of this particular instrumentation. So, it is to be covered
very extensively. Let us look at the contents of this particular lesson.
Contents; principles of nucleonic instrumentation, radioactive sources, radioactive source

elements. What are the different elements which will code or I mean it will make the
radioactive sources? Basically radioactive sources means alpha beta gamma rays, but
what are the elements which we will actually used for used? We actually used in the
nucleonic instrumentation that we must study. Its application in industrial
instrumentations; that means, we will find the label gauge the there is typical
applications. All the measurements of level then typical applications of the
measurements of the coating of a sheets and the rolling mills. So, there are various
applications we will finding this particular type of instrumentations.
So, we have same or if the densities I mean the density might be different, but the atomic
weight it can be the same. For level measurement we use the following scheme.
This is of the scheme of the level measurements. We have a source you see here. We
have a detector here and. So, this is the level measurement using nuclear instrumentation.
So, I have a liquid here. So, how it works?
Let us look at… When the passing when passing through the fluid the absorption will be
more and hence the amplifier output forces the steeper motor to move the source and
detector to move upward right. You see what will happen? So, when passing through the
liquid this will. So, it will always try to move this steeper motor to move it upward right.
When they reach the interface the absorption falls and thus we know that the level is
reached clear? Let me do it again. So, the this principle is something like this if this 2
system this I mean. Let me take a; that means, this stepper at this source and detector will
be initially at the bottom here. And please note that in the nucleonic instrumentation. So,
the either it is a great problem.
So, we have to think of all this whenever you are using what type of I mean radioactive
materials using though the half life. And all those thing is not that large, but the amount
of ways to you thing off. So, the proper sometimes there is a are there. So, until unless
you, we can manage with some other instrumentation, which is non nonhazardous like
the ultrasonics and any conventional, instrumentation of level gauge or density
measurement that is to be used. But in some situations as I told you like the detections of
the leak in the pipe underground or suppose the in a close containers I want know the
level of the liquids.
And all these things very difficult to know it is specially it is sealed. If the container
might be closed it does not matter because if I have a inside I mean electronics already
installed inside. So, I can inside I can detect, but if it is a separate container suppose ah a
cylindrical drum I want to know that how much the liquid contains which we cannot see
from outside. So, that type of situations I can use this type of radiation technique or I
mean technique. So, this is to be basically I mean all of us you have to careful; that
means, that I have told you several times. That means f I can manage with other
instrumentation system conventional nonhazardous systems. We are not suppose to use
the radioactive sources right. So, with this I come to the end of this lesson 26.
Prof. A. Barua
Indian Institute of Technology Kharagpur
Lecture - 26
Nucleonic Instrumentation
Welcome to the lesson 26 of industrial instrumentations. In this lesson basically we will

covered a nucleonic instrumentation. Let us look at Nucleonic instrumentation means
that we will use some radioactive sources and there will be corresponding detectors. And
there are some advantages of this of this type of instrumentation, because we have seen
that the in the in the lesson 25 we have covered a ultrasonic instrumentations. So, there
are some typical advantages of the ultra sonic, but there are some cases like if you want
to launch those ultra sonic waves the air it is just impossible where the radioactive rays
can I mean move through earth that is the great advantage of this particular
instrumentation. So, it is to be covered very extensively. Let us look at the contents of
this particular lesson.
Contents: Principles of nucleonic instrumentation, radioactive sources, radioactive

source elements, what are the different elements which will I mean which will make the
radioactive sources? Basically radioactive sources means alpha beta gamma rays but
what are the elements which will actually used for used actually used in nucleonic
instrumentation? That we must study. Its application industrial instrumentations that you
find the label gage there is typical applications of the measurements of the level then
typical applications of measurements of the coating of shades and the rolling mills.
So, that various applications will find in this particular types of instrumentations. So, that
you will cover in details all the different applications we will see then detectors what are
the different detectors used in this type of instrumentation nucleonic instrumentation?
Then also even though its little diverts us will cover the nuclear medicine, because the
nuclear medicine also uses some detectors, some sources and some detectors which is
very similar. So, there is some instrumentation there also, so we will cover in these under
the head of nucleonic instrumentation will also cover the nuclear medicine.
Now, nucleonic instrumentation looks like this. Used for industrial measurement it is
based on radioactive material. So, can it can travel in air thus an improvement over
ultrasonic’s which cannot travel in air we have seen that if you launch the try to launch
the air ultrasonic’s waves in air it is almost impossible. And we have seen that in the
cases like if I want to travel I mean using this particular pitch we made the level gages
also. Because when the when the when the ultrasonic waves are traveling through waters
and it is when its reach the top surface it is almost I mean negligible portions will be
transmit in the air most of the portions will be reflected back from the top surface.
So, using that principle we made the level gages we have seen that thing. But this
disadvantages also, because in many situations we may need to measure the, I mean we
want to transmit the signals, because all these signals all these particular method. That
means, ultrasonic methods or nucleonic instrumentations or non-invasive type of
techniques. So, in some cases if I want to I mean launch this rays through the air it is
impossible in the case of ultrasonic, but it is possible through the nucleonic
instrumentations or radioactive sources.
Advantages is non-invasive we have seen that this non-invasive are just I right now, I
told it is non-invasive techniques. Particularly useful when other techniques fail in many
situation will find that the other techniques will fail, because you will see the radioactive
sources have some hazards unlike other sources. If I use any other conventional sources I
mean if you use then radioactive source also always saturated with some hazards until
unless it is justified that we have to use nuclear instrumentation. Nobody will allow you
to use electronic I mean sources and detectors. Disadvantages; first of all with its
associated problems of hazards as I told you.

In application; let us look at density measurements, thickness and level measurements.
This is the basic 3 measurements we will do; that means, density of a liquid you can
measure, thickness of the surface we can measure, thickness of a coating suppose walks
on a paper. That will also we can measure suppose very common examples we are telling
that we had a stitch at record at almost all of you know. A stitch at record as used
extensively in instrumentation, so it is even though we can make the data equations I can
we can stored in a computers, but always people wants some hot copy also. So, if you
want to store that type of signals suppose, I am I mean recording a signals of
temperatures over 24 hours time with sampling of 10 minutes at every 10 minutes I want
to record that signal that temperature on a strip chart recorder. One of the method is that
you use that strip chart recorders, the thermal style as where thermal style as we move
and the recording it looks like this.
That means you see that, if I have a let we take a blank page. So, I have a strip it is
moving right, so the style is coming down here. So, I am getting impression over this 1,
so this in time and this is in temperature take this centigrade. So, I will get a temperature
at any time what is the temperature of the particular process that can be viewed from this
particular impression. So, they are at ink style as ink as a problem, because we have to
refill every time. So, all this cassettes are also is there, but instead you if you can do that
this entire strip entire piece of paper that is which is rolled here roll unrolled here. It is
coming down in this direction. So, what will happen? You see that A impressions will,
because if this is blackened this paper or some other color and coated with wax. These
papers if it is coated with wax if this paper is coated with wax. So, what will happen?
You know that it will once if it is heated this suppose this style is heated then this style is
heated. Then what will happen? You see this will give an impression, because the wax
will melt up and it will give the impression from which I can record the temperatures.
This is the one of the method I mean one of the applications of the instrument of the strip
shot recorder in instrumentations. And, so they are the precisely I have I have to control
the thickness of the wax coating over the paper. Uniformly you have to go it should not
very thick, because it should be very thick then this hard style as cannot give an
impression if it is very thin that might be useless to us. So, those precise I mean precision
I mean coating we have to use. We can use a nucleonic instrumentations are method of
measurement of the thickness.
Humidity of grain is another method. I mean another examples of the, because in the
grain if this always to store it. I need some particular humidity that is to be maintained
for the good preservations of the grain. So, that measurement of humidity in the grain
also can possible through nucleonic instrumentation. Determinations of material
composition true detection. So, what is material composition? That can also be detected
is possible through the nucleonic instrumentation, leak detection. That means, if there is
a leak in the pipe and all those things; that can also be detected especial in underground
pipe.
It is very difficult to know where the leak is it is possible to record through it is possible
to find through nucleonic instrumentation. Also determinations of age of archaeological
samples, because you see many Indians I mean our is a great country. Find that many
places are old sculptures old I mean things are coming up I mean there are many
sculptures are coming up right with my 2000 years old might be more than that. So, to
know this age of this particular sample, we need some I mean some sort of radioactive
sources and detected techniques by which I can tell you the age. So, there is another
application though it is not much of industrial instrumentation, but this is another
application are nucleonic instrumentation.
Now, radiative source or radioactive source, so this should be actually I should say
radioactive source instead of radiative source. These are alpha particles beta particles
gamma rays and neutrons there are 4 basic. And I mean there are some advantage and
disadvantage of this particular method. So, we have alpha particles beta particles gamma
rays and neutrons. Now, alpha particles it is high ionization. It will ionize a gas through
which it will pass. Low penetrations only can it can pass 10 centimeter in air energy is
high. That means, 10 and in electron holes which is quiet high suppose to be quiet high.
Beta particles; it is less ionizations property than the alpha particles. It is penetration is
much higher 10 meter in air and its energy is medium energy. So, .0 .5, so you can see
that its energy is less.
Then gamma ray is its penetration is several inches of lead, because lead is very difficult
to penetrate through any other alpha beta particles will not penetrate through lead. So,
that only the gamma rays can penetrate several inches, but if we use thicker lead gamma
particles also cannot penetrate through it. Neutrons neutronic I mean neutrons has
different source it has a no ionization effect. It has large penetration and very small very
small energy 0.0 0.025 electron volt; you can see that how less it is.
Now, half life it is a very I mean common form of I mean it is called parameters.
Whenever you use you were using in a radioactive sources you must be aware of this
half life period. Sometimes we need shorter half life period sometimes we need larger
half life period. It is a time required by a radioactive material to decay to half of its initial
size. It is time required by radioactive material to decay to half of its initial size.
Mathematically it is I mean defined as half life equal to t 1 by 2 equal to 0.693 by
lambda while lambda is decay constant.

Now, source materials what are the different sources of this alpha beta gamma rays and
neutrons? Let us look that. We have cesium 137 it half life is extremely small 30 that is
problem we all of you are aware of the radioactive sources half life is extremely high.
So, it will continuously it emit all these rays which are some of the rays are harmful to
the human body. So, obviously, it will ray for 30 years half life is 30 years. Cobalt 60
half life is 5 years. Then we have americium 241 which is half life is 460 years is
extremely high you can see.
Now, detectors there are different sort of detectors we will use in instrumentations. In
disturbance instruments we will slowly detect we will slowly explain all this 1 by 1. We
have gas detectors gas will be ionized; that means ionization chamber; because once the
gas is ionized I can say how much the ray is. Because if the rays passes through some I
mean body; obviously some of the rays will be loosed and we will get some rays
accordingly I can calculate. Then the proportional counter all these are basically gas
detectors. Then we have Geiger Muller counter.
Geiger Muller is a basically German names there I am not writings. So, it will look like
this. So, Geiger Muller counters, so we have a scintillation counter its pensively slightly
different of the gas detectors I mean scintillation counters will discuss all these things in
details. We have solid state detectors this basically solid state detector we will see what
is the PIN junction diodes? We will see that 2 p and n junction in between 1 entry in
junction will come. So, it will make a solid state detector.
Now, this is a characteristic of radiation detectors. You see here its characteristics of the
radiation size of the pulse and are it I mean volts you see that is different region, which
region will utilize to which that we slowly avail to us.
The gas chamber operating in the different regions are region 2 and ionization chamber
region 3 proportional counter region 4 Geiger Muller counter.
Ionization chamber; you see here the chamber is normally filled with suitable gas such as
argon at atmospheric pressure. And sometimes at a higher pressure, it is always you can
know this very in hot gas. So, the chamber is normally filled with suitable gas such as
argon at atmospheric pressure and sometimes at higher pressure. An electric field is set
up between the electrodes by applying dc voltage. An electric field is set up between the
electrodes by applying dc voltage.

As radiation is incident on the chamber the gas is ionized. Obviously, and the electrons
and ions thus formed are attracted towards respective electrodes giving us the output
current, is it clear? Suppose I have we have discussed the ionization gages not yet the in
the case of low pressure measurement. So, same basic principle is saying actually you
see look at here. I mean if I have this 1; if I take this pen I have see gas is ionized
suppose these are the ion getting a suppose I am putting a meter here 2 plates. So, the
positive ion supposes positive ion will move this direction negative ion move in this
direction. So, resulting some current which will detect by ammeter. So, how much
current I will get that will be proportional to the ionization? The more the gases ionized I
will get more and more current and the gas will more. And more ionized if there are
more number of particles comes and fall on the chamber clear?
The length of the plateau in region 2 it depends on the nature of the gas inside the
chamber and on the specific ionization. Let us go back let us length of the plateau you
see here this is also we can refer we can go back to our origin discussions. So, the length
of the plateau in region 2 depends on the nature of the gas inside the chamber and on the
specific ionization. So, what type of gas is there? That is one factor and what is the
ionization specific ionization work it depends on these 2. So, the length of the plateau
will be there. A specific ionization is the number of ions is we have defined the specific
ionizations. What is the specific ionization? Number of ions formed per centimeter of
length per atmospheric pressure number of ions from per centimeter of length per
atmospheric pressure.
Proportional counter due to larger mobility and higher acceleration of the primary ions
produce secondary ions. This is known as gas multiplication. Now, you see the when the
ions move in a in a gas chamber, when the ions move in a gas chamber there are gas
molecules and it will knockout some of the outer electrons of this. So, it will further
produce some of ions; that means, a gas a when, because there is because you see that we
have we have plate is not we have discussed. We have we have plates here I have
positive ions negative ions and connecting this battery.
So, these rays are coming here ionizing the gas now, these will move with very fast
accelerations in these directions. So, whenever it is moving it will may knockout some of
the, it will may further ionize the gases or the molecules of the gas. So, it further ionizes
obviously, the more and more current I will get. So, these are the secondary ions,
because this the primary ions which is due to the incident rays whereas. When the rays
will give when this I mean the ions moves through due to the electric charge due to
electric potential on the both sides. So, it will produce secondary ions this is known as
gas multiplications.
The gas amplification factor A of 2 by n is obtained where n is the number of mean free
paths away from the centre wire anode.
The output voltage pulse remains truly proportional to the number of ions pairs. Initially
formed over the first half of the region 3 and the system are used as a proportional
counter clear. The output voltage pulse remains is truly proportional to the number of
ions formed, number of ion pairs. Initially from over the first half of the region 3 and the
system is used as a proportional counter.
Now, let us look at the Geiger Muller counter again I am telling this is should look like
this; that means, should be there. The region 4 is known as Geiger region of that
characteristics cover the detector. The property of the counter operated in this region is
the appearance of output pulses which are of the same size and independent of the
number of ions formed initially by the radiation.
Output pulse amplitudes may be as high at 100 volts it is quiet high I do not need any
further amplification. They are primarily used for the gamma radiations and in some
cases for more energetic beta radiation that is we must know. The end-on type counters
are also used for counting of alpha and beta rays and soft x-rays which can penetrate thin
windows made of mica or Mylar. In some cases it can plated in some cases it cannot.
Let us look at these scintillation counters that is figure shown 1. So, this scintillation this
incident radiations scintillation materials, we have photocathode photo multiplier photo
multiplier tube is please note it is nothing but it will multiply, because this signal the
current which will get is very small. Now, 41 function of this photo multiplier, because
incident light will fall on this photo conducting materials. It will give some current,
because some electron out of from will emit. It will give some current and photo
multiplier is nothing but the photo multiplier tube. Actually it will enhance the current
there is an immediate multiplications of the current. So, I will get higher output that is
the reason we are calling photo multiplier tube.
An incident radiation falls the energy is consumed and in the scintillation material and
the scintillation material releases photons, which are turns detected by the photo sensitive
cathode of a photo multiplier tube let us we repeat again. As the incident radiations falls
energy consume in the scintillation material releases photons which are in turn detected
whether photo sensitive cathode or photo multiplier tube.

Scintillation materials used are for alpha particle zinc sulphide then for beta and our
particles sodium iodide with thallium additive. Anthracene or naphthalene are also used
these are the basic.
Now, this is solid state detectors as I told you they have been n type and p type regions.
And we have been intrinsic I mean semiconductor in between the 2 I will get an output.
So, that would be utilized, because I need a power supply. So, that it can be utilized for
measurement.

PIN diodes are used which are nothing, but simple pn diodes with an intrinsic layer. As a
radiation falls on the intrinsic layer electron hole pairs are generated which moves in
different directions. Thus we have an output voltage which is proportional to the incident
radiations. PIN diodes are used which have nothing but simple PN diodes with an
intrinsic layer. As the radiation falls on the intrinsic layer electron hole pairs are
generated which move in different directions to different directions. Obviously, thus we
have an output voltage which is proportional to the incident radiations. Without doing
incident radiations I would not get any output voltage. However, there is always a
background thermal noise and hence they are operated at a cryogenic temperature.
Alpha detectors, what are those alpha detectors? Ionization chamber with CH 4 as a gas,
proportional counter. Beta detectors are for the beta rays, we need to have differential
mode sensing as they are generally radiated with alpha and or gamma rays.
You see we have this detector system for beta rays detectors beta which will block betas,
so through gamma, so it will come out of this. So, beta will be blocked. So, automatically
detector will come out detector output will get. The output of the above 2 is compared to
get the contribution of the beta rays, because here I am getting only gamma rays. So, I
am these 2 will be compared so that I will get only the contributions of the beta rays.

Now, neutron detectors are slightly different as you know in this case we use a indirect
method, so B 10 plus neutron lithium plus alpha. We use a detector and thereby detect
the neutrons. Obviously, will some detectors which will detect the neutrons.
Some applications density a level measurement of liquid in sealed containers. In sealed

containers if I do not if I cannot see from outside. Because in other all other conventional
methods I must take out a sample and then I can measure the density, but in the
nucleonic instrumentation is a non-invasive technique I can use to measure. So, it is for a
sealed container I can measure the density as well as level of the liquid inside the
containers. So, we have a test objects, so I am schematic of density of level measurement
I am giving a ray we are getting outputs. This I am actually this output we are measuring
how?
Let us look at I equal to I naught exponential mean into X this is basic expressions.
Obviously, what is that? Let us look at I will be I naught will be always higher than I. So,
I equal to I naught exponential minus exponential of mu x where mu is the absorption
co-efficient. Now mu is the function of automatic weight and density etcetera, mu is a
function of atomic weight and density etcetera. Obviously, that will vary the, I will be
varied in this case please look at right this will vary this I will vary clear? Hence
depending on the output signal we can tell the density of the test object. By looking at
the output signal I can look at the density of the test object because atomic weight if the
density changes. So, the atomic weight will change sorry if the, I can change the density
if the density changes that value of I will be different. So, atomic weight is the same the
densities density might be different that will be the same. For level measurement we use
the following scheme.
This is the scheme of the level measurement. So, we have source you see here we have a
detector here and. So, this is a levels measurement using nuclear instrumentation. So, I
have a liquid here. So, how it works? Let us look at when passing through the fluid the
absorption will be more and hence the amplifier output forces the stepper motor to move
the source and detector to move upward. You see what will happen? So, when passing
through the liquid this will. So, it will always try to move this stepper motor to move it
upward. When they reach the interface the absorption falls and thus we know that the
level is reached clear.
Let me do it again, so the something like this if these 2 systems, these I mean let me take;
that means, this stepper at the source and detector will be initially at the bottom here. So,
it will move up how it will move? So, it will be detected, because there is source and
detector ((Refer Time 28:35)) stop. Because incident ray and I mean this radiated ray and
the detector ray will be different, because it will slowly move excuse me we have I mean
stepper motor I mean assembly. So, that what will happen? It will slowly move now,
whenever it will reach the top surface what will happen? There is the difference of the
receiving signal, so at that time you can immediately stop it and see that what is the
level? So, that will tell you the level of the system right, clear.
So, when passing through the fluid the absorptions will be more and hence the amplifier
output forces the stepper motor to move the source and detector to move upward. So,
into the liquids the large absorptions of the open particles, so that I will get a less output.
And the detector side and when they reach the interface the absorption falls. So, no now,
we know the level has reached. So, now that I mean positions can be reflected position
that were movement of the stepper motor can be calibrated in terms of there, because you
see even though I am here. I am showing you actually in practice if it is sealed container
you cannot see them inside. So, outside should be calibrated sorry.
So, this outside should be calibrated that mean they should be graduated like this 1,
because these you cannot see this all this level even though this. So, whenever these
through this one absorption will be high. So, I will get a difference when the signals. So,
whenever it reads this top surface that is the top surface of this level absorption will fall
so; obviously, detector output will be different. So, that time I can I can stop the motor
and can see this one this is forcefully when the absorption always it will move up step by
step it will go out even that is not a continuous, but it will work. So, at this point I can
measure it and tell the level of liquid inside this vessel which is sealed from outside we
cannot see it from outside excuse me. When they reach the interface the absorption falls
and thus we know that the level is reached the level is then shown on the display.
The thickness measurement in rolling mills is very common you see here nucleonic
thickness measurements. So, I have rolling mills rolls are coming on this of the basically
rolls which is from the inwards from rolling mills; that means, I have a roller. So, rolling
measures thickness then which is I am getting a proper shape I want to measure
accordingly how much pressure I will give on the roll that can be determines. That
means how much pressure I will give on this particular roll? How much pressure it will
give on this accordingly if I give more pressure the thickness will be small. The
thickness will be reduced whereas, if I give less pressure. So, the thickness will be large
and large, but I have to measure the thickness at this is a continuous measurement. So,
that it will control and I can give the signal feed back to this rollers. So, that accordingly
pressures will be given on this and I can control the exact thickness of this devices an
exact thickness of the roll sheet which is coming out from the rolling mill in a steel plant.
Let us look at principle I will source and detector find no problem.
We know that again I equal to I naught exponential minus mu x. Thus the detected signal
will directly give us a measure of the thickness. So, the, if I have a calibrated output
suppose I have something I have made the calibrations. Accordingly I know that for this
suppose 5 centimeter thickness of the rolling sheet I know the what will be the value of I
if I I naught is constant. So, I will use that principle so; obviously, if I continuously I will
detect I which the other set where I detect I. You see I will detect I at the detector and
there is a pre-assigned value of I for these thickness 5 centimeter thickness what will be
the value of I if I do not get it suppose if it is less; that means, thickness is more . So, I
should press this roller. So, the thickness will be a proper size. Thus the detected signal
directly will give us a measure of the thickness right which can be utilized to control the
thickness of the rolling sheet which is coming out of the rolling mill.
A coating thickness measurement coating thickness measurement I have give an example

1 of the new example of the coating thickness is that, I told you in a strip chart recorder
that strip chart is coated with the wax. So, that type of situations I need to control the
thickness even though it is described it is a fine coating. So, it determinations, because
whereas, in the rolling its quiet large compact where is it is a fraction of a millimeter on
the coating, whereas there it is the much higher it is a multiple multiples of the
millimeters. Example; wax on paper plastic or aluminum foil this is a good example wax
and paper is a good example which is used as I told you in a strip chart recorder. The
above mentioned can be done the above mentioned method which have discussed can be
done in various was one of them is the differential beta transmission gage as shown in
next. What is that intensity of the detectors will vary according to the coating and the
differential meter gives the required output.
Let us look at fine I have a coating which is coating is coming through this 1 let us look
at. So, I have a coating this is a coating on the, I have a source I have detector on this
side. So, this is actually suppose this is actually paper on which I am making the wax
coating suppose this is wax this is a wax coating. So, wax is coming through this 1 and
making a coating. So, I have a source and detectors; obviously, this thickness of the
detector output you see here. So, there will get a some I mean I have a source I have a
detectors here. I have a source also see this is continuously coming through this one.
It is continuously going this role is continuously moving in this direction then what will
happen? You see here, so this source detector will give you some output I know some
differential the some pre-genering differential output I will, because why we had made
like this 1, because this thickness will be very small. So, I need some differential
arrangements. So, that it will I will get more deflections are more output. That is the
reason this differential arrangement you can say that whatever the principles which have
used for the thickness measurement of the rolling sheet can be utilized.
But if I utilize that principle then will chances lot of error, because the thickness there, it
is quiet high it is few centimeters whereas, here it is few millimeters. So, I need some
differential arrangements I have some output here and if I get this one I will get some
output so; obviously, this output will be lessened this output. So, this output will be
higher and this output will be lower. So, I will get a differential output that can be
calibrated in terms of the thickness of the coating of the wax clear.
Now, lead detection in underground pipelines see it is very difficult to if I have leak in
the underground pipelines special in the municipality in the cities. So, it is very I mean
utilize, but it should be safe as that same time, because if I use radioactive sources that
half life will be extremely high that will be problem. You see this one of the example
how you can detect? If I if there any leak in the pipe this is a taking a section of the pipe
there is a ground. And this pipe is obviously, installed few meters might be 1 meters
might be half meters below the hour or the surface level of the road right or the foot path.
Now, what actually the principle look like this we are using here Geiger Muller counter.
Basically Geiger Muller counter is used to Geiger Muller counters are using to detect the,
I mean leak in the pipe. How? Let us look at…
Inside the inflated rubber ball we have sodium 24 which is radioactive. Clear inside the
rubber ball let us look at again inside the rubber ball. We have sodium 24 which is
radioactive that with inside the rubber ball we have the sodium 24 clear fine which is
radioactive. The signal of this Geiger Muller counter is maximum at the position 2 that is
at the position of the leak? Suppose here we have a I mean position 2 this is a position 2.
So, signal at the position 2 will be higher, because there is an, if there is a leak in the
pipe, because it is slowly moving. So, leak in the pipe, so I can detect that way. So,
signal of the Geiger Muller counter is maximum at the position 2 that is at the position of
the leak, right.
We use 24 sodium as the radioactive material, because of the following reasons, because
of its half life basically the pipeline under question may be a water pipeline. And hence
we need to take a count the nucleonic hazards associated with it. So, it should only
nucleonic hazards the hazards that is we should look at 24 sodium I mean sodium 24 is
half life of 15 hours, so it is very short. So, its dissolves readily in water making sodium
chloride that is common salts; that is great advantage of this particular sodium right. So;
obviously, it dissolves and make the, I mean common salt as no problem, because almost
there is always some water is chlorinated for purifications. So, if we use the sodium;
obviously, there is a leak also of sodium 24 for our see it will not make any harm to
human being.
Now, also it is used for the nucleonic instrumentation used for sulpher content to
measure the sulpher content in crude oil. Contents of natural petroleum is carbon and
hydrogen that is it is an assume it is called hydro carbon industry all in industry we will
always call as hydro carbon industry. If sulpher is present then the atomic weight of
crude oil will increase. Because as we say we have seen that the atomic weight will
increase if atomic weight will increase. What will happen? With increase of atomic
weight the absorption coefficient mu also increases, what is that equations?
We have seen that equations what is that equations the equation looks like this I I naught
exponential minus mu x. So, the increase of atomic weight this mu absorption coefficient
mu will also will increase. Obviously, so the I will be different is not it right I will be
different if we it expand it. So, what will happen? It will be negative terms so; obviously,
if the mu increases. What will happen? So, the, I will different. So, by this I can measure
this sulpher thus the output I equal to I naught exponential minus mu x will change
giving us a measure of sulpher content in the crude oil.
Now, let us discuss little bit of nuclear medicine as I told you already are also, because in
context of the nuclear instrumentation we should discuss somewhat nuclear medicine.
Nuclear medicine takes into account radioactive material for the diagnosis of disease and
for assessment of the patient. It differs from radiography in the fact that the source of
gamma rays is not external, but rather within the patient that is the difference between
the radiography and the nuclear medicine. It also differs in the way that the radioactivity
can be attached to the material that the biologically active in the patient.

It is used to image organ function as opposed to simple organ morphology. I am sorry
this will be opposed. It measures the radioactive distribution inside the patient which can
be either static or changing with time.
This is a typical example of the sodium iodide scintillation detector you see what will
happen? We have a gamma rays coming we have a scintillation crystals sodium iodide in
your photo multiplier tube. We have a preamplifiers we have a delay line then we have a
further amplifiers I am getting a large gain. Obviously, there should be high voltage that
we have not drawn it should be a high voltage section is necessary, because photo
multiplier tube needs a high voltage to operate. So, I am getting a current and I will get I
mean further amplifiers say this is an attenuator, so accordingly I will get a full gain
here. Obviously, that is the simple system switch we were using this see I can get the
step gain or I can get different gain of the systems. This is the simple attenuator all the
instrumentation peoples always we know the people do like this one right.
Now, original pulse you will you see you look at here. I am sorry you see this actually
how it walks actually this I have shown how the pulse looks like, because I cannot show
it here. So, I took the next page how the pulse I mean this output after this scintillation
crystal coming from preamplifier how it look like? Let us look that. You see this original
pulse now, the reflected pulse is coming like this and if I can combine these 2 I will get
this pulse. This pulse is utilized for making the detection combining this 2 pulse because
the other will be canceled out. I will get a pulse like this pulse will be utilized for making
the detection and for the diagnosis purposes.

You see this is a cross sections of the focusing collimator used in the rectilinear scanning
system. Now, all the scanning system is basically you know the any scanning system
whether it is a CT scan or it is a I mean the television camera scanning. Basically you
know there is a whenever there is rectilinear motion; that means, the motion of the
detector. We makes by directions rectilinear motion is the motion in two direction x y
directions I will call it as scanning right. Here is a cross section of a focusing collimator
used in the rectilinear scanning system is a sensitivity to a point source of radiation
express as the percentage of the radiation of the focal point right, you can see here.
The sodium iodide detector is common to nearly all the instruments using nuclear
medicine imaging. It consists of 3 components the crystal itself which scintillates with
blue light in linear proportion to the energy of gamma ray losers in it. A photo multiplier
tube it converts this light into a proportional electrical signal. The support electronics it
amplifies and shapes the electrical signal into a usable form.
Simple nuclear medicine procedure do not involve images, but rather consists of simply
placing the detector near the surface of the patients skin and counting the gamma ray flux
This is very common example; obviously, non basic any process that involves rectilinear
motion of the detector is called scanning that I told you earlier.
A sodium iodide detector must be collimated in a scanning procedure in order to resist its
field of view both along the longitudinal axis and transverse to it. One such arrangement
to shown in the figure right which we have shown I mean what is that? See, how it is
collimated here shown here right? Source sensitivity of point source detection thus of
radiation expressed as a percentage of the radiation at the focal point a sodium iodide
detector must be collimated.
Now, the cross sectional view of the gamma camera, lead shield; lead shields are all
sides. You see we have a lead shield on all sides all the sides we have lead shield; we
have a hexagonal array of 19 photo multiplier tubes these are photo multiplier tubes. We
have a sodium iodide crystals here glass window through which you can crystal house.
You can see this crystal housing the optical light and this is sodium iodide crystals. You
can see here we are getting output signal x x plus and y y plus which can be given to a
like oscilloscope systems the horizontal plates and the vertical plates to get the proper
picture of gamma camera.
The gamma camera is also known as the anger camera. It is the stationary imaging
system that is simultaneously sensitive to all the radioactivity in a large field of view. It
does not depend on motion of the detector to piece together an image right.
Refer Slide Time: 46:56)
The radiation detector is a single sodium crystal 30 minus 40 centimeter in diameter and
1.2 centimeter thick. It is quiet large you can see the detector sodium iodide crystal 30 to
40 centimeter in diameter and 1.2 centimeter. The detector is viewed simultaneously by
an array of photo multiplier tubes arranged in a hexagonal fashion at the rear of the
detector as shown in the figure just we have shown. You see in the detector is viewed
simultaneously by an array of photo multiplier tubes arranged in a hexagonal fashion at
the rear of the detector.
As the gamma ray enters the sodium iodide crystal the scintillation light spreads and
each photo multiplier tube picks up a fraction of this light right clear? However the
gamma ray enter, let us go back again. Gamma rays enter through this thing we have
sodium iodide crystals it is scintillation lights it is going to the photo multiplier tubes I
am getting the vertical output clear? As a gamma ray enter sodium iodide crystal the
scintillation light spreads and each photo multiplier tube picks up the fraction of light.
The fraction of the light seen by each tube determines the position of the original point of
entry of the gamma ray clear fraction of the total light seen by each tube determines the
position of the original point of entry of the gamma ray. Now, voltage corresponding to
the x-y coordinates of the gamma ray are reconstructed from a signals of the photo
multiplier tube. This is done basically where the analog electronic circuit.
Now, in the single photo emission computer tomography it looks like this a SPECT uses
a large area scintillation assembly similar anger camera or the gamma camera and rotates
it around the patient it rotates around the patient. Many of the isotopes used in nuclear
medicine produce a single photon or gamma ray in the useful range of the energy clear.
The collimator is designed to collect radiation from parallel rays. And these are used in
the process of an image reconstruction which is analogous to the methods used in
computed tomography. Multiple planes or slices are reconstructed at the same time
which helps us helps us to show both information the volume of interest by constructing
several cross sectional views in addition to these offered by conventional nuclear camera.
Now, positron emission tomography is slight different. Certain isotopes produce

positrons that react with electrons and produce 2 positrons at 511 kilo volts at opposite
directions. Positron emission tomography uses the property to determine the source of
radiation.

The isotopes have two means of decay the nucleus can capture an orbital electron then
that combines with one positive charges. So, the nucleus can capture one orbital electron
that combines one positive charges and nucleus can emit a positive charge in the form of
positron that combines with the external electron right. So, these are 2 method by which
you can I mean method of decay of the isotopes; obviously, all the radioactive materials
basically is decay that you must know.
The combinations of negative and positive particles annihilates the charges and masses
of each. Energy and momentum are conserved and two 511 kilo volt gamma rays are
emitted in opposite direction. The property of simultaneous emission of gamma rays in
opposite direction gives a positron emission tomography and the ability to locate the
region of origin right. So, that is the necessary I mean you must know the origin.
Instead of multiple multihole collimeter as in most of the gamma cameras two imaging
detectors capable of determining x-y positions are used right. And each x-y pair is
accepted if the two scintillation effects are coincident and have energy levels close to the
expected value of 511 kilo electron volt. The simplest PET uses two modified anger
cameras placed at the opposite sides of the patient as shown in the figure.
You can see this is a paired camera how the, is rotating. So, that continuous, so it is
rotated in this directions. So, it will in directions, so the patient will be covered by 2
angular camera and please note that in the nuclear instrumentation the horizontal is a
great problem. So, you have to think of all this whenever you are using what type of I
mean radioactive materials you are using though the half life. And all these things is not
that large, but the amount of ratio you have to think also the proper hazards. And it is
sometime environment list they are show until unless you can manage with some other
instrumentation which is known as hazard. Hazard as like ultrasonic or any conventional
instrumentation of level gage and density measure that is to be used.
But in some situations as I told you like the detections of the leak in the pipe
underground or suppose the in a close containers I want to know the level of the liquids.
And all things, it is very difficult to specially it is sealed of the might be closed it turned
does not matter because if I have a inside electronics already installed inside. So, I can
send some installed inside I can detect, but if it is separate content suppose a cylindrical
drum I want to know the how much the liquid contents which we cannot save from
outside? So, that type of situation I can use this type of radiation technique. So, this is to
be basically I mean all you have to be careful that I told you several times. That means,
they can manage with other instrumentations systems and mention non-hazardous other
systems. We are not suppose to use the radioactive sources right. So, with this I come to
the end of this lesson 26.
Welcome to the lesson 27 of industrial instrumentation. In this particular lesson, we will

study basically the measurement of magnetic field. Basically 2 sensors we consider
though the title of the lesson of magnetic field, but we will find that specially the sensor
the second sensor which we had. So, we will we will study the measurement of current
measurement of power all those things, but the first sensors that is the search coil is
basically used for measurement of magnetic field. And second 1 is the, that will that can
be used for measurement of magnetic field as well as current.
In fact, we will find that this particular sensor is extensively utilized though it is named
that it is used for the measure of the magnetic field. But you will find it is extensively
used for the measurement of current especially high current as well as the measurement
of position of the proximity sensor measurement of power multiplication of the 2 signals.
This all those things we will see what are the different applications areas of this all
effectors users.
Let us look at the contents and all those things.

Contents are search coil magnetometer; search coil will basically discuss here. So,
magnetometer search coil looks like this that you have a as we know that if a magnetic
field that if you know some coil anywhere some voltage will be develop. So, using that
principle that voltage will measure and that voltage will be calibrated in terms of
magnetic field. If the other things remains constant; obviously, we find that the output
voltage will directly proportional to the magnetic field. So, that is the search coil or
magnetometer search coil sometimes people call magnetometer search coil sometimes
we call it simple search coil right. And second transducers, we will consider in details is
half effect transducers as I many times amplifies or something like. That means, I will
apply magnetic field which is coming through this going out this right I am applying a
current electric current which is entering through this and going out of this. That means, I
have applied some voltages that it is in some current is flowing. If I apply a voltage
magnetic field here we will find that across these 2 surface this and this a potential will
be developed that potential is actually we call it half coefficient right.
Let us look at the problem again this is our problem. So, given hall coefficient for
germanium crystal equal to minus 8 into 10 powers minus 3 volt meter ampere Weber
meter square.
Solutions; Eh equal to KhBi by t we have seen. So, B equal Eh into t by Kh into I. So,
this you just put all this value of B. So, B equal to 0.012 Weber meter squared.
So, now, when the half field equal to Lorenz force must have a balance. So, I can write it
is the moment of the velocity; V is the velocity of the electrons inside the all sensors. So,
eEh by b. So, V equal to Eh by B by small b let me go, yes. So, which will give you the
speed of the electrons as 972 these are all the dimensions of the hall plates, b is the
dimension hall plates 9722 meter per second. With this I come to the end of the lesson 27
of industrial instrumentation.
Prof. A. Barua
Lecture - 27
Measurement of Magnetic Field
Lesson 27 of industrial instrumentation, in this particular lesson, we will study basically

measurement of magnetic field. Basically 2 sensors will consider here though the title of
the lesson is measurement of magnetic field. But you will find that especially the sensor
second sensors which we had. So, we will study the measurement of current
measurement of power all those things. But the first sensors we that is the search coil is
basically used for measurement of magnetic field. And second one is the Hall Effect
transducers that will that can be used for measurement of magnetic field as well as
current.
In fact, we will find that this particular sensors is extensively utilized thought it is named
that it is used for the measurement of magnetic field. But you will find that it is
extensively used for the measurement of currents especially the high current as well as
the measurement of position of the proximity sensor. Measurement of power
multiplications of the 2 signals these all those things we will see the, what of the
different application areas of these hall effect transducers?
Let us look at the contents and all those things. Contents are search coil magnetometer
search coil will basically discussed here. So, magnetometer search coil looks like this
that you have a as we know that if a magnetic field that if you move some coil anywhere
some voltage will be developed. So, using that, principles that voltage will be measured
and that voltage will be calibrated in terms of terms of the magnetic field right. If the
other things remains constants; obviously, we will find that the output voltage will be
directly proportional to the magnetic field. So, that is the search coil or magnetometer
search coil sometimes people called magnetometer search coil sometimes we call it
simple search coil right. And second transducer we will considering in details is the hall,
effect transducers right.
At the end of the lesson the viewer will know the relation between various parameters
like the magnetic field strength, flux density. The output voltage of the coil total flux
through the loop etcetera in case of magnetometer search coil right. So, this thing
basically will know and in the case of in the case of Hall Effect transducers. We will
know the how parameters like flux density, dimension of the crystal, its hall coefficient
current density, hall voltage etcetera are related. In case of Hall Effect transducers and
how to use it to measure the current flux density power etcetera will also study in this
particular lesson.
Now, magnetometer search coil is basis principle is something like this a schematic of
magnetometer search coil is shown in figure 1. A flat coil let us go back and see this one
then will this come again here.
See this is the a coil which is and it is placed under magnetic field right and its cross
sectional area is A length is L, and I am getting output voltage here E. Actually these
output voltage is to be to get some output voltage, please go the 2 things is important
either the magnetic field must vary or this coil was change its position. There should be
some movement of the coil there is 2 option by which I will get a nonzero output
voltage. One is that either H or B if the H varies, because as you know B equal to mu H
H or B will vary or the position of the coil will vary.
Then only I will get a nonzero output voltage this output voltage can be calibrated in the
terms of the magnetic field. So, that is the basis principles of the magnetometer search
coil. Let us go back to the previous slide right. So, schematic of magnetometer search
coil is shown in figure 1 that is already we had shown right. A flat coil with N turns is
placed in the magnetic field as shown in figure right is a flat coil please note its coil can
be 2 types as you know. What is that flat coil placed let us go back and see a flat coil
means like this.
The, you know the coil can be wound on a and it have a shape like this one. It can be I
will usually before making inductance you make like this one right and a coil can be like
this one also.
That insole a same I mean plane I can have the several turn; that means, I have can have
a several turn of the coil. If I look from the side it will have a just a several length of the
wires and which is coming right. So, it will look like this one that is the reason we
calling it flat coil there is some reasons why we are calling all these flat. If it is the
dimensions of the coil is very important to properly predict the output voltage. If though
we will calibrate the instrument, but that is very important.
A flat coil with N number of turns is placed in the magnetic field as shown in figure 1
right N numbers of turns of there it is a flat coil. Obviously, the more number of turns I
will get more output voltage that is also very important. But as we increase the number
of turns the size of the coil also will increase the whole flatness. Whatever we are
assuming that will be valid and more over the weight of the coil also will be quite large.
So, that creates other problems. The length of the coil is L and the cross sectional area is
A right. What is this cross sectional area let us again look at.
I want to mean that suppose I have a coil of N turns. So, this filled up area I am taking of
the cross sectional area. A right this cross sectional area of the well. And it is not the
cross sectional of the wire by which I am making the coil it is a cross sectional area of
the coils; that means, if I just simply drown 1 simply look like this one right. So, I am
taking of the this cross this is our cross sectional area this is the output voltage I will
measure right which will be calibrated in terms of magnetic field. So, length of the coil is
L and the cross sectional area is A will all talk about SI unit A. So, this is the all we will
talked about. It is a flux density corrected to magnetic field H by mu and this is the cross
sectional area A this E output voltage which is coming from the coil.
The magnetic field strength H and magnetic flux density B are in the direction shown
and are related by the following expression. We have already shown the same direction it
is working you see magnetic field is also working in this directions flux density is also;
obviously, work in the same directions right. So, the magnetic field strength H and the
magnetic flux density B are in the direction shown, and are related by the following
expressions B equal to mu into H. What are these where B is a flux magnetic flux density
which is in Weber per meter square Weber per meter square. And H is a magnetic field
strength which we want to measure that is ampere per meter right. We are interested to
measure the magnetic field strength and mu is a magnetic permeability for the free space
which is 4 pi. I am sorry this will be 4 pi. This will be 4 pi into 10 to the power minus 7
Henry per meter 4 pi 10 to the power minus 7 Henry per meter right.
So, the voltage output of the coil E is given by now you see one thing is very important
you see here in the previous slide we are interested in the measurements of H right. The
how much is the magnetic field actually the value of B this point be constant this coming
from some magnetic material. The value of H depending on the medium whatever the
mediums, which we are applying these, magnetic field I will get a value of B. So, B B
may vary depending on the particular medium suppose in water I have some difference
value of B; obviously, mu will vary in the free space. We will have different value of B
because mu will some other value, but H will remain constant is for some specific
application. So, we are more interested to measure the H rather than the measurement of
flux density, clear? The voltage output of the coil E is given by E equal to NA into cos
alpha into dB by dt suddenly why how come this alpha came?
Actually we are assuming in the actual diagram you see if I go back to the diagram I
have shown I have shown it like that. That H is perpendicular to the plane of the flat coil
is not it. You see the H is perpendicular to the plane of the flat coil, but that is not the
fact that be not be the fact though it is also less desirable, that it should be perpendicular
that is output will be higher.
If it comes in an angle that; that means, I am saying if it comes in angle like this 1. So, I
have suppose I have a. So, I am applying this magnetic field like this 1 suppose the
magnetic field which is making an angle alpha right. So, I am taking of that alpha please
note right. Again I am remain reminding this is pi naught omega the voltage output of the
coil E is given by E equal to N into A into cos alpha into dB by dt.
That is derivative of the flux density with respect to time. Now; obviously, it will be dH
by dt, because mu will come out which is to be constant for a medium. Where A is the
area of cross section of the coil, which is in meter square; obviously, in SI unit and alpha
is the angle formed between the directions of the magnetic field. And a line drawn
perpendicular to the plane of the coil just what I have whatever I have shown. What is
that exactly like this one?
Just I have shown; that means, the I have a coil here even though I could not draw it in a
single plane we assume that it is in a single plane. So, this is the perpendicular to the
plane. So, this is our alpha clear or this is our now alpha is the angle formed between the
directions of the magnetic field and a line drawn perpendicular the, to the plane of the
coil, clear? N is the number of turns of the coil as we increase the number of turns;
obviously, the output voltage will increase that I told you earlier also. If you A increase,
because, but the magnetometer search coil is usually we are of this small dimensions.
Because it is very easy to say that the, if you increase A, you will have the I mean larger
value of the output voltage. But if you think of the large A keeping the shape of the, that
entire coil intact I mean a large A and rotating it at a high speed it is not very task right.
The total flux through the loop is; that means, the loop of the coil is given by phi equal to
A cos alpha into B right this is the total flux in the loop. So, that E equal to N d phi by dt;
that means, derivative of phi into dt that is; obviously, always is not I mean voltage. That
will be N multiplied by d phi by dt if it is an angle A cos alpha right. So, voltage output
of the transducer is proportional to the rate of change of magnetic field as you rotate it in
a high speed. So, the, you output voltage also will that mean rate of change of flux will
be directly proportional to the output voltage faster the rate of change of flux through the
coil; that means, faster rate if it is rotates I will get the larger output voltage. So, that is
we are telling voltage output of the transducer is proportional to the rate of change of the
magnetic field.
If a stationary coil now is placed in a steady state magnetic field; that means, a DC
magnetic field. That means, a field which is produced by a DC voltage or a permanent
magnate there will be no output voltage, because d phi by dt will be 0. So, there will be
no change of flux no variations of the flux with respect to time. So, the output voltage
will be; obviously, 0. That we can see, because we have seen this part will be 0. So,
output voltage will be 0. The search coil is thus a transducer that transforms a magnetic
field signal into a voltage. That means, it converts the magnetic field into z voltage
which is can be easily transmitted can be converted to the digital domain. So, all those
things facilities are already there which can be easily calibrated also if you have a
standard magnetic field.
In order to perform a measurement of a steady magnetic field it is necessary to provide

some movement of the search coil quite. Obviously, as I told you if it is steady that is the
first bullet you see if this is a stationary. Obviously, if it is a stationary placed magnetic
field no output or you will not get any output right. So, in order to perform a
measurement of a steady magnetic field it is necessary to provide some movement of the
search coil right. Some movement of the search coil will be necessary. So, I cannot
change the magnetic field which I am going to measure. So; obviously, I have to make
some movement of the coil. How let us look at how actually I am doing? A typical
method is to use a rotating coil as shown in the figures with some motors and it will
rotate at continuous speed.
So, let us look at that the rms value of the output voltage for such a device will be given
by E rms 1 by root 2 NAB into omega. What are the other things, what are the regions
where omega is the angular velocity of rotation of the coil the figures I will show you.
Let us go back to figure first because it is very reticules if I see this is our coil.
It is again flat coil in turn flat coils. So, it is rotating by angular speed of omega you see
angular axis of rotation. So, it is continuously rotating I have placed. This entire coil is
placed in a magnetic field of H with flux density B the coil is of length L. So, it will
produce the output voltage which will be measured at this 2 terminals which term which
are those terminals this terminals I will get the output voltage B, clear? So, this will be
the output voltage E let us go back the search coil is thus a transducer that transforms a
magnetic field signal into a voltage. And in order to perform a measurement of a steady
magnetic field it is necessary to provide some movement of the search coil. So, some
movement of the search coil is necessary that actually I am doing in the figure 2 as I
have shown right. So, this is our coil now there is some constraint here please note 1
thing that the constraint is that first of all the coils should be of small size.
And the dimensions of the coil should be very well known. The exact precise geometry
of the coil should be known to predict the value of the output voltage. And as I told you
the smaller the size of the coil; obviously, you will get the better results. That you must
be mention I mean it is worthy to mention that the size of the coil should be very, very
small. So, this is the constraint that the coil should be small enough the magnetic field
constant over its area. And this coil should move in a sufficiently high speed right this is
another important thing to get a steady state value.
Now, let us solve some problem. So, that we will be more comfortable in the, so the
problem is a search coil has 10 turns and with a cross sectional area of 10 centimeter
square. It rotates at a constant speed of hundred revulsion per minute the output voltage
is eighty mille volt calculate the magnetic field strength. Let us take a white page I will
give the solution to this.
So, this is the solution to problem 27.1 like this 1 you see it is given that B equal to root
2 E rms by NA omega. So, which will be root 2 into 0.08 by 10 it is 10 centimeter square
the cross sectional area number of turns 10 right. So, the, it is in the centimeter square.
So, it is converted in the meter square. So, 10 to the power minus 4, so it you see this is
the number of turns right this is the dimension in the problem it is given 10 centimeter
square, clear? So, I convert it in the meter into 100 because it is 100 multiplied by 2 pi by
60. So, which will give you 1.08 Weber per meter square? So, the H will be given by B
by mu which will be given by 1.08 divided by 4 pi into 10 to the power minus 7 10 to the
power minus 7 into 8.6 10 to the power 5 ampere per meter right. This is the problem
number solution to the problem number 27.1 right we have another problem.
This is the problem 27.2. The problem is a magnetometer search coil has a nominal area
of 1 centimeter square with 100 turns the rotational speed is nominally at 180 rpm.
Calculate the voltage of output when the coil is placed in a magnetic field of 1 Weber by
meter square please note it is not omega this is W right. And this will be simply W or I
should take please note this is W right Weber per meter square. So, let us make the
solution of this problem. So, again I am telling this is W.

So, again you see the solution to problem 27.2, B equal to root 2 E rms NAW E rms
equal to BNA omega by root 2 you see it is 1 then, because it is 1 centimeter square. So,
multiplied by 100 turns right into magnetic field is the this is the B is 1 Weber. So, it is 1
N is 100 into 1 converted into meter square 10 to the power minus 4 multiplied by 180
excuse me 2 pi by 60 divided by root 2. So, this will give you 1.27278 volts right. So,
this is the solution to the problem 27.2 right. So, we have solved 2 problems.
Now, let us come to the more I mean I mean the other sensors rather I should not I mean
say this is the more important and that is the less important. We have this you see this is
even though which is very easy to tell that you rotate the magnetic field I mean coil all
those things. It is not very easy to do that and its dimensions everything is very I mean
very difficult to predict all those values. So, people what they do, because we have much
better sensors much better devices which is called the Hall Effect transducers. It is
actually developed by the 1 scientist engineer I should say name is Edwin Hall.
So, the name also the sensor is also given Hall Effect actually this effect he discovered
do while during the other experiment. And the sensors later came I mean came on. So,
the sensors also named under him; that means, the hall effect transducers or hall effect
sensor now what is hall effect let us look at. The Hall Effect refers to the potential
difference which is calling? We are calling hall voltage on opposite side of a thin sheet
of conducting or semi conducting material in the form of hall bar through which a
electric current is flowing. And a magnetic field is applied perpendicular to the hall
element. I will repeat it I am a bar looks like this 1, if I take this 1 as a I mean I mean
many times amplified. So, something like that; that means, I will apply a magnetic field
which is coming through this and going out this right.
I am applying a current electric current, which is entering through this and going out of
this; that means, I have applied some voltages. That is the reason the some current is
flowing. If I apply a voltage I mean magnetic field here you will find that across these 2
surface these and this the potential will be developed. That potential is actually we are
calling it a hall coefficient right hall voltage rather. So, the Hall Effect refers to the
potential difference all hall voltage on opposite side of a thin sheet of conducting or semi
conducting material in the form of a hall bar through which electric current is flowing.
And a magnetic field applied perpendicular to the hall element.
The ratio of the voltage created to the amount of current is known as the hall resistance.
And the characteristics of the material of the this is the characteristics of the material of
the transducers and the hall effect sensors is shown in figure 3.
Let us look at Hall Effect transducers this is our Hall Effect transducer this is a thin metal
plate of thickness t. So, I applied a voltage B sorry a potential B and you see this is a
basically a Lorentz I mean Lorentz effect will come. In fact, some people say that the
current carrying. That the current actually carrying by the electrons the actually that is
first proved by the hall Edwin Hall right. So, I am applying a potential here. So, battery
you see the current is flowing through this 1.
And some voltage will be developed across this that is I have shown like this 1, clear?
Very clear, I will look at this 1 the same thing I am applying a magnetic field here. That
means I am applying magnetic field here a current is flowing through this 1 applying
current it is coming and it is going out through this right. So, hall potentials if you take
these 2 terminals voltage will be developed that voltage is called as a hall voltage clear
again right. So, it should be sufficiently thin.
A semiconductor plate actually you will find that the hall coefficients or hall voltage is
moved predominant in the case of semiconductor though in the all metals supposed to
have some I mean hall voltage. But you will in the case of semiconductor this is more
predominant. A semiconductor plate of thickness t is connected as shown. So, that an
external current I passes through the material due to some battery voltage and when a
magnetic field is impressed on the plate in a direction perpendicular to the surface of the
plate. There will be a potential EH the, that is the hall voltage is subscript it generated as
shown.
This potential is called the hall voltage and is given by EH equal to KH into I into B by t.
This is our expressions of the hall voltage right. So, a semiconductor plate as a I told you
it will have a larger value of the hall voltages the hall voltage of thickness t is connected
as shown. So, that an external current I passes through the material when a magnetic
field is impressed on the plate in a direction perpendicular to the surface of the plate.
There will be a potential EH generated as shown and this potential is called the hall
voltage and is given by EH and KH IB divided by t, clear?
Where I is in amperes B is in gauss and t is in centimeters right the proportionality

constant is called the hall coefficient. And has the units of volt centimeters per ampere
gauss right. So, it is a hall coefficients and is the characteristics of the particular material
bar has the unit of the volt centimeters per ampere gauss.
Hall effect device produce a very low signal level. So, that it is to be amplified. So, it
needs an amplifications, because the voltage is very, very small please note it it is not a
very large voltage. So, it could amplifier known as amplifier is necessary though the
voltage DC. So, that simplifies a lot of problem right. Hall Effect sensors are in fact, a
device containing both the sensors described. And a high gain IC amplifier in a single
package. Nowadays, these are made I think in 1000s and 1000s, because of the, because
the hall effect transducers. We can see that if I know the, if I can predict that EH I can
measure the current also it is not it if I know the potential let us go back it will be, clear?
You see here if I know the EH of the material if I know the B if I know the, if I i know
the EH if I can measure the EH. This EH can be calibrated in terms of I is not it if the, I
is unknown.
So, by measuring EH I can tell the, I clear because this is the only thing variable. So, it
Hall Effect sensors the, that is the reason hall effect sensor are extensively used for
measurement of current large current very large current around 500 amperes and all
those things. There are various types of Hall Effect sensors are also used and more over
suppose if this I is constant if I can vary B I can use this Hall Effect voltage EH to use it
for proximity sensor, clear? How let us look at Hall Effect sensors are in fact, a device
containing both the sensor and described and a high gain IC amplifier in a single
package.
Advantage of the hall effect sensor hall effect devices when appropriately packaged are
immune to dust dirt mud and water in many places in hostile environments I may use
this. It does not depend on how much dirty it is, it will work same right calibration I do
not have to recalibrate the instrument that is the great advantage of the. So, Hall Effect
devices when appropriately packaged or immune to dust dirt mud and water. So, any is
that type of environments I can use this Hall Effect sensor especially for the
measurement of current. That is it is most extensive nowadays not for the used not for
the measurement of magnetic field though we only just started this lesson. And, so I said
that the, I mean measurement of we are actually we are discussing the measurement of
magnetic field.
You please note this since it is Hall Effect transducers; we are actually discussing
historically actually the all effect transducers are used for measurement of magnetic
field. But it is nowadays extensively used for measurement of current excuse me. These
characteristics make Hall Effect devices superior for position sensing. This is another
important proximity sensing that I told you compared to alternative means such as
optical and electromechanical sensing. Electromechanical sensing does not work in the
dirt as you know, because dust will always block the of light. So, this optical system also
does not work for a very, very small change. So, I can use this hall effect we will show
that how would actually it works.
Hall Effect current sensor; this is the current sensor, based on Hall Effect you see the you
see here a current. I want to measure this current, which is flowing through this wire.
This is the section of the wire, which I have shown section of the wire you see this is the
section of the wire it is placed under a magnetic field the ferrite core. This is our hall
sensors right and I am sending a constant current through this where current is entering
here current is coming out of this you see the current is coming to through the here. And
going through this and coming out through this one right. I am measuring the output
voltage across these 2 terminals these, this terminal and this terminal the output voltage.
Let me, the take another colors I think this is not very, clear?
These hall and this I am measuring the voltage right. Now, what will happen? You see
that this since it is a ferrite core. So, the amount of magnetic field also will depends on
the current which is passing through this one. As we have shown that the, this EH the
output voltage or hall voltage depends on the magnetic flux density. And this flux
density will depend on the sensed current which is the current which is flowing through
the wire right. Along with in conjunction with the ferrite core, so that will give you
different value of B for different value of the current right. So, accordingly what will
happen this output voltage will vary? So, this output voltage will be celebrated in terms
of the sensed current. This is the basis principles of our Hall Effect current sensor, clear?
If currents of both AC and DC are to be measured sensor based on the Hall Effect or the
magnetic I mean resistance magneto resistance effect can be used right. In a typical open
loop sensor the current carrying conductor is passed through the hole in a gapped ferrite
core used to concentrate the magnetic field that is we have shown. It is the gap ferrite
core what is the gap you see it is not continuous it is there is a gap after this there is here.
So, there is a gap. So, inside that gap we put this 1 why this gap to concentrate these
magnetic field. And this magnetic field depends on the current which is flowing right it
does not depend on the ferrite core only depends in the current which is flowing through
the line right.
So; that means, B, so the B will change due to if the current changes. So, the output
voltage will change. So, output voltage calibrated in terms of sensed current. So, in a
typical open loop sensor the current carrying conductor, because it is a open loop
sensors, because in the hall some hall effect sensors are used in closed loop control also.
So, that is the reason we are using the term 2 different term open loop and closed loop. In
a typical open loop sensor the current carrying conductor is passed through the hole in a
gapped ferrite core used to connect concentrate the magnetic field. And the hall device is
in the gap.
As we know the hall sensor is a magneto sensitive semiconductor that provides an output
voltage proportional to the product of its current, which held constant. And the
component of the magnetic field that is perpendicular to its surface. We have seen
previously this is. So, many times right the hall sensor is a magneto sensitive
semiconductor. That provides an output voltage proportional to the product of the current
which is held constants. So, if I use a batter voltage constant, so it will remain constant
and the component of magnetic field that is perpendicular to its surface right. Since this
field is proportional to the current being measured that I told you several times twice the
device output voltage is proportional to the sensed current. This is the most important
this second point right. What is that since this field is proportional to the current being
measured right field will depends on the current. And the device output voltage is
proportional to the sensed current. So, using this principles we are making this hall effect
current sensors as simple as that. A typical device has a range of has as I told you 0 to
350 ampere and a frequency response from a DC to 1000 Hertz.
Hall Effect current sensors is like this one. What is this DC? I am talking about suddenly
a typical device has a range of 0 to 350 ampere that is understandable it can measure a
current of 350 Ampere. And please note it is a non invasive device not contact device if
you want to measure I i do not have to put nanometer I do not have to cut the wire right.
That is the great advantage any meter measurement of current will be I mean is necessary
we have previously the CT BT. And all those things current transformer potential
transformer a little absolute that type of instruments. So, I can use the Hall Effect
transducers there, clear? Totally you do not have to cut the wire to measure, because
whenever you want to measure a current in the wirer you have to break the wire.
And connect a nanometer, but which is not necessary here right. So, when the online;
that means, when the circuit is working fine you can bring a hall effect transducer set up
like which I have described. And measure the current which is passing through the coil.
Frequency response of 0 to 1000 Hertz what is this suddenly frequency response. That
means, can I measure AC. AC current here yes why not the output voltage which will get
that 1 also will be AC, but the magnetic field remains constant that is DC no that also
will vary. Obviously, if this is AC that B will also vary. So, the output voltage will also
will be AC.
The Hall Effect current sensors the magnetic field may be that provided as a
consequence of electrons flowing through a conductor. So, this is as I told you earlier,
because this Hall Effect hall voltage actually developed due to the flow of electron in the
semiconductor. In the conductor or in the metal or in the semiconductor, because some
people say that the actually the metal that this the current flowing through the metal is
due to electron was first established by the Edwin Hall.
Magnetic field may be that provided as a consequence of electron flowing through a

conductor. It is thus possible to create a non contacting current sensor which I told you
sometime back in which the conducting cable with current to be measured is threaded
through a hole in the sensing devices. Just through a hole in the sensing device. So, I will
show you such a devices that more clear to you. The device has 3 terminals, let us look at
this is our current sensors.
You see the the wire through I mean which will be in which we are interested to measure
the current will pass through this please note. The wire will pass through this one. So, it
will go out like this 1, clear? So, this should pass through this 1. So, this is our compact
systems or our current sensor right. Hall Effect current sensor with internal integrated
circuit amplifier this is total 8 mille meter opening 0 current output voltage is midway
between the supply voltage that maintain at 4 to 8 volt differential. Non-zero current
response is proportional to the voltage supplied and is linear to 60 ampere for this
particular which it 25 ampere device.

Across 2 of these is applied a sensor voltage because I have to give a voltage is not it.
And from the third is taken a voltage proportional to the current being measured right.
So, the across the 2 of these I have applied a sensor voltage which actually will give you
the current I capital I you remember our basis expressions. What is that let us look at
when I am telling that you see here KH IB into divided by t. So, this I is remaining
constant across the 2 of these I have applied a sensor voltage. So, the sensor voltage will
make this I constants KH B t everything remains same I have measured EH will measure
the current.
So, current also will vary the B actually. This has several advantages. So, no resistance
are shunt needed, because you see the typical measured of current high large current like
500 Ampere I mean 6 and 1000 ampere we are using a shunt. What is that shunt is basic
principles all of you know form the first year classes.
I have a nanometer, which can measure suppose I have a nanometer suppose it can
measure only 0 to 1 ampere. But I want to measure a hundred ampere. So, what I will
do? I will put a shunt a low resistance across this 1 R shunt. So, large current will pass
through. So, this Rsh is much, much smaller than the resistance of the nanometer ram, so
the large current most of the current if I want to measure. Now, we want to change it to
100 Ampere. So, most of the current will pass through shunt the small current will pass
through this. Again the same problem the shunt and all that is not necessary in this
particular case right.
So, this has a several advantage. So, no resistance a shunt needs to be inserted in the
primary circuit. And also the voltage present on the line to be sensed is not transmitted.
To the sensor and a characteristics which enhance the safety of the measuring equipment
that is most important thing. You see what is I am saying also the voltage present on the
line to be sensed is not transmitted to the sensor. So, the, whatever the voltage, because I
am measuring the current, so that that is will be associated with the large voltage might
be. So, that voltage will not come to the come to EH, because EH whatever we are taking
about that will be these voltage what I am saying that you see these EH, I am sorry.
This EH KH IB by t B is changing getting changed due to the sensed current. This

sensed current was associated with large voltage fatal voltage where the order of kilo
volt. But that voltage is no way connected here is not coming to the measured I mean
output voltage of the sensor, because that will give you some value of the B. So, that is
the, I changes; that means, as the currents which I was sensing that is changing will you
please do not confuse. These with the, I these I with the actually the current we are
measuring. These current this small I this I sorry this I is actually the the DC current,
which is established in the hall sensor by a steady voltage that is a very small voltage.
We are not measuring this current I we are a measuring a current which is changing the
value of B right and we are measuring the EH. So, that would might be associated with
large voltage fatal voltage, but that is not come to the picture that is what is that will not
come to the sensed output. That means, the terminal which we are using for
measurement of that voltage the hall voltage is no way getting that high fatal voltage is
not it, because dangerous voltage I mean high tension voltage. So, that is no way
connected, because it is no way totally non contact sort of devices. There is no physical
connection between the current, which I am going to sense or I am going to measure and
the output voltage right no mechanical connection nothing else.
Range of the current sensor the range of a given feed through sensors may be extended
upward and downward by appropriate wiring. So, by I mean you can wiring; that means,
I can make it larger and smaller. To extend the range to lower currents multiple turns of
the current carrying wire may be made through the opening that, because there is an
opening through which current sensors will go.
To extend the range or to the higher currents say current divider may be used with a
portion of the current carried by a large wire flowing through a smaller parallel wire with
the thicker wire passing through the sensor. This is the same principles of turn, but that is
not very popular in that sense; that means. I have to take some other I mean is that same
shunt business I am introducing again. So, I should not say it is very popular or usually
typically you see the hall sensors are used for range. The same hall sensors which are
used for measurement of current of 100 Ampere you are not using that hall sensor for
measurement of 1000 Ampere for 1000 Ampere. You passed of different I mean all
together different sensor assembly.
We can have a split ring on the clamp on current sensors also. A variations on the ring
sensor uses a split sensor which is clamped onto the line rather than the threading the line
through the sensor enabling. The devices to be included in the test equipment and it is
not permanently installed in the device being tested excuse me. This also simplifies the
permanent addition of current sensing to existing circuit as they need not to be
dismantled to perform the installations. You do not have to lose I mean dismantle those
thing right this is one way I mean though. It is I mean I mean we are using a clamp on
sensor this is one way it is going to because you do not have to dismantle the entire
system.
Now, analog multiplications because you see the, you will be surprised suddenly how
come. We are come to the analog multiplication I mean we are taking about you see the
analog multiplications is a, we are talking about, because though. It is a signals I mean I
cannot I mean 2 signals can multiplied we are, so for taking about this excuse me.
Suddenly how it can develop multiplication business let us look at the output is
proportional to both the applied magnetic field and the applied sensor voltage right.
Now, output is proportional to the both the applied magnetic field applied sensor voltage
it is not it you see the applied sensor voltage, what is let us go back again, sorry.
Again we are telling that we are writing EH KH into IB into t right KH is constant t is
constant. So, output voltage EH I can say that output voltage EH is proportional to I
multiplied by B is not it that I can tell quite; obviously, if KH and t t are constant. But the
is that you see that how I am saying that the, these 2 signals will be multiplied. What is
B? Again B will depend on the magnetic field now if I can make B which is proportional
to some other current to this. I we will be I I will depend on the current, which is flowing
through the sensor B will. If I make the B proportional to some current so; obviously, I
can say that the output voltage will be multiplied output voltage will be proportional to
the multiplications of the 2 currents is not it? That is exactly what we are doing in the
analog multiplication let us look at.
Now, let us go back to the output voltage output is proportional to the both the applied
magnetic field and applied sensor voltage right. If the magnetic field is applied by a
solenoid then the sensor output is proportional to the product of the current through the
solenoid and the sensor voltage is not it? So, using that principles I am making the
analog multiplication.
Power sensing: by sensing the current provided to a load and using the device applied
voltage as a sensor voltage it is possible to determine the power flowing through the
device right. This power is for direct current devices the product of the current and the
voltage. With appropriate refinement and the devices may be applied to the alternating
current applications where they are capable of reading, the true power produced or
consumed by the device.

Position and motion sensing; that is most important thing which is when it is applied to
the mechanical motion sensing and the motion limit switches, the hall effect device can
offer enhanced reliability in the extreme environment. As there are no moving parts
involved within the sensor or magnet there is a far greater useful life expected than from
the electro mechanical switches right. So, again the, I mean position sensing is possible;
that means, if I by some means I am changing the position I can change the value of B
capital B we have capital B we have written. So, many times, so the; obviously, the
output voltage will be proportional to the B. So, I can sense the position, so the position
sensor will you make position sensors using this principle these all principle.
The sensor and the magnet may be permanently and completely encapsulated in an
appropriate material.
Now, let us look at the problem last problem 27.3. Let us read it which is the a
germanium crystal having a dimension of a dimension of 6 into 6 into 3 millimeter is
used for the measurement of flux density using hall effect sensor when the hall field and
the Lorentz force balanced each other. It is observed that the current density in the
crystals is 0.3 ampere per millimeter square. And the hall voltage developed is minus
0.35 volt we have to find the value of the flux density and the velocity of the electrons in
the all sensor itself right. Now, suddenly I mean you can see that how we have induce a
Lorentz force Lorentz force is basically a force which will try to oppose the which will
try to oppose the basic I mean acquisition of the charges on the 2 extreme side of the
plate.
We have seen that we have told several times that you see that when we apply some
voltages on the opposite surface of the plates a voltage will be developed. Why this
voltage was developed? This voltage is developed because there is a acquisition of the
charges. Now, Lorentz force is basically when this Hall Effect and the Lorentz force will
balance each other we will get a steady output voltage H right. Whatever the small it
time might be 10 to the power minus 14 seconds, but within the short time it there will be
balance will be achieved is it clear? That is I am saying that the when I applied a in a in a
hall effect transducers when I applied a voltage I mean which will read will give me a
steady current I and it is and on the steady magnetic field B.
What will happen that there will be the redistributions of the charge of the charges or the
electrons inside the hall to be 2 sides of the plates? And the Lorentz force will balance
that thing and due to this acquisition of the electrons I am getting a a definite non-zero
voltage which is we are calling it a hall voltage right. Let us look at the problem again
this is our problem. So, given hall coefficient for the germanium crystal is minus 8 into
10 to the minus 3 volt meter ampere Weber per meter square.
Solution EH is equal to KH is BI by t we have seen. So, B equal to KH t by KH by KH

by into I. So, this you just put all these value of B. So, B equal to 0.012 Weber per meter
square.
So, now when the hall field a Lorentz force must be I mean balance. So, I can write it is
the moment of the velocity B is the velocity of the electrons inside the hall sensors. So, e
EH by b, so V equal to EH by B by small b let me go yes. So, which will give you the
speed of the electron as 9 7 2; these are all the dimensions of the hall plates, B is a
dimension of the hall plates 9 7 2 2 meter per second right. With this I come to the end of
the lesson 27 of industrial instrumentation.
Prof. A. Barua
Lecture - 28
Optoelectronic Sensor – I
Welcome to the lesson 28 of industrial instrumentation, in this particular lessons we will

study optoelectronic sensor. In fact, the optoelectronic sensors, I have divided in the 2
lessons in the electronic sensor 1 and optoelectronic sensor 2. In the optoelectronic
sensor 1, we will look at this various types of photo resistors then photodiodes photo
transistors, we will cover very brief. And all this things also, some brief introduction to
fiber optics, because as you know there are as a day’s going on. So, there are lot of fiber
optics senses coming up and we should discuss that details in the lesson 29. But in basic
introduction to the fiber optic sensor also will be fibre optics techniques transmission
techniques. And all this things will be covered in this particular lesson that is the second
I mean, later part of this the lesson 28.
Let us look at optoelectronic sensor 1.
Contents are photo conductivity, we will consider here photo conductivity. Based on
that, in the photo resistors or light depended resistors has been developed and which is
used is extensively in instrumentation. So, that we will cover in the photo basic principle
of the photo conductivity, then on based on this what are the different photo resistors?
We have developed I mean, engineers and scientist developed that we will discuss right.
Then we have photo resistor, photo diode then and applications. We will look at some of
the application typical application; that means, how we can utilize make some
measurements? Basic measurements; there are various applications of this photo diodes
per. So, far as the instrumentation is concerns in context to instrumentation, we will take
2 examples and see how this photo diodes are utilized to make that above measurements.
Also we will cover the introduction to fibre optics we will discuss the fibre optics in
details and fibre text, fibre optics sensors in lesson 29 for basic principles of the fibre
optics will be covered in the lesson 28.
Photo conductivity; what is photo conductivity? The conductivity of a material is

proportional to the concentration of charge carriers present right. You know conductivity
s basically depends on the charge carriers in the case of metals. We have some fixed
amount of charge carriers, which is thermally generators? In the case of I mean, semi
conductor devices, these can be varied depending on incident light. So, which will
change the resistance; obviously, the photo current also will change. The current density
in amperes per meter square will be expressed as follows, I equal to n mu n plus p mu p e
multiplied by E.
Where n is the magnitude of free electron concentration, p is the magnitude of hole

concentration, sigma is the conductivity, mu n is the electron mobility in centimeter
square of an by volt second and mu P is a hole mobility in centimeter square per volt
second. Now, sigma conductivity will be given by n mu n this is mu n this is mu n p plus
mu p multiplied by e, where e is the electronic charge E is electric field.
For intrinsic semiconductor, that is semi conductor without, any impurity. We call it
intrinsic semi conductor in the case of; that means, with without any doping. Once you
dope, what will happen that the number of hole electrons will change. Now, if you dope
with the p type impurity number of hole will increase, if you dope with the n type
impurity the number of electron will increase right. So, for intrinsic semiconductor n
equal to p equal to ni which is called a intrinsic concentration, when the semi conductor
is doped we call it extrinsic semiconductor right. In that case np cannot be same, because
it depends on the, what types of dopings? In may be in the case of I mean, travel and
impurity. What will happen? That you’ll find that the hole will increase number of hole
concentration will increase so on and so forth.
Radiant energy supplied to the semiconductor causes, covalent bonds to be broken and
the hole electron pairs in excess of these generated thermally, are created excuse me.
This increased current carriers, decrease the resistance of the material, whenever the light
falls is increased there is increase of charge carriers. In fact, what will happen? Where is
as you know there is in a metal ((Refer Time: 05:56)) forbidden energy. So, some metal I
mean, some electrons will cross that forbidden energy gap and come to the conduction
level. So, which will contribute to the change of concentration of the charge carriers?
Hence such a device is called a photo resistor or photo conductor, clear? For a light
intensity change of 100 foot candle the resistance of photo conductor may change by
several kilo ohm, right. The minimum energy of photon required for intrinsic excitation
is the forbidden gap energy. Now, electron volt, which is in electron volt of the
semiconductor material, if you remember when it appears to be like this one.
I have a conduction band. So, when the light falls, which is comes here, some of the
electron; this is the forbidden gap energy EG. So, these energy these amount of energy
must stored so that the electron will come to the conduction band. This is the conduction
band and this is our forbidden gap energy, right.

Now, threshold wavelength of incident light; that means, the light which will be
sufficient to generate the excess carriers in excess to the thermal energy is given by a 1.2
4 by EG in microns or micrometers. For Silicon EG is equal to 1.1 electron volt for
lambda c equal to 1.13 micron and for germanium EG equal to 7.2 electron volt points 7
2 electron volt. Which will give you lambda c are the of that, because this is necessary,
because you have to first note the what the incidence light will be make this incidence
frequent of the incidence light, which will you must I mean, must fall on a
semiconductor devices photoelectric devices. So, that the electron excess electrons or
excess carriers will be created right.
Photo current; now, steady state current is given by I equal to e multiplied by Pr into tau
upon T subscript small t. What are this regions i is the steady state current in a photo
resistors or photo conductor where Pr is a rate? At which the carriers are produced by
light rate? At which the carriers are produced by light, because if the light falls strong
light falls more numbers of carriers will be generated which is in excess of the thermally
generated carriers. Tau is the average life time of newly created carriers, because you see
that, whenever the hole electron pair is generated, there is the chance of recombination.
So, this the average light time, because once it is recombined that current cannot to be
detected at the output. T subscript small t is the average transit time for carriers to reach
the ohmic contact. Because any semi conducted devices, I must collect the current at the
outside. So, that it must reach that carriers, must reach the ohmic conduct. So, that it will
pass through, some ammeters and some deductible, current will be observed and e is a
electronic charge e is an electronic charger.
Now, one of the commercial photo conductivity cell is cadmium sulphide, one of the
most widely used photo conductivity cell or photo resistor or LDR. Whatever the name,
you call is light depended resistors or photo resistors or photo conductivity cell. It
basically same; that means, ((Refer Time: 10:08)) made of cadmium sulphide with the
pentavalent antimony impurity. Impurity must be there, otherwise in cannot be
pentavalent impurity are basically, the charge carriers will be here the electron right
since it is pentavalent. The resistance is over 2 mega ohm in dark and 10 ohm in the
bright light right. It looks like this, it looks like this, if I take a.
So, it is looks like this, I mean if you look at the encapsulation, I will draw a little bigger
2 leads are coming up and there by transparent plastic cover on the top. Unlike the other
semi conducted devices and we can you can see this photo resistors inside right. So, light
should fall on this 1, if you bring it to the sun light, you will find immediately, the
resistance. We measure the resistance across this 2 immediately, it will reduce right that
resistance, you can make several circuits by which I can. Suppose if I have a circuit like,
this 1 I have a I am giving a voltage of suppose 5 volt and this is my photo resistors, this
is my photo resistors. So, light is falling on this, so what will happen? So, the charge
current will change. So, I can do in other way, you can we can take I mean, I can do it in
other way; I can take a new slide there is no problem. So, it will go so let me take the
pencil, let me take a new. So, it looks like this.
I can take a right. So, I am giving a 5 volt here, and this is my photo resistors light is
falling on this looks likes as I told you earlier its capsulations. So, it is looks like this 2
leads are coming up, you can see the photo resistor here. Lights are falling on this right
and this voltage; obviously, what will and if the light falls, you will find that this voltage
will change right. So, if I measure, because this resistance will change. So, this voltage
Vo across this will also change.
So, this Vo can be utilized for making so many sensors or switches also right I can use it
as a switch also these photo resistors, clear? Because this voltage will change as a light
falls on this in dark. It is very large resistance around 2 mega ohm in a bright light, it is
very less 20 ohm clear. In fact, the dark it should be very, very, very high it should be
infinite, but there is always a dark current, because there will be always some electron
hole electron pair in the in the which is thermally generated. So, we there will be a some
current right. So, resistance is over 2 mega ohm in dark and 10 ohm 10 to 20 ohm in
bright light cadmium sulphide has a high dissipation capability excellent sensitivity in
the visible spectrum.
Power dissipation is more than 300 milli watt and a selenium cell. Selenium is also used
as a photo resistor; selenium cell is sensitive throughout the visible and particularly,
towards the blue end of the spectrum right.
Now, semiconductor photodiode now, we talk about only the photo resistors so; that
means, I have some means, that the light dependence resistors. We are calling also it, I
mean, also as a light dependence resistors, but more applications in instrumentation. You
will find is that is in the semi conductor photodiode photodiode and phototransistor is
almost same in case of phototransistor. I will get some amplification, but the time
constraint is large. Whereas, in the case of photodiode current is small, but the time
constraint is small. If a reverse biased p n junction is illuminated the current varies
almost linearly, with the light flux the effect is utilized in the semiconductor photodiode
right this effect is utilized in the semi conductor photo diode. The reverse saturation
current Io in pn diode is proportional to the concentrations pno and npo of minority,
carriers in the n and p region respectively
If you have a reverse biased photodiode then in that case ((Refer Time: 15:26)) there is
no minority carriers. So, the only the sorry majority, carriers will not play the current,
because only the minority carriers will play key role to have a current because in all semi
conduct. It does not matters, if it is p type semi conductors the hole is the majority
carriers and the electronic minority carriers. In n type semi conductor electron is the
majority carriers and the hole is the minority carriers. So, if it is reverse biased then what
will happen that the reverse saturation current Io in Pn diode is proportional to the
concentrations of pno and npo minority carriers and n and p type region. You see the
pno, we are calling the minority carriers in n type region and npo is a electron, which is
as semantically as p region respectively.
The volt ampere characteristic will be given by I equal to Is plus Io 1 minus 1 minus e
exponential to the power V by eta Vt. Please note this is eta, this is eta right, this is not n
this is eta like this 1. Where Is the short circuit current and it will be proportional to the
light intensity, current under the large reverse bias V is positive for forward bias and V is
negative for the reverse bias right. Now, for the ((Refer Time: 16:55)) bias is large then
obviously, what will happen? This part will be very, very small it is not it. Is the reverse
bias is very large the reverse bias is large. So, what will happen? If is reverse bias is
large then what will happen?
You will find you see that this will be negative. So, this is V is very large then what will
happens? So, this will be almost negligible. And only it will be I equal to Is plus Io
where Is is the short circuit current and it will be proportional to the light intensity
current. And large reverse bias V is positive for forward bias and negative for reverse
bias. The parameters eta I should write here, eta is unity for germanium and 2 for silicon,
this eta this not n please note eta the parameter eta is unity for germanium and 2 for
silicon and Vt is the volt equivalent of temperatures right.
This is a equivalent circuit of a reverse bias photo diode. You see these, we have one
capacitance, we have a current generator here C is equal to 10 pico farad P is I mean R is
quite large 50 mega ohm and there is a small resistance r equal to only 100 ohm. So, is
that saturation current is equal to K into L, what is K?
We will see the symbol L represents the light, light flux in lumens you see the Is equal to
KL the symbol L represents the light flux in lumens. And k is the proportionality,
constant in a range of 10 to 50 milli ampere per lumen, clear and the reverse saturation
current Io doubles for every 10 degree centigrade reverse saturation current. We have
seen also this 1 equal to 4 reverse saturation current Io doubles for every 10 degree
centigrade and its typical sensitivity is 10 milli ampere per lumen, right. The lumen
actually, is defined as a 1 watt of light at lambda equal to 0.55 micron can produce 680
lumen. These are standard for lumens that mean a 1 watt of light at lambda equal to 0.55
micron can produce 680 lumens.
Now, you see this photodiode characteristics, we have drawn here. This diode current in
micro ampere and this the reverse voltage, as we increase the reverse voltage, we will
find that, there is saturation. And this incident light, which is given in foot candle right,
this incident light in foot candle 3000 foot candle, 2000 foot candle, 1300 foot candle
335 foot candle. This one and there is the dark current even, if you do not I mean, throw
any light on the photo. I mean, diode this basically, you are falling, you are sending a
light to the junctions of the photodiode please remember right pn junctions. So, in the pn
junctions, you are sending you are throwing the light. So, these are dark right this reverse
bias photodiode characteristics reverse voltage V. And this diode current as you can see
that, if we increase the light; obviously, the diode current also in getting increased.
One lumen per square feet is equal to 1 foot candle 1lumen per square feet is called the 1
foot candle n-p-n photo transistor provide electrical amplification, but has a larger time
constant. So, you actually, you will find at this photo diode is basically, a diode, but you
will find that it is a it is put in a plastic in capitulation. So, that the light can fall on the pn
junction pn junction it is more sensitive, than the photo diode; otherwise basic principle
is same as a photodiode. If it is more sensitive large current, you will get, because you
will get some amplifications resin inherent to any transistor characteristics. Suppose if
the base ammeter, we I mean throw some lights; obviously, at the collectors. We will get
a large current with amplifications, that is the only advantage, but time constant is large.
Some typical applications of this photodiodes let us look at some typical applications.
Now, this is a scheme, which we have shown that a detail evolution counter using
photodiode, how a digital evolution counters? We can make using the photodiode, you
see here. So, I have a light source, light source is coming here, this a disk when this light
source light emitting diode. Then this hole on the disk this circular disk and the photo
detector, which is photodiode comes in a 1 line, then only some light are found will fall
on this line light will fall on this photo detector otherwise not. So, you see if it rotates, if
this disk rotate as a there is a shaft and the disk rotates, for each rotations. I will get a
pulse here, at the output of the photo detector now, this type of pulse as you know this
type of pulse will very irregular shape, right.
So, this pulse will be very irregular shape, this pulse will be very irregular shape. So, you
must sent through, Schmitt trigger through, which is called a pulse shaper to give a
regular shape pulse and then to a counter. So, as the speed of this disk increases number
of count number of count per second or per minute will also increase right. So, I can
make a direct rpm meters or a revolution per minute meter by these types of scheme. We
have some other electronic circuitry, but this is a some applications of the photo diodes.
There are numerous applications, with that is not much concerns, with the
instrumentation or measurements, that is I am not discussing automatic opening of the
doors. And all these things you know automatic switching on the street lights in many
countries you see that there is no person to switch on the street lights. So, what they do?
There is a photo diodes sensitive switches they have. So, what will happen? That when
the light I mean, the sunlight goes down.
So, automatically, the, this photo diode will be activated, and is will turn on the switches
of the all the street lights. So, there are some and that above many applications are there
especially, in a in countries like a Europe. You will find when the tramps and trains are
going usually, to save the power; they do not turn on the lights inside the bus or tramps.
So, whenever the tramps or bus are going inside a tunnels so; obviously, the light is less
all the lights, inside the tramps or bus should switch on. So, those are photo sensitive, but
there is not much of the instrumentation there some measurement techniques. We are
discussing; that means, some revolution counters, we are using now, we have developed
using the photo diode, let us look at how it works.
So, you see this is our circuit schematic of the revolution counter. Here you see that we
have a 7490 decade counter most significant bit of the 7490 is going as a clock. Here is a
clock which is coming from the n gate again; the most significant bit of the 7490 is going
to the clock of the next 7490, clear? So, is the decade counter, it will start from 0 is the
decade counter, it will start from 0. And it come up to 9 then again, comes to the 0. So,
its 10 different states of modulated counter, it will start from 0 count up to 9 then again,
come to 0 right like that. Now, how it works? You see this is 7475 is basically, D latch 4
bit D latch flip flop. You know the D latch is basically, a device which is basically, a
master slip flip flop. But it will latch to the previous state, until unless new clock comes
right.
Whenever a clock cycle ((Refer Time: 25:06)) whatever the whatever, you have at the.
So, sometimes we call the data latch flip flop, that is it is we call D flip flop or data latch
flip flop. Data latch flip flop means, whatever the previous signals we have. So, it will go
it will stick to that until unless a new signals comes, new state comes, either it may be 0
or one it will go back to that is it will go to that is particular state. When the when the
now, is basically, when a clock pulse comes at the tailing edge of a clock pulse in the
case of master shift. Obviously, it should activate at the tailing edge of the clock pulse.
So, it will go to the state, if it is one is at the input of the D flip flop this basically, JK flip
flop. You know this basically, JK flip flop the data which I want to transmit that, data is
directly, coming to the j input and data and the and the k input is whatever the j.
Whatever the data input is coming in a j inputs, we are passing through a inverter and
coming to the k input right.
So, automatically, what will happen? See if it is 1 is coming, so in the j input is getting 1
and k input is getting 0 and if the 1 is 0 is coming, j input is getting 0 and the k input is
getting 1 see it is important. So, what will happen? Under the clock pulse if it is a master
((Refer Time: 26:20)) to the trailing edge clock it will trailing edge clock activated. So,
what will happen? That that inter change will occur at the trailing edge and you whatever
the data input you have either 1 or 0 it will stick to that right. Now, 7475 is such as data
flip flop with the 4 bit data latch flip flop seven four 7475 is the 4 bit data latch flip flop
it is a 4 bit. So, what will happen? You see it is very interesting, you see I have in the
AND gate inputs. We have a clock pulse, high frequency clock pulse right and you see
here.
Now, this clock pulse is coming from, where it is coming from? That photodiode,
depending on the number of speed depending on the speed of the disk, number of clock
pulse per minute will also increase right. So, what we are doing here? You see that in 1
input of the AND gates, I am giving this clock pulse and other inputs; I am giving a 1 a
pulse, where the mark is 1 minute. Then it will have something and then again it will
come. So, it is only 1 minute width it is repeating pulse obviously. Now, what will
happen? You see, so this whenever this is high. So, it will activated and the the counter
will start to count from 0 right.
Now, you see what will happen? During that time this is inverted here. So, 7475 will not
be activated. So, counter is going on counting now, what will happen? At the trailing
edge of this clock pulse, when the 1 minute is over. So, this pulse will go down and no
more it will it will no more, this output from the photodiode can enter the 7490 7490 will
stop counting during, that time when this is low, this will be activated, because it is
inverted what does it mean? So, whatever they have counted. So, long that 2 7475, it will
it will come to the ((Refer Time: 28:29)) coder and then to the display unit clear?
We have a 7 segment display. So, we will introduce display unit now, during that time
you see what will happen? This duration I can make very small, I can make 1 minute
then 1 second and again 1 minute at time pass I can make it clear. So, during that again,
what will happen? Whenever this pulse is off, next clock this, whatever the data, we have
at the end of the 1 minute. That will hold for the next until unless, next clock pulse
comes, that is a reason it is called data latch flip flop. So, what will happen again? Again
it will start to count. So, for 1 minute, so after this when this falls it will stop counting
same for 9 0 will no more count anything. So, one this falls, this will be activated,
because this will be 1. So, whatever that data, we have it is output of the 7490 in a 4 bit,
that will come to the 7475. So, previously it will latch to the I mean previous values.
So, then it will come to the display and it will ((Refer Time: 29:35)) coder right, this way
it will work. So, this you can see this 1 application, even though I mean, this photodiodes
I mean, as a very small part. But the, until unless you have photodiode this system will
not work you see. So, I will light source, when the light source which is rotating in a very
high speed and since photodiode has a very, very small time constant. So, small amount
of whenever, even it is speeds at a very high speed it does not matter, it will detect each
and every poles. Whenever, this photo detector this hole and the light source comes in
one line. So, it will make 1 pulse and this pulse is to be shaped through, Schmitt trigger
then it is coming to the counter clear?
This is our scheme, this very interesting scheme that is will be activated for 1 minute will
count for. So, whatever the reading, we are getting here, it will be directly in revolutions
per minute, is not it? Because we have made the width of 1 minute, mark to space mark
will be for 1 minutes space might be for 1 second it does not matter. So, during 1 second,
what will happen? Again until unless this 1 minutes comes at the end of the 1 minutes
these data whatever the data there that will be displaced automatically, that will remain
in the same whole position clear?
Now, limitations it has limitations, lack of 0 references and it cannot sense the directions,
of the rotations. Whether it is I mean disk is rotating in the clockwise, directions or anti
clock wise directions. It cannot sense for that reasons, I have to we have different we
have seen that, we needs some phases, it modulation to have a that type of sensing.
Measurement of steel strip in hot rolling mill, this another I mean, scheme by how the
photo diodes can be utilized to make a, because in the hot rolling mill, it is very difficult
it is. So, hot temperature is, so high it is very difficult to measure the thickness of the,
because thickness measurement of the rolling sheet is also very important accordingly, I
can use the feedback poles, so that the we can put more pressure on the roles. So, that it
should be thinner or the remove the pressure.
So, that it should come to the proper thickness right. So, we in context of that, we will
see that the photodiodes, because a hot rolling mill in the hot surface will work as a
photo diode photo, I mean light emitting diode. Now, the question comes, we need little
bit of mathematics; that means that auto correlation and the cross correlation functions
that will be discussed. Now, you see the correlation function the auto correlation function
of a signal is closely, related to the power density spectrum of the signal. The auto
correlation function of waveform is defined as Rxx limit T tends to infinity 1 upon T t to
0 xt xt minus tau into dt.
The waveform xt is multiplied by a delayed version of itself, xt minus tau and the
product is averaged over T seconds, the waveform xt is multiplied by delayed versions of
itself xt minus tau. The produce is average over T seconds, if it is multiplied by the same
signal delayed by some time, we call auto correlation function. Now, let us look at the
cross correlation, the cross correlation function, between the 2 identical waveforms xt
and yt are given by Rxy t xt minus tau yt average. Which is limit t tends to infinity 1
upon T t to 0 xt minus tau yt in to dt right. The cross correlation function is used to
measure the non contact velocity measurement of rolled metal sheets, non ((Refer Time:
32:23)) type of techniques. So, non contact with out of the velocity of the rolling sheet,
right. So, the velocity is most importance thickness is there, but the velocity is most
important, because it why it is important? Because in the hot rolling mill, if it is not I
mean, if you do not put in the quail form when it is steel hot that will create problem
right. So, that reason…
This is a scheme for measurements of velocity of hot rolling strip, you can see here. So,
we have a hot rolling strip here you can see ((Refer Time: 33:57)) is coming. So, it is
coming through a roll and the thickness is getting reduced like this 1. This is a thickness
and this velocity should be as, I told you this velocity is to be maintained, because until
and unless it is hot you cannot put in a quail form right. So, that is the reason the velocity
is important, we have a roller here these two’s are called the rollers. Now, see it is since
it is hot, so it will work as a source light source, we have a collimator lens these 2 are the
collimator lens.
So, light will be come and it will focus and will form the photocell. We have the 2 photo
cells here, 1 photocell is here, 1 photocell there. We have a variable delay here; we are
assuming these 2 signals are I mean, same type of signals. But that it is not the exactly,
same signals it is coming from the 2 different sources. So, we have to make the
correlation functions of this 2 signals, this is the correlators, which is make the cross
correlation of the 2 signals, which is coming yt and xt and the cross correlation
coefficients we have getting at here.
The measurement of velocity of rolled steel of sheet material is difficult, particularly,

when it is red hot. The basic scheme of the correlation measurement is shown in figure 4
which have already shown. The fluctuations of intensity of the light reflected from the
metal surface. Are transformed in to electrical signals by means of 2 photocells mounted
at a distance d apart.
Let us look at yes the cross correlation coefficient is a maximum, when the variable
delay equals the time equals the time for a point on the strip to travel, between the
measurement points. There are 2 measurement points, so the cross correlation coefficient
is maximum, when the variable delay equals the time for the point on the strip to travel
the between the measurement point. So, let us look at, you see this is the measurement
point, when this variable delay equals the time for time to travel the same points of these
to these, I will get that peak output of the cross correlation coefficient. So, let us look at
that the cross correlation coefficient is maximum, when the variable delay equals the
time for a point on the strip to travel between the measurement points. Now, the sheet
velocity will be small d divided by tau d, because what will do?
Actually, you see here actually, I will vary the delay. So, whenever I will vary the delay.
Whenever I will get the peak, I will measure this tau d right. So, tau d, so the d divided
by tau d will give the give us the, because this is the cross correlation coefficient is the
correlation co efficient. So, d divided by tau d will give me the velocity of the roll steel.
Now, let us come to the, I mean this with this I come to the end of this 1; that means, that
the optical sense. I mean, the photo electric and photo conductivity sensor photo diodes
let us, come to the introduction to the fibre optic transmission systems. Because we in the
lesson 29 will covered the fiber optic sensor, because it otherwise it will very huge. So, I
will cover some of this in this lesson 28.
Fibre optics is the technology of using light to transmit information, light has a number
of advantage over electricity, as a medium for transmitting information. It is immune to
corruption by neighbouring electromagnetic fields 2 light source, can go parallely 2
cables can go parallel. Whether there it is very difficult to put cables 2 parallely, because
there is a curve, there may chance of cross talk, right. It is immune to the corruption by
the neighbouring electromagnetic fields attenuation in the all electrical circuit. There is
attenuations here, in this case attenuation over a given transmitted distance is much less
and it is also intrinsically, safe and since it is light.
So, there is no ((Refer Time: 38:08)) as such. Fibre optic technology is used in 2 major
ways, first the fibre optic cable itself can be used directly as a sensor, in which the
variable being measured causes, some measurable change in the characteristics of the
light transmitted by the cable. Secondly, the fibre optic cable can be used as a
transmission medium for data.
The later use of the fibre optic cables for transmitting information can be further divided
into 3 separate areas of application. First relatively short fibre optic cables are used as
part of the various instruments to transmit light from the conventional sensors to a more
conventional, convenient location for processing often in situations where space is very
short at the point of measurement.
Secondly longer fibre optic cables are used to connect remote instruments to controller in
instrumentation networks. Longer fibre optic cables are used to connect the remote
instruments to controllers in instrumentation networks, because ultimately, that signal is
to be transmitted to the controllers right. So, transmission electrically, so the
transmission the electrical transmission there is lot of problems. So, we can switch to the
fibre optic senses. Thirdly longer links are still used for data transmission systems in
telephones and computers networks. We are not much interested, we will not neither we
will cover that in this server in more details right. That is in the communication part
these three application classes have different requirements and tend to use different types
of fibre optic cables.
Principle of fibre optics; the central part of a fibre optic system is a light transmitting
cable containing at least 1, but more often a bundle of glass or plastic fibres. This is
terminated at each end by a transducer as shown in figure 5 at the input end. The
transducer converts the signal from the electrical from into light, right.
This is you see, I have a transducers which is electrical signal is coming from the left
hand side. It is coming the light, light is getting launched through this fiber optic cable
and basically, we will see this fiber optic cables we cannot light can transmit. It can go in
a curved path, we know from our childhood light never goes in a curved path. But you
will see using the total internal reflections in the fiber optic cables, light can travel in the
curved path also right. So, at the transducers we will get light again, suppose this is our
LED, LED light emitting diodes. It is launched though this one and these is a, we have a
photodiode or pi n diode. So, I will get the electrical outputs, so the electrical to
electrical, I will get at the 2 ends whereas in the medium, it is only light are taking them
our data or information.
At the output end the transducer converts, the transmitted light back into electrical form,
suitable for use by data recording manipulation and display systems. So, we have a data
recordings, we have it conversion, whatever you like you can do once it is in electrical
form. These 2 transducers are often known as the transmitter and receiver respectively,
what is the transmitter? That we have seen which one is transmitter, this is our transmit
this also transducers, But we are calling it, transmitter and these are we are calling it,
receiver since it is receiving signals.
We are transmitting the signals through this LED, where the, that signal the photo diode
or pi diode we are calling it receiver. In that case sorry, these two transducers are often
known as the transmitter. And receiver respectively, and the arrangement described
above is somewhat, modified when the cable is used directly as a sensor. You will find in
some cases, we will use a cable as cables as a sensor, we will modulate the light which is
going through this 1. So, it will be slight different in that above application.
In that case, the light injected into the cables comes directly, from a light source and does
not originate does not originate as an electrical signal. However, the same mechanisms
described, later have to be used to get the light in to the cable.
Fibre optic cable fibre optic cable consists, of an inner cylinder core surrounded by an
outer cylindrical cladding as shown in figure 6. What is that figure 6? Let us look at a
little far it went.
So, we have a inner core, we have outer cladding then jacket and all these things the
refractive index of this one. Then only you will get a total internal reflection, please note
refractive index of the this of 1 the inner core will be higher than the refractive index of
the outer cladding clear, this is always; otherwise you would not get the total internal
reflection. Fiber optic cable consists of the inner cylindrical core surrounded by an outer
cylinder, cladding as shown in figures 6, which I have shown the refractive index of the
inner material is greater than, that of the outer material and the relationship, between the
2 refractive indices affect the transmission characteristics of light along the cable.
The amount of attenuations of light, as it travels along the cable varies with the wave
length of the light transmitted. This is very important, we have seen that the, we have
find to out optimum, wavelength of light optimum colour of the light which will give us,
the minimum attenuation that is important otherwise, you have to use a repeater though. I
am saying that the loss is less, but there is a loss usually, which is describing ((Refer
Time: 44:08)) per kilometer something like that. So, what will happen? As it is less
repeated otherwise again, I have to convert that, signal to at what at the repeated station?
I have to convert that signal in the current electrical domain, again through led to
transmit that signal.
This characteristics is very non-linear and a graph of attenuation, against the wavelength
shows a number of peaks and troughs the position of these peaks. And troughs varies
according to the material used for the fibres, it should be noted that the fibre
manufactures, rarely mention these non-linear attenuation characteristics. And quote the
value of attenuation, which occurs at the most favorable wavelength.
The 2 forms of cable exist, known as monomode and multimode, we have a monomode
fibre. Monomode has some I mean, monomode has a I mean, launching the light in the
monomode fibre is very difficult. Because it is, we have to use a very directional light in
the case of laser whereas, in the case of multimode fibre. I can use ordinary light led, that
is very I mean, that is saves a lots of cost. Monomode cables have a smaller diameter
core typically, 6 micron, whereas the multimode cables have much larger diameter.
Typically, between 15 micron and 200 micron both glass and the plastic in different
combinations are used in various forms of cables. We can use glass or plastic.
One option is to use different types of glass fibres for both the core and the cladding.
You can use both the glass for the, but; obviously, of different RI refractive, index for the
core and cladding is second the cheaper option is to have a glass fibre core. Which has a
higher refractive index and the plastic cladding, which has a less RI value refractive
index, value this has a additional advantage of being less brittle than the all glass version.
So; obviously, if we use the plastic it will have, because ultimately, you see the
mechanical strength is also very important. If it is made of full glass, it is very difficult to
hold, it I mean it is very difficult to its very prone to the breaking. So, if we use a plastic
cladding; obviously, your chances of I mean survival is better.
Finally, all plastic cables also exist also 2 types of plastics where is the different
refractive indices? But they have a larger losses. So, that is the reason all plastic cables is
not very popular till now. This already, I have shown you inner cross section of the fibre
optic cable, inner core this might be all glass this might be glass. This is can be glass
whereas, this can be glass and this can be a plastic. Obviously, the refractive index this 1
whether it is glass or plastic it does not matter; obviously, it should be higher than, this
one or it can be using all plastic, but all plastic this attenuation is large. So, that is
another problem that all plastic fibre optics cable.
This is the cheapest form of all, but it has a disadvantage of having high attenuation
characteristics making it, unsuitable for the transmission of light over a medium to larger
distances for a shorter distance. Suppose I am using as a sensor pressure, sensor load
sensor that type of case not a problem, but if is used for a transmission only. So, it is a
not very suitable for using the plastic, all plastic fibre optic cables. Protection is
normally, given to the cable by enclosing the same time of insulating and armouring
materials. So, they have a outside plastic, I mean we have a like we have the electrical
cable that type of protection. We will give protection is normally, given to then cable by
enclosing. It in the same types of insulating and armouring the material, that are used for
copper cables as in the case of copper cables, we have a several, because this will give
you mechanical strength also.
It protects the cable against various hostile operating environments also against,
mechanical damage. So, mechanical damage, so that it will give you this, protections will
give you that type of things, when suitably protected the fibre optic cables can even with
stand the flames, if you engulfed by the flames fibre optic cables. If we have observed
that, the fibre optic cables can survive it is impossible, the cases of the copper cable and
all those things.

Fibre optic transmitter, let us come the light emitting diode LED is commonly, used as a
the transmitter which converts, usually light is used for multimode fibre at the
monomode fibre. We have to use laser please note because the light should be very well
directional. In fact, you will find the launching the light in a fibre optic cable is a very
very difficult task we have a lot of connectors and all those thing. So, we if we some
other or some way or the other that is simplified that that gives you lot of advantage lot
of flexibility while using the fibre optic cable, which converts an electrical signal into
light and transmits it into the cable the LED is particularly suitable for this task at it has
an approximately linear relationship between the input current and the light output. So,
the input current and light output is very much I mean this linear relationship is there we
have seen that the diode has the I mean that type of advantages.
The type of LED chosen must closely match the attenuation characteristics of the light
path through the cable and the spectral response of the receiving transducers. Because
receiving transducers also should have same characteristics or the, cannot convert that
light signal to the electrical signal that is one to one characteristics should be there. That
means, as I told you linear characteristic amount of light falls and the current it is a it is a
linearly variable. So, that should be a at the receiver end amount of current amount of
light falls. That is at the case of receiver end in the case of transmitting end amount of
current should be directly proportional to the amount of light which is going to launch to
the fibre optic cable. An important characteristic of the transmitter is the proportion of its
power, which is coupled in to the fibre optic cable. This is more important than the
absolute output power.
This proportion is maximised by making the purpose purposefully designed LED light
emitting diode transmitters which have a spherical lens incorporated. So, that it is like I
mean if you put a a source at the focal point of a lens then what will happen the light
which is coming out of the lens will be the parallel. So, that is very much in necessary.
So, this produces approximately parallel beam you see whatever I said this produces a
what is the this proposition is maximised by making the purpose designed LED
transmitters which have a spherical lens incorporated into the chip during manufacture.
This produces an approximately parallel beam of light into the cable with a typical
diameter of 400 micron right? The proportion of light entering the fibre optic cable is
also governed by the quality of the end face of the cable and the way it is bounded
bounded in the transmitter. So, this is a very difficult task as I told you it should be very
much well polished if there is a little misalignment. So, the light cannot be launched
through the fibre optic cable this intensity of light will be sufficient. Otherwise you need
a repeated very I mean after a short distance in the case of transmission.
Instrumentation that is not very much necessary. A good end face can be produced by
either polishing or cleaving in face end face of the fibre should be very much polished.
Polishing involves grinding the fibre end down with a progressively finer polishing
compounds until a surface of the required quality is obtained. So, the end will be
perfectly polished is this means this end will be perfectly polished. Because you see I
have I have a cable I have a cable and we have a its I have a cable and I have a
transmitter here.
So, whenever it is connected to that cable this transmitter this light cannot be launched if
this end surface of the cable is very rough. So, that should be very much polished. So, we
are using grinding technique and cleaving technique. So, through that it will be very
much polished and light can be easily transmitted through this one this light source as I
told you it is a spherical lens. So, that the light which is coming off from the suppose the
source is here it is falling on the spherical lens and which is from the lens when it is
going out it is going all parallel lines of lines of lights . So, that we will launch through
the fibre optic cables attachment to the transmitter is then normally achieved by the
gluing.
The proportion of the light transmitted into the cable is also dependent on the proper
alignment of the transmitter with the center of the cable right. The effect of the
misalignment depends on the relative diameter of the cable. And figure 7 shows the
effect on the proportion of power transmitted into the cable for the cases of cable
diameter greater than the beam diameter.
You see this shows that the some degree of misalignment can be tolerated except where
the beam and the cable diameters are equal. The cost of producing exact alignment of the
transmitter and the cable is very high as it requires the LED to be exactly aligned in its
housing, and the fibre be exactly aligned in its connector and the housing to be exactly
aligned with the connector.
Therefore the great cost savings can be achieve wherever some misalignment can be
tolerated. That is the reason multimode fibre is always preferred over the monomode
fibre is little misalignment will not affect much.

You see this is that we talked about effect of transmission here see the fiber diameter is
the more than the beam diameter. We can see that how much tolerances misalignments
we can I mean aligned that is multimode fibre has a always the more diameter is more
right. So, that is the advantaging whereas, the fibre diameters is equal to beam diameter
and fibre diameter is less than beam diameters whenever this is less it is more we have a
miss a small amount of misalignment. So, only will be tolerated.
Fibre optic receivers device used to convert the signal back from light to the electrical
form is usually either a PIN diode or a phototransistors right. Photo PIN diode is more
popular use. Phototransistors have good sensitivity, but only low band width the
bandwidth of the photo transistors is very small.
On the other hand, PIN diodes have much higher bandwidth, but has a lower sensitivity
we need a amplification obvious that is not a problems. So, we can have a amplifier. In
both high bandwidth and high sensitivity are required then the special avalanche
photodiodes are used. But at a severe cost the same considerations about losses at the
interface between the cable and receiver apply as for the transmitter and both polishing
and the cleaving are used to prepare the fiber ends.

The output voltages from the receivers are very small and the amplification is always
necessary, right. That is I told you I mean I can I can use a PIN diode, if I use a PIN
diode I. need further amplifications, but that is a advantage which we were getting,
because it has a larger bandwidth right. The system is very prone to noise corruption as I
even though I am telling this is not corrupted by any interfering noise or any electrical
noise. But the noise will come at the transmitting and the receiving end, because there is
a position there is a where it is signals optical signals compact to electrical signals again.
So, noises sources have introduce then even though if a fibre optic cables are going on
and over there are many crises cross of the other electrical cable. So, the any spikes in
the electrical cable does not I mean effect the information which are getting transmitted
to the fibre optic cable. System is very prone to noise corruption at this point; however,
the development of receivers which incorporate an amplifier are finding great success in
reducing the scale of this noise problem.
Transmission characteristics: Let us look at the monomode step index fibre have very
simple transmission characteristics, because the core has a very small diameter the light
can only travel in a straight line down it? This types of fibre can be sustain only one
mode of propagation and equivalents requires coherent laser source that is a problem,
because light will be transmitted in straight line.
On the other hand multimode step index fibers have quite complicated transmission
characteristics, because of the relatively large diameter of the core. Many modes can be
propagated in the multimode fibres because of the larger core diameter. It is also much
easier to launch optical power into the fibre also that connect the fibres together 2 fibre
can be easily connected by together. Moreover ordinary LED source can be used to
launch light that is most important I do not need laser.

Whilst the transmitter is designed to maximize the amount of light, which enters the
cable in a direction parallel to its length some light will only will inevitably enter
multimode cables at the other angels which is quite obvious. And it is going to total
internal refraction at coming to the end one end. Light which enters a multimode cable at
any angle other than the normal to the end face will be refracted in the core right.
It will then travel in a straight line until it meets the boundary between the core and
cladding materials. At this boundary some of the light will be reflected back into the core
and some will be refracted in to the cladding, right.
Now, multimode graded index fibres is also there where the I think we talked about the
step index fibre. Both in the case of I mean your monomode and multimode fibre
basically this is only graded index there is the of the index refractive, index of the change
abrupt change of the refractive index of the hum refractive index of the material of the
core of the fibre and also the cladding. Whereas in the case of graded index fibre, what
we will find that the the core itself is made of a material where the it is changing slowly
slowly. So, that the light can go in a core in curved path itself there, right? So, multimode
graded index fibres have a core with with nonuniform refractive index. These fibres are
characterized by the curved ray paths, which offer some advantages, but are more
expensive than the step index type.
So, you will find the next day in the, I mean in the lesson 29 we will covered this in more
details. Because the, what is the base principle, what should be incident light? All these
things and then we will go to the some fibre optic based sensors. With this we come to
the end of the lesson 28 of industrial instrumentation.
Prof. A. Barua
Lecture - 29
Optoelectronic Sensor-II
Welcome to lesson 29 of industrial instrumentation in this particular lesson. We will

consider will cover the optoelectronic sensor 2. Basically the contents we will like this
that, we will first cover the basics of fibre optics. And then some basics of fibre optics,
because we have already, discussed some basics in the industrial in the lesson 28. We
will cover details the transmission applied in the optical fibre, then I will come to the
fibre optic based sensors. Let us look at the lesson, so optoelectronic sensor 2 contents.
Introduction to fibre optics traveling of light through the cable, then monomode cable
multimode graded index cable and fibre optic sensors. These are the basic things; we
should cover at the end of the lesson.
Now, please one thing you note that, in this particular lesson. Sometimes, I will tell the
cable, but cable means basically I want to mention that it is basically, a fibre optical fibre
nothing other than that right? So, at the end of the lesson, viewer will know, how does a
light travel in a optical fibre, fibre optic switches, intrinsic fibre optic sensors, extrinsic
fibre optic sensors, right?
Now, you see, this is the, this the typical a fibre optics, you can look at lights are I mean,
how the light travels in optic ((Refer Time: 02:15)) this basically, we have shown a step
index fibre. So, in the step index fibre, you see here, please note that I am using the
British English. So, it will be fibre like this, one or I can take the other pen. So, it is fibre
and ((Refer Time: 02:34)) in the light travels, it always goes through a total internals
reflections. You see here light is traveling and these incidence is such a way, that it will
have a total internal reflections. So, that nothing no none of the lights, will be lost in the
cladding and totally, it will be reflected. So, that refracting index of this one of the core
this is the core this is actually, the core and this is a cladding right these are the cladding.
So, the refractive index of the core is higher than the that of higher than that of the
cladding and incident lights we actually, arrange to and fall in such a way. So, that the
light will get a total internal reflection, so it will follow zigzag fashion, please note it will
be like this one, So, it is not like this, it will be like this light will follow, like this one.
So, this phi should be greater than, the critical angle of incidence, because if it is not not
more than critical angles. So, there will be no total internal reflection, so the, we have to
arrange for the total internal reflection. So, that even this fibre is not steady, if it is bend
like this one. So, light can travel through, this that is a idea of this I mean, optical fibre.
So, even though, we now, the light travels in only in a straight line using this total
internal reflection. We can see, that the in the optical fibres, light can go through a bend
path right. That is a tremendous, advantage and using that principles, we have used many
though, if a bend like this. You will find that there is a loss of intensity and all those
things if the light travels straight. So, loss of intensity will be, even though we are
saying, that it is as a total internal reflection. And all those things, it will be some light
will be dispersed. So, using that principle, we will find that, we may made several
sensors right. We will come to that later on, now let us come to basic the principle, how
the light travels in a optical fibre.
Transmission of light ray; the light ray shown in figure one this just the previews figure
known, as a meridional ray as it passes through the axis of the fibre core. If cores, the
fibre cores, we will call meridional ray this type of ray is the simplest to describe and is
generally used when illustrating the fundamental transmission properties of the optical
fibres. This always we use these fundamental transmission properties of the optical fibre.
Acceptance angle look like this, you see the what is the acceptance angles? What is the
numerical aperture? These are the basic definitions of the optical fibre. This we will
discuss in this particular lessons at the beginning of this lesson. Now, the geometry
connected, concerned with launching a light ray into an optical fibre is shown in figure 2.
which is which illustrates a meridional ray A at the critical angle phi within the fibre at
the core cladding interface right, see it is you see that here if you look at…

What is the, this the figure 2 you see, it is a is a conical half angle this is the phi theta a.
So, theta is a I mean, because we must know, what is the, what is the maximum value of
theta? A is allowed, because this face, if this is a face of the fibre. This face should be
highly, polished if I fall the light in the normal directions that is, but that is not always
possible. Specially, in the case of a if we use a light source as a lead light emitting diode
there will be a different angle light can fall. So, I have to we must find, what is the
maximum angle value of theta a. So, that, when the light enters the core, there is a total
internal reflections right. So, that we must find it, light you see, travelling like this one
and some light say, if travelling like this one having the total internal reflections. This
light you see, this is the more than the acceptance angles.
So, this light, we will not get the total internal reflections. So, it will be reflected within
the core and it is lost. So, this is the lost I mean, this will not be transmitted, through the
core, this will be lost in the cladding itself. Whereas, if I can make within this theta angle
of incidence is within theta a, we will find that we will get a total internal reflections,
within the core of the fibre. So, we must find what is that? Let us go back again. Now,
the geometry concerned, with launching a light into the, into an optical fibre is shown. In
figure 2 which we have shown; which illustrates the meridional ray A, at the critical
angle theta c within the fibre at the core cladding interface. It may be observed that the,
this ray enters the fibre core at an angle theta a to the fibre axis and is refracted at the air
core interface, before transmission to the core cladding interface at the critical angle
right.
So, we have chosen theta a in such a way, so that it will be refracted in the air core,
because when it is launch in the air. I mean, one launch from the transmitter, we are
assuming, that it is for some distance. The light will travel in air then, it is coming to the
core. So, when it is in the core it is refracted and it bends towards, the axis right, so that
bent was the axis of the core. So, that the I mean and that bends will be in such a way
that will be more than the. So, that when the light falls in the core cladding junction, so it
will be more than the critical angle of the, this 2 this 2 medium. So, that we will get a
total internal reflections let us look at.
So, this is the thing at which I talked about, you see this is a angle this is a ray and this is
a maximum angle. I mean, if we go beyond that, I would not get any total (( )) this we
have shown here. You see, it is more than theta a, So, light is falling another light is
falling see. It is not, it is refracted there is no doubt, about this, because here the medium
the refracting index here. This one of the core is higher than, that of the higher than that
of the cladding. So, it will be, because if it is a normal, it will go away from the normal
and I will get a total. I mean, I will get a bending like this one right, because there is a
jacket ((Refer Time: 08:36)). So, the light will not go, so it will be cladding, it will be
lost right that I told you earlier also.
Now, hence any rays, which are incident into the fibre core at an angle greater than, theta
a will be transmitted to the core cladding interface. At an angle less than phi c and will
not be totally internally, reflected. See there is no totally internally, reflection as I told
you and theta a is sometimes referred to as the maximum angle. Or total acceptance
angle of the fibre optics, it is a total accept, you known it is a solid cone right. It is not a I
mean, I mean, 2 dimension, it is a solid cone theta is basically, a solid cone.
If the fibre has a regular cross section that is the core cladding interface are parallel.
There are no discontinuities an incident, incident meridional ray at greater than, the
critical angle will be continue to be reflected. And will be transmitted through the fibre
that is a very good.
Now, what is the numerical aperture? The numerical aperture very I mean, it is a less
specification of the fibres. So, must know, what is the numerical aperture? So, we will
calculate here, the numerical aperture for a optical fibre. You see, figure 3 shows a light
ray incident, on the fibre core, at an angle theta 1 to the fibre axis, which is less than,
theta a acceptance angle for the fibre, because you see rays to be transmitted through
total internal reflections. So, incident ray might be should be always less than, the
acceptance angle, if it is more then, it will not be totally, internal reflectance right. So, it
should be less than the acceptance angle for the fibre theta a. What is figure 3?
Let us look at this is our figure ((Refer Time: 10:19)), since theta 1 the acceptance angle
is slightly higher. The acceptance angle will be like this one right; this is our theta a
right. So, it is less than that, so I getting an I mean, bending I mean, because it is a higher
refractive index. So, light will get, because this is the normal to the plane of the surface.
So, it will be bend like this 1 this is theta 2 then it will go like this 1, clear let us go back
again. The ray path for the meridional ray, so launched into the optical fibre in air at an
input angle, less than the acceptance angle for the fibre.
Assuming that the entrance face at the fibre core to be normal to the axis, then
considering the refraction at the air core interface and using Snell’s law. We are adding
no into sin theta 1 equal to n 1 sin theta 2, right. So, this is n 1 this no this is the
refractive index of air and this is the refractive index of the core. And this angle of
incidence of theta 1 and this is the angle of refraction theta 2. So, that we can easily,
write the Snell’s law n 0 sin theta 1 equal to n 1 sin theta 2 this equation number 1 where
you can see here, if the phi is greater than, the critical angle at the core cladding
interface. If the phi is I mean, hence equation 1 becomes, if we write that the, you see if
the sin theta 2 again you can go here.
Now, this is if you take this one, so this is you see this is a 90 degree. So; obviously, this
is this is theta 2 so; obviously, phi will be basically, 90 degree minus theta 2. So, if it is
there so; obviously, I can write that, where phi is greater than the critical angle at the
core cladding interface. If it is phi is greater than the critical angle of the otherwise, there
will be no total internal reflections. I can n 0 sin theta 1 n 1 it will be actually, sin of phi
by 2 minus phi, because if you take that angle as a right angle triangle, what were here?
This is right angle, triangle ABC is a right angle triangle. So; obviously, I can write
what; obviously, that n 0 sin theta 1 equal to n 1 into cos phi, right.
Or n 0 sin theta 1, so n 1 sin square phi to the power half, we can write this equation
number 3. Again for the core cladding, we can write again, you see for the core cladding
interface. n 1 sin phi equal to n 2 why see look at? You see, when light will be reflected
here, because this is n 1 sin theta 2 n n 1 sin theta 2 equal to n 2 here. You see, when you
it is reflected, so it will be phi. So; obviously, if I apply, the total internal reflection
principles, I can immediately, write that expression. What is that again, for the core
cladding, we can write n 1 sin phi into n 2 is not it, because that will be 90 degree so;
obviously, it will be 90 degrees.
So, where phi is the critical angle of incidence of light from the core to cladding, because
in that case, if it is critical angle. If phi is a critical angle, this will be sin n 2 into sin 90
degree. So, sin 90 degree is equal to 1 so; obviously, it will be n 1 sin phi equal to n 2.
Where phi is the critical angle of incidence light from the core to cladding, so which I
can, write sin theta a equal to n 1 square minus n 2 square. What is n 1 square? I put n
one inside, so it will be n square n 1 square and we have here, we have we have n 1 n 1
square into sin square phi that is equal to n 2 square. So, in case it is 2 to the power half,
so fine.
Equation 4 apart from relating the acceptance angle, you see this is actually, this n a theta
a we got a calling the acceptance angle right. How to find the acceptance that is the
reason. We made it sin theta n 2 equal to sin phi equal to 1 otherwise, we cannot do it.
Equation 4 apart from the relating the acceptance angle to the refractive indices serves as
the basis for the definition of the important optical fibre parameter, which is called the
numerical aperture. Hence the numerical aperture or in a can be defined as na equal to n
naught sin theta a. Theta is the acceptance angle equal to n 1 square minus n 2 squire to
the power half equation number five, where theta a is the acceptance angle of the fibre I
should actually, write the theta in the previous equation equation number 4.
Since the NA is often used with the fibre in air. So, no is actually, unity with the
refraction index of the air is taken as the unity, so it is simply, equal to sin theta a. So, it
may also be noted, that the incidents, incident meridional rays over the range of theta
theta 1 is the incident angle. So, it should be lied between 0 and theta, if it is 0, there is
no question of total internal, if light will straight will flow will pass through the phi core.
But if it is more than 0 degree, so obviously, more than 0; obviously, there will be total
internal reflection, because it is less than theta a, we have chosen theta in such a way. So,
that the, it will be any light within this range, if it falls it will go through a total internal
reflections. So, the light will be propagated within the fibre, the numerical aperture may
also be given in terms of the relative refractive index, index difference delta.
Between the core and the cladding which is defined as delta equal to n 1 square minus n
2 square by 2 n 1 square, so which we can write n 1 minus n 2 by n 1 for delta. If it is
delta less, less 1 we can write like this one equation, number 6 hence combining equation
5 with equation 6. We can write that, numerical aperture is equal to n 1 2 delta to the
power half, which is equation number 7.
The relationships given in equation 5 and 7 for the numerical apertures are a very useful
measure of the light collecting ability of a fibre, because we a must know accordingly.
We will make our transmitter, even though I mean, fibre optics sensor also, we need
transmitters we need receivers though we talked about as, if we are only using for
communication that is not actually true. You will find here also the light will travel to the
fibre. They are independent of the fibre core diameter and will hold for diameters as
small as 8. Even, if it is a diameter of the core is 8 micron the light will I mean this
equations will follow.
While the attenuation has been minimized, there is a remaining problem in that the
transmission, time of the parts of the beam. That travel in the zigzag manner will be
greater, than the light which enters the fibre at 90 degrees to the face. And so travels in a
straight line to the other end. So obviously, if it falls on a 90 degree is not, it if the light
falls on a 90 degree, because we are talking about. So, much you see if I take a dark 1.
(Refer Slide Time: 17: 58)
So, we are talking about like this one this is our fibre and we have a cladding here. So,
this is our central, so light is falling like this one right. But at the same time light may fall
on a straight line, if it falls in straight line. So, the angle of incidence is 90 degree is not,
it is not very good, I can take this one. This is 90 degree, then is not it does not matter,
because it is even in that case. Even it falling on perpendicular to the face, but the angle
of incidence theta a in that, case will be 0 degree is not it theta should be always, less
than that value. Otherwise there will be no total internal reflections anyway let us go
back.
While the attenuation has been minimized, there is a remaining problem in that that the
transmission time of the parts of the beam. That travel in the zigzag manner, will be
greater than the light, which enters the fibre at 90 degrees quite obviously. A light
travelling in a zigzag fashions, with the total internal reflections. It will have some time,
where the light? When the light? I mean, travels straight away, through the fibres. When
the light falls with the theta equal to 0 degree; obviously, in that case, it will take less
time is not it.
In practice, the incident light rays to the cable will be spread over the range given by the
acceptance angle and. So, the transmission times of these separate parts of the beam, will
be distributed over a corresponding range, right. It will be distributed transmission times,
of different ray, so in the case of multimode fibre, so this will be very predominant.
These differential delay characteristics of the light beams are known as a modal
dispersion in a optical fibre. So, is the dispersion of light, so lights are getting more and I
mean, time is not equal time, it reaches at the end, but this I mean, problem can be
eliminated in the case of graded index fibres. So, it, because this problems, what we have
talked about is a multimode fibre? Multimode step index fibres will, this will be more
you will find easy to see this one in the graded index fibres right. It will not the problem,
not be there at this modal dispersion, will almost eliminated in the case of the graded
index fibre.
Now, step index fibre the optical fibre with a core of constant refractive index n 1. And a
cladding of slightly lower refractive index n 2 is known as a step index fibre as simple as
that. It is, because the refractive index profile for this type of fibre makes, a step change
at the core cladding interface clear. So, let us look at that type of I mean, step index fibre
or step index fibre. So, this step index fibre can be 2 type, it can be monomode can be
multimode. So, let us look at first monomode.

Let us take white page say, it will look like this you see, that I have a I can take a large
dark pen. A cladding here, I have a fibre here. Let me draw, one the end joining and the
center line, which we studied in the first year the drawing like this one right. So, what
will happen? You see here, this is the cladding both side cladding and center. We have a
is a cylindrical in shape, what will happen? In the case of multimode fibre. So, light will
travel in these various directions. So, it will travel like this, then it will travel like this
right. It will like this one. It can also travel like this, it can travel like this it can travel or
it can go like fashion, it can go in a more like this right in all these different cases. Now,
the index profile; that means, refractive index profile, we look like this one, it will have a
step change; that means, you see it will look like here. So, actually, it will look like this
one right, so this will right.
So, let me get back the pen again and see it will look like this right. So, here the
refractive index here, is n 1 and here the refractive index is n 2 right. So, this refractive
index, we are plotting RI refractive index, we are plotting. So, this is the multimode
fibre, so this is core, this is cladding, this is multimode fibre. And in the case of step this
is say please note, there is a there is a step change of the refractive index, RI is not it.
There is a step change here, at this point there is a step change that is a reason, we are
calling step change multimode fibre right. We have a step and we have a, we have a
single mode fibre also right, so its look like this.

So, I have a core as before then I have. So, in the case of multimode fibre I am sorry in
the single mode fibre the core diameter is extremely, small. So, that is the problem, we
need different sort of source here, while I mean, directional sources. So, the light will
travel in this case the refractive index profile. If I plot the refractive index RI, it will look
like I have a small change here. So, this is our n 1 this is our n 2, so this is our central
line. So, the light also travel to this one, like this one this is core and this is cladding as
before. This, I talked about the single mode fibre, single mode step index fibre, these are
we called again the step there is step index. There is a step change of the refractive index
see, index means basically, we are talking of the refractive index please note.
So, there is a step change of the refractive index, that is the reason, we are calling its
single mode step. I mean, index fibre, because there is only 1 mode light can transmit to
the core core diameter is usually, very small right let us go beck. So, this is about the step
index fibre clear, So, the optical fibre with a core of constant refractive index n 1. And a
cladding of slightly, lower refracting index n 2 is known as a step index fibre. It is,
because the refracting index profile for this type of fibre makes a step change at the core
cladding interface.
Now, a multimode step index fibre with a core diameter of around 50 micrometer. Or
greater that is larger enough to allow the propagation of many modes in many incident
situations. Many modes within the fibre core, it means that the there are many possible
ray paths through the fibre. Multimode is basically, there is a different ray paths or
multiple ray paths, through the fibre, that is we are calling multimode. However
monomode step index fibre allows the propagation of only 1 transverse, electromagnetic
mode and hence the core diameter must be of the order of 2 to 10 micron or micrometers
right. It is quite small, we have seen that if find graded index, it is in between the 2 right,
it is we will come to that.
Now, comparison of the 2 modes 2 modes of fibre, if I compare the 2 modes the
monomodes step index fibre has a distinct advantage of low modal dispersion. As I told
you earlier, which is basically, defined as a broadening of transmitted light pulses. Since
only 1 mode is transmitted whereas with the multimode step index fibre, considerable
dispersion may occur due to the different, differing group velocities. As we told you,
because there is a I mean, light ray which is going straight to the along the axis, that is
one speed of watts is going towards total internal reflection. Because in there is large, I
mean I mean, incidental or, so that we will have different group velocities. So, that the
end will there is a modal dispersion right. When the bandwidth requirement is low the
multimode fibre has a several advantage. Now, again the problem is that in the case of
multimode fibre; obviously, the bandwidth will be narrower right as several advantage.
Now, what are the advantage? If I look at the advantages, that you will find that in the
case of multimode fibre, we do not have any restrictions on the transmitter. Because we
have to transmit light either you have to use a lead or light emitting diode or laser
sources. Since the light emitting diode has a little disperse though I mean, the ray will be
dispersed that can be used for transmission of lights in the cases. So, we do not need any
very stringent requirements of the coherent sources and all those things, which is
necessary in the case of single mode fibre. So, this is a great advantage, because the cost
of the transmitter will be reduced, even though for the low. So, therefore, the, if the
bandwidth requirements is low; obviously, we go for the multimode fibre.
Because what will happen that I can save the money on the actually they on the
transmitters. Whereas, in the case of single mode fibre, we cannot use lead light emitting
diodes, we have to use the laser sources, which is very coherent and all those things
right. Now, graded index multimode fibre; the graded index fibre do not have a constant
refractive index. That is no step change of the refractive index in the fibre, but a
decreasing core, refractive index with radial distance from a maximum value of n 1 at the
fibre axis to a constant value of n 2 beyond the core radius in the cladding right. So, that
is step change, we will find there is some tremendous advantage. So, for that modal
dispersion is concerned, we will get.
So, we are getting the advantage of the graded index fibre, you see that we can use the
same the light sources. But I am mean, you can use the here lead, but the bandwidth will
be slightly better I mean, because here what advantages that there the light will reach at
the end. Even whether it is going in a straight the core or through a refractive total
internal reflections in different angles is does not matter, it light will reach the end of the
fibre at the same time. So, this, the great advantage, so a multimode graded index fibre
has a parabolic refractive index profile, So, let us look at that what is that actually talking
about? Let us take a white page.
It will look like this; that means, you see the refractive index profile will look like this,
we have is a multimode fibre. So, it has its look like, this is going like this, this is n 2 n 2
and this is refractive index n 1, but please note this n 1, I should should not write here
actually, there is a. So, actually what will happen? You see that n 1 is at the at the center
point n 1 actually, if I look at. So, it will the refractive index n 1 is at the center point
here actually. So, the refractive index will slowly, decrease to the n 2 value as I go from
a center to the cladding, because this is the cladding and this is a core. So, at the center
the refractive index is here you know as I move towards the cladding. The refractive
index slowly, changes slowly decreases at the up to the at the point n 2, it is at the point
the clad the this whole junctions it is now n 2, right.
So, that is now, the refractive index profile, if I plot it will look like this one no more
step change. So, it will look like this, so I have, so it is parabolic like this one right. So,
this is our n 1 at the center, so I can erase again. So, I can, so , so here it is n 1 and here is
it is actually, this will be like this 1 this is n 2 right. So, this is how light will travel like
this? One you see light fall and it will travel like bend path like this one, because slowly
light will bend right that is a reason, we are calling it right. So, light will come like this,
one. So, do not need say that it is to be go to the at the cladding is not necessary, because
before in the ((Refer Time: 32:58)) angles between the layer, because there are multiple
layers. So, it will move like this one clear, this is a difference. So, this is core, this is
cladding clear; that means, if I draw it will looks like this one.
That means if I draw, if I have a fibre like this one. So, the light will fall, this is our axis,
so I have a several layers. So, suppose this is n 1 this is n 2 now, a ultimately it is coming
to the cladding. So, this is our cladding, so light will fall like this, it will bend again to
this one, this will bend again, until and it is gets a total internal reflections clear. So, this
is a some I mean, I can write n 1, n 2, n 3 and n 4 right. So, n 1 is greater than n 2 greater
than n 3 greater than n 4 this ((Refer Time: 33:56)) I mean, for visualizations or
understanding the, what is greater, how the greater get into server box? These way we
are dispersing. So, it is not necessary, all the race with go at the at core cladding interface
in between, it may have a total internal reflection, if the incidence angle is more than the
critical angle clear. So, multimode graded index fibre has a parabolic refractive index
profile the gradual decreases in the refractive index from the center of the core to the
cladding creates many refractions.
Advantage what are the advantage? The multimode graded index fibre exhibit for the
less modal dispersion, than the multimode step index fibre reason. I will show tell you
although, many different modes exist. The different group velocities of the modes tend to
be normalized, by the refractive index grading. Because of the refractive index grading,
because you see the light which is travel at the center at along the axis of the fibre core
((Refer Time: 34:53)). So, ((Refer Time: 34:54)) if it goes to the extreme I mean, a core
cladding interface, it has a higher speed, because the refractive as you know, the
refractive. If the refractive index increases a speed of life also decreases, the rays
travelling close to the fibre axis, that I what I told you have shorter path. When the
compared to with the rays, which is travel into the outer regions of the core, this is path
is shorter, but the speed will be lower. So, ultimately, the, I mean the at the same time it
will reach. So, the modal dispersion is not be there, right.
However, the near axial rays are transmitted through, a region of higher refractive index.
And therefore, travel with lower velocity, then the more extreme rays and it compensates
for the shorter path length and deduces dispersion of the fibre which is going at the end,
which has a I mean, less days. I mean it will take I mean less time, because even though,
it is getting total internal reflection that the refractive index is less. So, the speed of light
will be higher those case. The multimode graded index fibres with the parabolic
refractive, index profile, which we have shown have transmission bandwidth may be the
orders, of magnitude greater than the multimode step index fibre bandwidth. So, this is a
great advantage of this one.
Now, let us come to the fibre optic sensors. Now, there are basis of the operation of the
fibre optic sensors is the translation of physical quantity measured into a change in one
of the ((Refer Time: 36:19)) or more parameters of a light beam.
The light parameters that, can be modulated are one or more of the following intensity
can be modulated, phase can be modulated, polarization will be there wavelength can be
modulated and transmission time. ((Refer Time: 36:42)) see in most of the case, we find
the intensity of light is very easy to be modulated, we will some sensors we will discuss
where the intensity of light is modulated.
Fibre optic sensors usually, incorporate either glass or plastic cables, because now, all
plastic cables are coming up very fast a glass is a cheaper than, the silica based I mean,
fibre. So, we must give some emphasis on the all plastic cables, because it is loss is more
though in the case of all plastic cables, but it is cheap also, it is a mechanical strength is
better than the all glass fibre. All glass types are rarely used, because of their fragility
and plastic cables have particular advantages for the sensor applications. Because they
are cheap and have a relatively, large diameter 0.5 to 1 millimeter making connection to
the transmitter and receiver easy using multimode, there is no problem.
So, the connection transmitter and the receiver transmitted to the cable, to the to the fibre
optic cable also from the again, from the cable to the receiver is easier. No in the case of
monomode fibre, it is most difficult to launch light I mean, to couple a transmitter and
the light itself, because is a small diameter due to small ((Refer Time: 37:52)) where is in
the case of plastic cables we can have a larger diameter. So, the launching of the light is
easier. So, that type of ((Refer Time: 38:00)) will not be there.
The cost of the fibre optic cable, itself is insignificant for sending the applications for
sensing applications as a total cost of the sensor is dominated by the cost of the
transmitter and receiver. So, always preferred lead light emitting diode and photodiode
for a which are to be used as a transmitter and receiver respectively. Fibre optic sensors
typically, has a long life for examples, the life expectancy of reflective fibre optic
switches is quoted as 10 million operations. Their accuracy is also good with for instance
plus minus 1 percent of full scale reading being quoted as a typical inaccuracy, level for
a fibre optic pressure sensor.

Further advantages are their simplicity, low cost smaller size and high reliability and
capability of working in many kinds of hostile environment. So, hostile environments I
mean, where is there is dust? When there is a chance of a large electromagnetic field.
Where the electrical connection is very ((Refer Time: 38:58)) or some you cannot use
electrical taking suppose. I have a thermocouples. And there is a it is ((Refer Time:
39:04)) I mean, transformer where it is large one it ((Refer Time: 39:07)) influence,
because if there is a surges or something in transforms, that will affect our readings, so
right. And, so if we route at the strip ((Refer Time: 39:15)), which is recording the
temperature that also will create problem. So, that type of problem does not come in the
optical fibres.
So, that is a great advantage of this type of fibres more over, please note in many, many
process industry. This is it is forbidden to use a high voltage; I mean especially, the
electrical cables voltage should be around 40 volts and especially, in hydrocarbon
industry which is highly inflammable industry, with that type of applications of fibre
optic sensors. Though previously use people are using a pneumatic system where the
safety is very high, but the pneumatic system has high it is rugged and I mean it is
rugged. Obviously, but it is very large in size all those things are problems and the
maintenance is difficult. Whereas, if you use fibre optic sensors in that type of situations
which will I mean satisfy both the thing that means, which is safety will be there.
Because since it is very low voltage and entire transmission is like through light there is
no electrical connection as though right.
However in spite of these obvious merits industrial usage is currently quite low that is we
must admit right even though this is. Because there might be there is some vicious circle;
that means, people those who are are using the conventional sensors whether the new
additions. These was not I mean, not new, when the, we are using pneumatic sensors in
the process industrial pneumatic system, process in the industrial. When the electrical
system came, which is now, converting all through 15 PSI of control signal to 2 4 to 20
millions, but there is also lot objections, but later on we slowly, we found over the
details. All this I mean, pneumatic system were replaced by the by the electrical system
like, where is the current is 4 to 25. Similarly, we also hope that in future, this also this
all will be replaced by the fibre optic sensors.
Now, fibre optic sensor, let us look at 2 major classes of fibre optic sensors are exist,
what is the intrinsic sensors and extrinsic sensors. In intrinsic sensor the fibre optic cable
itself is a sensor whereas, in the extrinsic sensors the fibre optic cable is only used to
guide a light, to and from the conventional sensor right right. So, that is not actually,
sensor or such I mean, I am talking of the extrinsic sensors. But we are using actually,
the extrinsic sensors that the cable to take out some light actually, to the, the
conventional sensors, which is might be far away right, we will see some that type of
typical situation, one of the example is the measuring the Jet engine exhaust
temperatures.
I can use a optical and pyrometer, but it is very difficult one line measurements right,
because jet engines I mean, jet is running at a all digit of four 40 thousand feet’s and it is
one line measurements is difficult. So, I have to install optical fibre where the, which is
send the light to the transmitter at the receiving end. I can use one optical pyrometer
might be, it is happening ((Refer Time: 42:10)) time to measure the engine temperature,
because the thermocouple will not for, because this a temperature is extremely high.
Now, intrinsic sensors, Intrinsic sensors can modulate the intensity phase polarization,
wavelength or transit time of the light right. This one sensors, which modulate light
intensity tend to use mainly, multimode fibres, but only the monomode cables are used to
modulate to other light parameters.
A particular features of intrinsic fibre optic sensor is that, it can if required provide
distributed sensing over the distances of up to 1 meter. Light intensity is the simplest
parameter to manipulate in intrinsic sensor, because only a simple source and detector
are required right.
The various forms of switches shown in figure 4 to 6 are perhaps the simplest form of
these as the light path is simply, blocked and unblocked as the switch changes position
right changes state. Let us, look at modulation of the intensity of light of the transmitted
light takes, place in various simple forms of proximity, displacement pressure pH and
smoke sensors. And some of these are sketched in figurers 7 to 8.

Let us look at in proximity and displacement sensors the latter are often given to the
special name Fotonic sensors. And the amount of reflected light varies with the distance
between the fibre ends, and a boundary in pressure sensors. The refractive index of the
fibre and hence the intensity of light transmitted varies, according to the mechanical
deformation of the fibres, caused by the pressure, we will show some example.
This is you see, the, a simple shutter switch; that means, if the shutter comes down;
obviously, will find if the shutter comes down. The shutter comes down like, this one
what will happen? You see there it will modulate the light, because if the light is coming
intensity would not get in here. So, this is this is a shutter switch.
This is the reflective switch, if the switches changes positions, suppose if the switches
changes position like this one. Then what will happen? In the switches position, switches
come to this positions so; obviously, the light sources or the light switches light positions
light source will not get in light I mean, light out. So, at the receiver, we would not get
get any light. So, this is reflective switch this is also can be used for the as the switch for
the fibre optic sensors.
.
This is the fibre optic switches; that means, is optical micro switch; that means, if I the
fibre bends like this one the fibre bands in this directions. So; obviously, if the light have
a light sources, have a light receiver, so obviously I would not get in get a light output.
So, this is all the use of optical micro switch right. So, this is also a example of fibre
optic switches.
This is you see, the applied force we are using a which as we say the some sensors is a
force measurement. What will happen? If the light falls on this one this straight light
fibre is the fibre . So, this will be corrugated in shape is not it if it is in corrugated in
shape. So, the; obviously, light intensity will be modulated, by looking. So, if it is the
more and more pressured, it will be more and more corrugated. So, the intensity light
here, which will receive if it is straight; obviously, I will get more light. So, if it is I
mean, bends like this one less light, so obviously the sensing the intensity of light can be
caliberated iin terms of forces right.
This is example intensity modulating sensors a roller chain pressures sensor you see that
with the chain pressure is higher then what will happen this the roller chain pressure is
higher and higher. This is the same example if you whatever pressure four sensors we
have discussed the more and more bending of the light being here. So; obviously, I will
less less light will received at the we at the end. So, that can be caliberated in terms of
the pressure. So, the how much is the chain pressure can be calculated by you can
intensity of light receiving at the receiving end.

You see this is the some with these intensity modulating proximity sensors if the light
goes if the light is coming through this one and it is getting reflected and it is light is
coming out if it is coming more and more. So, if it have a reflecting surface here so;
obviously, I will receive more and more light if it is far away. So, I will get less light, so
the intensity of light again here it is modulated according to the position of the, this
proximity sensors right here. It is pH sensor, which will be the principles will explain
after sometime.
In the pH probe, the amount of light reflected back into the fibres depends on the pH
dependent colour of the chemical indicators in the solution around the probe tip right.
What we are launching and what we are receiving? Obviously, the intensity will not be
same finally, in the form of smoke detectors, 2 fibre optic cables placed either side of a
space detect any reduction in the intensity of light transmission between them caused by
the presence of smoke. Very simple suppose I have a I mean I have a light source here.
And sorry let me take it is I have a fibre here and I have a fibre here. So, I am getting the
smokes comes here what will happen the intensity of light received at this end this is the
transmitter and this is our receiver. So, what will happen intensity of the light will be
decreased since the smoke will; obviously, decrease the intensity of light. So, the;
obviously, this can be used as the smoke detectors. So, it will turn on some alarm or
some announciation right.

Now, a simple form of accelerometer can be made by placing a mass subject to the
acceleration on a multimode fibre. The force exerted by a by the mass on the fibre causes
a change in the intensity of light transmitted. And thus allowing the acceleration to be
determined and the typical accuracy quoted for this device is plus minus 0.02 g in the
measurement range of plus minus 5 g 6 extremely good. I should say and plus minus 2
percent in the measurement range up to 100 g.
A similar principle is used in probes, which measure the internal diameter of the tubes
the probe consists of 8 strain gauged cantilever beams which track changes in diameter
giving a measurement of resolutions of 20 micron right. So, in many situations we need
this term. A slightly more complicated method of effecting light intensity modulation is
the variable shutter sensor shown in ten where it consists of 2 fixed fibres with 2
collimating lenses and a variable shutter between.
It looks like this you see what will happen, you see that this movement I can get from
many places I can get from the pressure sensors and I can get from the Burdon tube we
have already discuss the Burdon tube. Now, we will discuss that in some details light is
falling in I have a collimated lens. So, if I can put the shutter here; obviously, at the
output also will be less right. Because if you use a collimated lens; obviously, there will
be no dispersions of the light from the, at that light end. So that I can use a simple lead
light emitting diode in the multimode fibres right further as a transmitter.

You see why I am saying this shutter can be used you see that is A suppose I have a, you
have started the Burdon tube see tips looks like this is not it. So, the pressure increase we
have seen that this tip moves like this one this with the some arrangement. We have
connected to the needle or pointer and that dial is calibrated in terms of pressure here you
see this movement I can put on a shutter, is not it? So, light is coming in and the light is
coming out with the call emitter lens and all these things whatever we have. So, light is
coming parallel like this one it is from then I think it is coming to the right then what will
happen if it moves? So, the this shutter will move inside the shutter will move inside this
I should draw like this one; that means, that if the tip moves; that means, if the shutter
will all will go inside. So, it will their intensity light will modulated. So, if it is goes
entire end. So, I can show I should say it has reached the large scale. So, this intensity of
light also will be varied according to the pressure which is giving inside the Burdon C-
Tube. So, that can be calibrated in terms of pressure, right.
Now, moment of the shutter changes the intensity of light transmitted between the fibres
this is used to measure the displacement of various devices as I told you Burdon tubes.
Burdon tubes actually we know that we have basically tip movement is an with a some
sector and all those things we will find it is converted to a a circular movements on a
scale. And we can use it for a diaphragms also for the bimetallic thermometers same
intensity, but diaphragms what will happen we know that very easily. We now have a
diaphragms.
So, with pressure in diaphragms stretch diaphragms it look like this. So, if there is
pressure increase it moves like this one. So, if I now include a shutter here and again
lights will come here lights are coming out light in light out. Then what will happen the
shutter the position in the shutter will give you the intent will modulate the intensity of
light. So, this intensity of light can be calibrated in sound of pressure, because here I am
giving the pressure. So, diaphragms will bend like this is not it also in the case of
bimetallic strip.

As we know the bimetallic strip what will have which is used for the measurement of
temperature also for it can be switched if the temperature increase because of the 2
different alpha 1 suppose this is alpha 2; that means, different coefficients expansions its
bends like this one right. So, now if this strip is moved connected to a shutter so;
obviously, what will happen? You will find that this will be modulated right. So, this is
the one of the way of doing that thing. See yet another type of the intrinsic sensor uses
cable where the core and the cladding have similar refractive indices, but different
temperature coefficients.
This is used as a temperature sensor. Temperature rises causes the refractive indices to be
even closer together and losses from the core to increase. Because if the refractive index
is almost same there will be more and more lost to the cladding there will be no total
internal reflection. So, thus reducing the quantity of light transmitted at the, which we
receive that is the. Refractive index variation is also used in the form of intrinsic sensor
used for cryogenic leak detection. The fibre used for this has a cladding whose refractive
index becomes greater than that of the core when it is cooled to cryogenic temperatures.
The fibre optic cable is laid in the location where cryogenic leaks might occur. If any
leaks do occur light traveling in the core is transferred to the cladding, where it is
attenuated. Cryogenic leakage is thus indicated by monitoring the light transmission
characteristics of the fibre. Another use of the refractive index varation is found in the
devices which detect oil in water in many case it is nesscery to detect.
So, that case what is the principle you see these uses a special form of cable where the
cladding is sensitive to oil any oil presents diffuse diffuses into the cladding. So, oil will
be diffused in the cladding and changes the refractive index of the cladding. So;
obviously, light again will be light will be if the is refractive index changes. So, it will
change the transmission of light because a light will, because in the multimode fibre;
obviously, there will be total internal reflection. So, that will prevent the total internal
reflection. So, at the receiving end whatever the light we will receive that will be less
intensity right unclad fibre are used in the smilar way in these any oil present settles on
the core and allows the light to escape right. So, that is another thing.
The extrinsic sensors then extrinsic sensors. Extrinsic fibre optic sensors use a fibre optic
cable normally a multimode one to transmit the modulated light from a conventional
sensors. As I told you just given a example that, the in measurment of a temperatures of a
jet engine exhaust jet engine where we can use an optical pyrometer. But the light is to
be transmitted to the optical that is the type of sensors we are calling it extrinsic sensors
instead of a intrinsic sensors. So, let us not fibre optic as a sensors as such, but the
transmission of the light from the in the optical mode to the conventional sensors. A
major feature of the extrinsic sensors, which makes them so useful is their ability to
reach places which are otherwise inaccessible. As I told you in jet engine which is
traveling at a height of 40000 feet and it is a inaccessible. So, it is in that case optical
fibre is very suitable.
One example of this is insertion of fibre optic cables into the engines of aircraft to
measure temperature by the transmitting light into an optical pyrometers located
remotely from the engine.
Extrinsic fibres optic sensor provide excellent protection of measurement signals against
noise corruption. So, that is inherent to the any optical fibre sensors it is immune to the
electrical noise a surge anything. It does not matter and moreover 2 optical fibre can go
side by side without having any crossed or this is not possible in the electrical fibre. If I
take the thermocouple lead wires to thermocouple lead wires goes parallel then if there is
surge in one that will affect the other thermocouple lead wires also. So, obvious that will
affect the output which we will receive unfortunately the output of many forms of
conventional sensor is not in the form which can be transmitted by a fibre optic cable.
Conversion into the electrical form must, therefore take place prior to transmission.
For example in the case of a platinum resistance thermometer temperature changes are
translated into unbalanced voltage of a Wheatstone bridge right. The unbalanced voltage
is modulated and launched in the fibre optic cable through the usual type of transmitter.
Because the unbalanced voltage I can unbalanced voltage also will always keep the
measurement of temperature. But that would can be modulated that can be converted
through a, that voltage and later accordingly it will give the light intensity. So, the
intensity of light we will be modulated will be depended on the unbalanced voltage. So;
obviously, the intensity of light can be measured of temperature of the object which we
put.
This complicates the measurement process and means that the low voltage power cables
must be routed with the fibre optic cable to the transducer. One particular adverse effect
of this is that the advantage of intrinsic safety is lost because we are using the electrical
system. So, that intrinsic safety which we talked about, so much of the optical fibre that
is lost there.
Now, fibre optic instrumentation networks looks like this that little or any saving arises
of the installing the fibre optic links in instrumentation networks indeed there may be a
cost penalty. Because many sensors we are using; however, there are great advantages in
terms of the links immunity to corruptions of the signal carried.
Immunity to current surges caused by the stray electromagnetic field also affords
protection to computers in the networks as such can causes software corruptions. As with
the fibre optic sensors the cost of the short fibre optic links in instrumentation networks
is dominated by the cost of the terminating transducers that is the problem. That means,
everywhere I need a transmitter and receiver that it means the use of the increasing the
cost, but in some applications as I told you safety is most important. So, in that type of
situation we should not consider for the cost right. However as a length of the fibre link
becomes greater the cost of the fibre optic cable becomes more significant.
The cheapness of the plastic cables is attractive for instrumentation networks, but they
cannot be generally be used. Because it has a large loss as I told you is a large signal
attenuation in a plastic cable, but in the short range it is very useful. Because that will
reduce the cost of the fibre though as I told you repeatedly the cost of the fibre is
insignificant compared to cost of the transmitter and the receiver. One disadvantage of
the fibre optics compared with electrical conductors is a networks is that the light
connections at the ends of the cable are much more costly. Then the electrical
connections and branching of the light through the cables is not is is difficult to
implement. What is that you see it is very simple what its look like this.
I have a electrical circuits like this one I can have branch it I can make a another branch I
can make another branch or I can make another branch like this one. I can do like this
one I can put like all the different branches in possible where in the optical fibre it is very
difficult if I need branching I need another transmitter there . So, the bend of the light is
not possible which is very easy in the case of optical in the electrical systems.
The current research is focused on the distributed sensors measuring different process
variables along a fibre optic cable. Alternatively, sensors of the same type, which are
located at various points along a cable are investigated as a means of distributed sensing
of a single measured variable. That is only possible, but it it coming out the fibre optic
sensor are coming up very fast. When you we think that in the in few decades I think
within next decade, the fibre optic sensors will dominate. It will replace all the electrical
sensors or electrical transmission system in the instrumentation networks in any process
industry. With this I come to the end of the lesson 29 of industrial instrumentation.
Prof. A. Barua
Lecture - 30
Synchro
Welcome to lesson 30 of industrial instrumentation. In this particular lesson, we will

synchro, which is basically inductive based devises. And this can be used as an error
detector or position sensors, not the linear position rotary positions can be can be rotary
position can be detected can be measured and we will get the voltage output. As well as
synchro is widely used for the feedback, control systems in AC position control systems.
And also in continuations with the synchro, we will consider the micro synchro or which
is called more popularly named as microsyn. That is also will be covered in this
particular lesson.
Synchro is actually a steel now, you will find it is used though there are digital shaft
encoders, which is used for knowing the position of the system positions of the rotary
shaft. But the synchro along with the servo mechanisms where that means, where the 2
phase servo motors can be used for position control system that means, I will give some
particular position. That can be easily attained by the use of synchro when the synchro is
used as a error detector. As well as for remote transmissions of the synchro some
synchro is also can be used, over a long distance synchro can be used.
So, let us look at the contents of this particular lesson. Lesson 30 - synchro.
Contents are the synchro as well as you know that it has a transmitter and it has a
receiver. So, we will consider the transmitter first as well as receiver the 2 are almost
same thing. I mean it has 1 rotors and stator, both in the case of transmitter and the
receiver we do not call it, we call it control transformer in the case of synchro. So, it is
always is to be used in the pair form; that means, transmitter cannot be used alone or
receiver cannot be used, alone which is called as a control transformer. So, this always is
to be used in pair. So, that is the reason sometimes we call it always or sometimes we
call it synchro pair, instead of calling just synchro, because synchro alone 1 transmitter
or 1 receiver will not work, right.
So, keeping that in mind, so always please note that the synchro always see whether it is
error dedicator systems or reciprocation control systems or as a mechanical shaft error
measurements. I mean all these things, I mean involved with the synchro pair; that
means, one transmitter and one control transformer. So, we will discuss these details, this
transmitter and control transformer. The first we will discuss the transmitter then we
have control transformer which is nothing but a receiver, but what we are calling it
control transformer, because the there is a transformer action. The transformer action is
also there in both in the transmitter and the control transformer, but we are calling it
control transformer.
Then we will discuss the synchro error detector. Synchro can be used as the error
detector, because there is a misalignments of the 2 shaft can be detected. Because we will
get a professional output voltage from the misalignment of the shaft that we utilized to
use in control systems. We will also discuss the microsyn or micro synchro which is a,
which is a smaller versions of the synchro with it is also synchro; that means, it can be
also used as a position sensor. But the advantage of the microsyn is that it is a basically
used for a low torque devices and you will get the there is no winding on the rotor of the
microsyn. Whereas, in the case of is synchro we have windings; that means, coil
windings both in the case of rotor as well as in the stator. Because since it is a motor
devices you it must have sub stator and it has a rotor also I mean stator and rotor.
At the end of this lesson, the viewer will know the principle of operation of a synchro,
the construction of rotor of synchro transmitter. The construction of rotor of synchro
control transformer, AC position control systems. We will see that AC position control
systems can be used that we will discuss in details here.
Introduction let us look at synchro is an AC electro mechanical system which basically

makes the angle measurements. Angular measurements of the shaft can be detected thus
I mean rotor it sense it is not actually for the linear measurement, but it is a angle
measurement devices, right. This is slightly differing I mean, so for the in the all the 39
lessons we have not discussed anything on the angle measurements. Now, with the rotary
angles measurement is possible; that means, if there is a misalignment of the shaft that
can be I mean that can be detected by the help of synchro.
Synchros for angle measurements are most widely used as a components of servo
mechanism. Servo mechanism is nothing but a feedback systems automatic feedback
control or automatic control systems in short form we call it servo mechanism where
they are used to measure the compare the actual rotational position of a load, with its
commanded position. Why there will be some difference, because of we have some
desired position of the load rotary position angle position that can be achieved with the
help of synchro and in conjunction with the servo mechanism if the torque is large, right.
Otherwise, we can use a synchro pair we will discuss all these in details.
To perform this operation a synchro pair is used, and they are called transmitter and the
controller transformer. Construction is almost same stator construction is almost same
the rotor constructions are different, we will look at. The pair is used to detect shaft
position later synchro pair as I told you, sometimes we are just calling it pair is used to
detect the shaft position error.
Theory: synchro is a rotating device that operates on the same principles as a

transformer. And produces a set of voltages is correlated to angular position. There are 3
stator, so from each stator across the each 2 stators; that means, if you have a 3 stators
between the stator 1 and stator 2 stator 2 and stator 3 stator 3 and stator 1. We will get
some voltages that can be calibrated in terms of the position of the rotor right. So, as the
rotor position changes the voltage output also will change.

This is the diagram of a symmetric diagram of synchro, I mean transmitters. Because it is
in I mean it is a big diagrams we cannot accumulated in a single slides. So, in the it is
splinted over in the 2 you see this is a single phase excitations, we are giving here this is
a rotor. So, this is a rotor, this is the rotor, and this single phase excitation goes to the
rotor. Through a slip rings; obviously, we need slip rings because rotor; obviously, we do
not need in any slip rings in the stator. So, we will get some voltages across the stator;
that means, between S 2 and neutral and between S 3 and neutral as well as this 1 S 1 and
neutral, right. We will get a voltage let us look at the second slide.
You see here, this is completes schematic diagram of the synchro transmitter. That is as
just diagram this is a complete diagram of the synchro transmitter. Let us look at, here I
am giving between the R dash R double dash since it is rotor. So, I am giving the name R
dash R double dash through slip rings and to the rotors of the synchro. We have 3
windings in the stator which is which is in this complete systems. We call it the
transmitter this is the rotor of the transmitter it is slightly difference, if you can look at
this is the dumbbell shape we will discuss all this details. Why it is dumbbell shape?
Because actually the command we will give to the, this rotor.
So, it should not roll actually the system in which we are connecting this shaft whereas,
the control transform rotor which is typical cylindrical shape that we will see later on.
Synchro transmitter; let us look at stator is Y connected stator similar, to a 3 phase
induction motor. Though it looks like a 3 phase induction motor, but we will see that we
will get a single phase output, at the stators of this synchro transmitter. Rotor is salient
pole, dumbbell shaped. Why to reduce the weight of the rotor? Because if the there is a
mechanical loading will not be there if the weight of the rotor is less with a single
winding. Where the single winding, on the rotor of the synchro where the excitation
single phase excitation is going through slip rings. Let the voltage applied to the rotor of
a synchro transmitter be Etr time domain, Etr this is the magnitude of the signal ((Refer
Time: 10:11)) into sin omega t. This is the omega is the excitation, frequency of circular
frequency of excitations of a single phase excitations which we are giving to the synchro
rotor, clear?
When the rotor is in position as shown in the figure, let us go to the figure. This is the
rotor position when the rotor is aligned with S 2 if the position is something like that,
when the rotor is in position shown in figure in one, which is defined as the electric 0.
The voltage induced across the stator winding between S 2. And the neutral n is
maximum, in this particular position rotor or the output voltage will be maximum. The
induced voltage across the stator windings between S 2 and the neutral is n is maximum.
And is written as this is the neutral; neutral we are abbreviated as S. Please note this is
the neutral we had abbreviated as S. S means this is a neutral we had been abbreviated as
n, and is written as e S 2 n between the stator winding 2 and the neutral is K the factors
we are introducing into Etr sin omega t no phase shift nothing, where K is the constant of
proportionality and 2 other stator voltages can be written as these are the constant of
proportionality. And the 2 other stator volt, there are 2 other stator voltage between the
stator 1 and neutral and stator at the 3 and neutral.
So, those voltages will look like you see e s 1 n KE tr cos 240 degree into sin omega t,
and minus 0.5 into KE tr into sin omega t. This is our e s 3 n time domain t K into E tr
cos 120 degree into sin omega t equal to minus 0.5 KE tr, because 120 degree will be KE
tr Etr is there, because it is in transmitters we are giving it Etr subscript tr. So, I have
abbreviated these transmitters by tr, right into sin omega t.
Therefore the voltage across the stator windings are e S 1 S 2; that means of instead of
neutral if I measure the voltage across S 1 and S 2. I will simply subtract the voltage e S
1 n between winding 1 and neutral and winding 2 and neutral which is equal to minus 1.5
KE tr into sin omega t. E S 2 S 3 equal to e S 2 n minus e S 3 n equal to 1.5 KE tr into
sin omega t. E S 3 S 1 equal to e S 3 n minus e S 1 n. So, these 2 will be in phase, so this
should be equal to 0, clear?
So, these are the 2 different voltages we are getting. So, it will give the position of the
synchro rotor right; that means, transmitter rotor position can be predicted from this
voltage. Because we always we said that we within all synchros we are getting the 3
different stator voltages. That stator voltage will indicate the position of the rotor of the
synchro. It is to be noted that despite the similarity between the construction of the stator
of a synchro, and that that of a 3 phase machine only the single phase voltage is induced
in the stator. As if we is the thing I mean three phase indication machine they are looking
very similar to the 3 phase induction machine, but you will see that the a single phase
voltage induced in the stator.
Now, what I did? I actually I rotate, the previously you see this stator is in this alignment
is not it stator were like this 1 right. Previously what we had? I had the stator like this 1
like this 1 I had a stator like this 1, clear? Now, I have rotated by angle phi in
anticlockwise directions. I have rotated the stator in anticlockwise direction by an angle
phi with respect to S 2. In fact, the positions, which I have shown that is considered to be
the electric 0 right, you see this is angle phi which is making with the stator winding S 2.
If now the rotator is rotated. To a position as shown it figure 2 that is figure 2. I think,
yes. The voltage the voltages in each stator winding will vary as a function of the cosine
of the rotor displacement angle phi. The voltage magnitudes are e S 1 in KE tr cos phi
minus 240 degree. And this will be; obviously, multiplied by the sin omega t please note,
because sin omega t will be there. So, I am not writing it everywhere sin omega t is there,
because that is excitation, is not it? That single will be always there.
So, I am not writing into sin omega t E KE tr cos phi minus 240 degree e S 2 n equal to
KE tr into cos phi and e S 3 n; that means, the between the winding 3 and neutral stator
winding stator between 3 and neutral between 2 and neutral between 1 and neutral. So, it
is 1 and neutral KE tr cos phi minus 240 degree e that between 2, and neutral when the
rotor is shifted by the angle phi from the winding 2 in anticlockwise directions KE tr into
cos phi, and e S 3 n equal to KE tr into cos sorry KE tn.
KE tr cos phi minus 120 degree clear, with sin omega t; obviously, with that will be
there. The magnitudes of the stator terminal voltage will be expressed as e S 1 S 2. That
means, that in this position thus voltages across the stators will be given by e S 1 n minus
e S 2 n root 3 KE tr sin phi plus 240 degree e S 2 S 3. That means, between the stator of
the winding 2 and 3 between that is can be given by the voltage, which will get at the
stator 2 and neutral and the stator 3 and neutral root 3 KE tr sin phi plus 120 degree into
sin omega t everywhere sin omega t is there. And e S 3 S 1 equal to e S 3 n minus e S 1
equal to root 3 KE tr into sin phi, clear?

You see this is the plot of the stator voltages of a synchro transmitter versus phi. What is
phi, which is in counter clockwise directions? Plot of the stator voltage of this you see
here e S 3 S 1. So, you have plotted like this right is not e S 3 1. There is KE tr into sin
phi sin omega is there forget about that, because we are drawing only with respect to phi.
So, e S 3 1 like this 1 whereas, S 1 S 2 you see S 1 S 2 S 1 S 2 has a, I mean 240 degree.
So, it is it looks like this you see at 240 degree it is becoming 0. Whereas, S 1 S 2 will
like this 1 at 180 degree it should become 0, clear?
Construction of a synchro; The physical construction of the transmitter and the control
transformer are identical, except that the transmitter has a salient pole dumbbell shaped
rotor while the transmit transformer has a cylindrical rotor. This is the distinct difference
or only difference between the rotors because stator in the 2 cases are same. Both in the
case of transmitter and control transformers, only the difference is in the case of design
of the stator. Because to reduce the weight we have made dumbbell shaped whereas, the
control transformer rotor is the cylindrical shape. The construction is similar to that of a
wound rotor induction motor and the figure 4 shows the coil arrangement of the
transmitter rotor.
See this is the coil arrangement of the transmitter rotor, and this is the control
transformer rotor. There are 2 null positions of the rotor of the control transformer. This
is the synchro rotors constructions. You see this is dumbbell shaped whereas, this is
cylindrical shape, why this is cylindrical shape? Dumbbell shapes already, I have
defined. And why it is cylindrical shape? That we will define after few minutes.
The rotation of the rotor changes the mutual inductance between the rotor coil and the
stator coils. That is the reason I am getting different voltages is not it. You see write I
will write it again rotation of the rotor changes the mutual inductance, between the rotor
coil, and the stator coils. For a given stator coil, the open circuit output voltages is
sinusoidal in time, and varies in amplitude with the rotor positions also sinusoidally as
shown in the figure 3. Already we have shown that with the changes of rotor positions is
not it we have seen in the figure.
Where is that it figure. You see here, with the changes in rotor positions. This signals
varies sinusoidal is not it. You see the signal varies sinusoidal, with the changes in the
rotor position. Rotation of the rotor changes the mutual inductance between the rotor coil
and the stator coils. For a given stator coil, the open circuit output voltage is sinusoidal in
time, and varies in amplitude with the rotor position, also sinusoidally as shown in figure
3. This output voltage will be; obviously, sinusoidal is not it.
The 3 voltage signals from the stator coils uniquely define the angular position of the
rotor. This will uniquely define the angular position of the rotor, in which position
whether it is clockwise or anticlockwise everything will be known from this when these
3 voltages are applied to the stator coil. These 3 voltages will be applied to the stator
coils of the control transformer. They produce a resultant magnetomotive force, aligned
in the same direction as that of the transmitter rotor.
This will produce the magnetomotive force and which will be aligned in same position.
That means, it will result in magnetomotive force aligned in the same direction as that of
the transmitter rotor, the rotor of the transformer act as search coil in detecting the
direction of the stator field. Rotor of the transformer acts as a search coil; that means,
you see the rotor of the transformer can be act as a search coil. This will give you,
because is there any mismatch between the rotor position of the 2. I mean of the rotor
positions of the transmitter and the control transformers I will get some output if the axis
of this coil is aligned with the field. The maximum voltage is induced into the
transformer rotor coil.

If the axis is perpendicular to the field 0 voltage is induced and giving the null position
mentioned above. The output voltage amplitude actually varies sinusoidally with the
misalignment angle, but for small angle the sine and the angle are nearly equal giving a
linear output. How it is happened?
Let us see the control transformer stator winding is similar to that of the transmitter that
is already we have discussed. The rotor of control transformer is cylindrically shaped to
make the density of magnetic flux same, at all positions of the rotor. Otherwise the
output voltage will be the function of the position of the control transformer rotor itself
right. So, that is the undesirable characteristic previously nowadays you know that is
almost all the I mean the output voltage should not be a function of that. So, you want to
intimate the independent for that particular part. So, that is the reason the density of the
magnetomotive magnetic flux is same at all positions of the rotor.
Otherwise the output voltage will be the function of the position of the control
transformer rotor itself that I do not want is not it. I want that the rotor output will be the
mismatch of the position of the transmitter rotor to the control transformer rotor. But if
the control transformer position of the control transformer gives additional change of
voltage that is undesirable. That I do not want, that is the reason the control transformer
rotor is cylindrical shaped. So, that in all positions; the same density of magnetic flux
will be available. That is distinct difference please note, why it is done like this? This is
done purposefully, so that the output will be independent of the position. Out only, I
mean if there is mismatch of the shaft positions of the rotor with that of the control
transformer I will get the output.
But I do not want to get any change of output for the different positions of the control
transformer rotor. That is the reason we made it it cylindrical shape. I will read it again
rotor of the control transformer is cylindrically shaped to make the density of magnetic
flux same at all position, if it is same at all positions what will happen in the case of rotor
in the transmitter. In the case of transmitter, because it is dumbbell shaped. If I change
the position there are different linkage mutual inductance are difference I am getting
different sets of voltage. But here I will not get in the control transformer I will not get
that thing right that is the reason it is cylindrical shaped. Now, the voltage only deferred
only what were the output voltage I will get from stator of the control transformer. That
which will depend on the position or the position mismatch of the transmitter and the
control transformer, clear?
Error detectors, we will that we will more the function of an error detector is to convert
the difference of 2 shaft positions into an electrical signals. That we want that the if there
is any mismatch of the if the functional error if the mismatch of the difference of the
shaft positions, I will get a electric signals. Therefore, a single synchro transmitter will
not apparently inadequate as I told you single synchro transmitter will not occur. We will
always need a single transmitter. I mean we need the transmitter along with the control
transformer; that means, if we call it synchro pair.
This is the, you see we have talked about; you see stators are look at it very carefully
stators are connected. You see all the stators corresponding this stators is connected this
stators is connected to this stator is connected to this I have a rotor and, I have a single
phase excitation here. Now, if the, these 2 rotors as the same positions I will get the 0
output voltage if these rotor positions are different. This position and these positions are
different. I will get output voltage which can be given by E naught equal to sin of phi tr
into phi ct this phi ct is the position of the control transformer that is phi ct whereas, the
phi tr is the position of the transmitter that is the reason. I will get you phi tr.
So, multiplying factor will be there KE that is not out routing this is the rotor output we
will peak up the voltage at this point. R dash and R double dash whereas, I will take the
voltage here, I will give the excitations R dash and R double dash are there. This is the
shaft of the synchro transmitter and this is the shaft of the synchro control transformer.
What I do not want that the, for the different positions of the control transformer I will
get a different output voltage. This output voltage I will get that will be only the
mismatch of the 2 rotor; that means, phi tr minus phi ct.
But if it is not cylindrically made this rotor of the control transformer is not cylindrically
made then. What will happen? You will see for different positions I will get a different
sets of voltage, even only for that. Now, if the, these 2 are same phi tr and phi ct same. I
will get the 0 output. Because here at that position because these will this output voltage
will not change due to the position of the phi ct only, position of the rotor of the control
transformer. It is the difference of the 2 transformer will give you output volt. That is the
reason this is dumbbell shaped and this is cylindrical shaped, clear?
Now, synchro error detector is made of 2 synchros or a synchro pair or a synchro pair.
that is 1 transmitter and 1 control transformer. For small angular deviations between the
rotor position, a proportional voltage is developed at the rotor terminals of the control
transformer right. That means, for small deviations it will find we have seen that it is the
sinusoidal voltages. So, for small deviations near the vicinity of the null, we can use it as
a linear sensor right.

You see this is a you can see here. So, you can see for 15 degree almost you can see it is
used as a linear sensor the rotor voltage of the control transformer along with the verses
error of the; or mismatch or a misalignment of the transmitter and the control transformer
rotor right. So, you see here it is voltage is increasing it is coming like this 1. So, near the
vicinity of the null near the null I can say sorry. Say the near the vicinity of the null, I
can say it is a linear senses. So, during these positions I can say it is the linear sensor. So,
for these positions I can say it is a linear sensor. Otherwise, it is a non-linear device.
Obviously, the sin is not linear function it is a non-linear function.
Synchro error detector is a non-linear devices, but for small angular deviations, up to 15
degree. In the vicinity of the null positions, the rotor voltages of control transformer is
approximately proportional to the difference between the positions of the rotors of the
transmitter and the control transformer. Synchro error detectors is a non-linear devices,
but for a small angular deviations of up to 15 degree in the vicinity of the null position
the rotor voltages of the control transformer is approximately proportional to the
difference between the positions of the rotors of the transmitter and the control
transformer. Mathematically K S equal to we can write, E equal to phi tr minus phi ct
equal to E by I mean phi e for small deviations. We are writing it is, because sin theta or
equal to sin theta for small e is not it? That is we all we know; that means, sin theta is
almost equal to theta when theta is small right.
Where E is a error voltage phi tr is a shaft position of rotor, of synchro transmitter in

degree, phi ct phi subscribes ct shaft position of the rotor of the control transformer in
degree, phi e is the error in the shaft position in degree. And KS is the sensitivity of the
error detector. It is in volts per degree.
For typical control transformer, the sensitivities lies lie between 100 to 500 millivolt per
degree. And synchro accuracy is usually specified in terms of the minutes of the error as
you know 60 equal to 1 degree. A synchro having plus minus 10 of accuracy will
produce the correct stator voltages for a given angle with no more than plus minus 10 of
error signal.
Apart from their use as a displacement transducer such synchro pairs are commonly used
to transmit angular displacement information over some distances for instance to
transmit gyro compass measurements in an aircraft to remote meters. That is also another
use of the synchro transmitters. They are also used for load positioning allowing a load
connected to the control transformer rotor shaft to be controlled remotely by turning the
transmitter rotor. By transmitting rotor automatically what about that given signals I have
given rotate rotary angles I have given to transmitter that will be followed in the control
transformer.
You see this is a, I mean this is for low torque measurements for low torque applications,
we can use a synchro transmitter receiver pair. This is especially for the low torque
measurements you see the rotor is connected to rotor and stators are connected to stators.
So, what will happen you see that whatever the rotations I will give to phi tr whatever,
the rotations I will give to the phi tr same phi ct will rotate by the same angle. Whatever
the rotations I will give in the phi tr that the same angle will given to the phi ct, right. So,
there will be no error between phi tr and phi ct, because these 2 rotors are connected
together.
I am giving a single phase supply here I am giving a single phase supply here, single
phase supply, single phase supply. I am giving here. So, this is basically it is a I mean it
can be previously you see this type of systems were used. Suppose in a I mean I am
giving the digital clock was not that I mean commonly available. Some 30, 40 years back
you can you can see that people are using this type of synchro, because in railway
stations and all those places air I mean airport previously you know the all the watch
should have the same time.
And that time what they do they have master clock, and they are connected through a
synchro transmitter to a several slave clocks, and since all are connected together. It is
because you know the rotations or the it is basically low torque devices. I mean rotating
a, the arm of the second arms or the minutes arms or the hour arms does not need the
large track. So, this is very well suited for this type of applications where a master clock
whatever way it is rotating the slave will follow exactly the same. So, there will be no
error or mismatch of timing between all the slave clocks right, but basically as I told you
this is a low torque devices.
For these application the transformer rotor is free to rotate and is also damped by
preventing oscillatory motion. The simplest arrangement the common sinusoidal
excitation voltage is applied to the both rotors. If the transmitter rotor is turned this
causes an imbalance in the magnetic flux patterns. If the transformer is the transmitter
rotor turns these causes an imbalance in the magnetic flux patterns. And the results in a
torque on the transformer rotor which tends to bring it into the line with the transmitter
rotor.
The use of a synchro transmitter receiver pair is limited to relatively low torque loads.
And overall accuracy is determined by the friction of the rotor of the control transformer
and load bearings if it is in large. So, it cannot be used, because that that we have to use
some servo mechanism. Otherwise, for low torque divisions when bearing problem is not
that much excuse me I mean we can use this type of device this type of I mean
connecting 2 rotors shorting 2 rotors and shorting. I mean the 1 stators are connected
together in all the pair I mean whenever you are using the synchro pairs all the stators
will be connected on that.
Like that, but if you use a servo mechanisms we do not we will not connect 2 rotates
together right. Excuse me in ordinary applications such as driving an indicator pointer as
I told you, indicator pointer might be the, I mean clock also wall clock pointer at remote
location. The receiver shaft will follow the transmitter shaft with less than 1 degree. This
is the high accurate you can see the 1 degree is nothing right. So, is in ordinary
applications such as driving an indicator pointer at a remote location. The receiver shaft
will follow the transmitter shaft with less than 1 degree clearer.
Now, this torque is typically small and for small displacements. So, that this technique is
only useful if the load torque on the control transformer shaft is extremely small
otherwise you cannot right.
It is necessary to incorporate the synchro pair within a servomechanism, where the

output voltage induced in the control transformer rotor winding is amplified and applied
to the servomotor, which drives the transformer rotor shaft, until it is aligned with the
transmitter shaft.
These we have shown, it is also spread up in the 2 this is a error synchro error detectors
which I am going to show you, single phase excitations we have phi tr, we have phi ct.
But it is spread over 2 slides, because this is a big diagram which cannot be
accommodated on the single slide.
Let us look at the next line you see here, previous slide let us look at. This is now we
have a another, what we have? We have one another. This goes to the next slide please
note right. So, we have another right, this will work as an error in voltage. What is that?
This is phi ct. So, in any mismatch of phi tr and phi ct we will be get will be received as
a error voltage here. You see here this is the things just I have drawn this is our synchro
controlled transformer rotor output.
Because this error voltage we will available in the control transformer rotor out output.
This rotor output is going to the amplifier, and this amplifier is fed to a 2 phase
servomotor. And is to the servomotor is there. So, this is another phase and these 2 are in
quadrator, right. So, the servomotors always will try to with proper sin to reduce this
error voltage. So, it will it will rotor in a direction, if I have a large load in a gear train.
So, that the phi ct will automatically follow the phi tr until or unless this error voltage
becomes 0, right.
We will see the details, this is our excuse me block diagram of AC control systems. We
have phi tr, this is our command rights. I want this particular I mean position I am getting
phi c our goal is phi ct should be exactly equal to phi tr. So, if there is mismatch this is
coming as a error, if this is there is no error phi c phi tr minus phi ct will be 0. If there is
an error there is an non 0 voltage will come down here. This will be amplified this will
be go to the 2 phase servomotor. Servomotor will automatically make the correction until
unless will rotate in a direction until and unless this error voltage becomes 0 and phi tr
will be equal to phi ct. This is the block diagram of AC position control system, right.
KS is the sensitivity of the error detector has an opposite signs at 2 null positions. It has
opposite signs in 2 null position. In the closed loop systems there exists only 1 true null
true position. And the other one corresponds to an unstable operating point on the
system, right. Now, when the synchro positions are close to the true null and the rotor of
the control transformer lags behind the rotor of the transmitter.
A positive error voltage will cause the servo motor to turn in the proper direction to
correct this lag. If this is positive it will always correct the lag right. When the synchro
positions are close to the null, true null, the rotor of the control transformer lags behind
the rotor of the rotor of the transmitter and a positive error voltage will cause the servo
motor to turn in the proper direction to correct this lag. But if the synchro rotor in not in
the true null position, then what will happen? Let us look at…
If the synchros are operating close to the position which is not true null, for the same lag
between phi tr and phi ct the error voltage becomes negative. And the servo motor will
rotor in the direction that will increase the lag. So, this will continue until unless we get
the true null position. The larger the log in the rotor position of control transformer
higher will be the magnitude of the error voltage. Larger the lag in the rotor position of
control transformer higher will be the magnitude of the error voltage quite; obviously,
right and it will cause the servo motor to rotate in the same direction. Until the true null
position is reached, it will go is like this one. Now, the synchro error signal can be
properly written like this.
If I take a blank page and if I take a blue, so it will look like this, it will be error signal, et
will be equal to you can write K phi et into sin omega t. And let me erase sin omega t
you can see this is a this is a wave form of a modulated wave with the suppressed carrier
right. So, it is very similar to the modulated wave with a suppressed carrier, because sin
omega t will be there always. So, it is ((Refer Time: 41:46)) always will be there. So,
error signal will be modulated wave with suppressed carrier, right.

Now, we will discuss next our topic is a microsyn. Microsyn is nothing but micro
synchro. So, it is a trade name as you would say some companies developed. And only
features it has it has that it has no winding on the rotor. Since, it has no windings the
tremendous advantage there is no question of slip rings right. So, obviously, there is a
great advantage of the, this type of devises. This is also a position sensor that means,
angle measurement is possible; obviously, the lode is less torque size is less. It can be
used for the as a position sensors or for the angle measurements let us look at the details
of this one. It is a rotary position sensor based on the variable reluctance, right. It is a
variable reluctance sensor right, based on the reluctance as simplest that.
This is our microsyn you see this is a diagrams of the microsyn, which I have shown we
have a excitation voltage excitation you see these 2 coils are different. If you look at very
carefully you see this excitation is going through this 1 single winding it is coming like
this 1 it is going like this it is going like this it is going like this. And in output voltage
these are also separate there is no connection between the, this expatiation coil and the
output which we are taping, right. You see this is going like this this is going like this
going like this this is going like this going like this it is coming out, right.
So, this is our output and we are giving the input excitation single space excitation
voltage clear. Now, see this windings 1 and 3 we have done in such a way that this will
aid each other. Now, winding 2 and 4 we have made in such a way also that winding 2
and 4 open winding 1 and 3 1 and 3 will aid each other 2 and 4 will aid each other, but
will oppose 1 and 3. Now, in this position in this particular position, so which I have
shown this is the rotor this is basically made of iron this is the rotor this is the null
position.
Why this is null position? You see in this particular position the reluctance of the coil
between 1 and 3 or between this and this between the iron and the coil and between iron
and the coil number 3, will exactly balanced by the reluctance. Because if the reluctance
is large; obviously, the inductance will be small. What will happen you see here the
reluctance of this between angle between soft iron is iron and the coil 2 and the iron and
the coil 4 will cancel each other we will get 0 voltage. If it moves I will get a unbalanced
voltage, right. That is a typical and beauty of the synhro the microsyn they say you can
see here there is no winding on the rotor of the we had taken a section there is no
winding on the rotor of the microsyn, right.
As shown in figure ten the rotor is connected to in the input shaft when the input shaft is
in the null position, where the voltage induced in coils 1 and 3 which aid each other are
just balanced by those of the coils 2 and 4 which also aid each other, but opposes 1 and 3
right. This is the balanced conditions; that means, the as I told you the voltage induced in
the coils 1 and 3 which aid each other will exactly balanced by those of the coils 2 and 4
which aid reach other. But opposite to the coils 1 and 3 that I told at the beginning of this
microsyn right.
Now, if now the shaft rotates in clockwise direction then what will happened if the shaft
rotation in the clockwise direction from the null position? The reluctance of the coils 1
and 3 increases let us go back you see here if it rotation in this direction. Then what will
happen reluctance between coil 1 and 3 will increase, because reluctance also depends on
this co iron. So, reluctance between 1 and 3 will increase and reluctance between coil 2
and 4 will decrease are you get unbalanced voltage. That unbalance voltage will tell the
position of this rotor or the angular measurements will be over. If now the shaft rotates in
a clockwise direction as shown in figure 10 from the null position reluctance of the coils
1 and 3 increases and reluctance of coil 2 and 4 decreases, what will happen in that?
As a consequence induced voltage in coils 1 and 3 degreases, because we have a large

reluctance if this voltage will degrees and that of the coils 2 and 4 increases because the
reluctance has decreases. So, I will get a large voltage. Therefore, a net output voltage E
naught will be available from the microsyn. So, I will get excitation. So, I will get output
voltage the output voltage will tell you the position of the rotor. Now, if the shaft moves
in opposite direction microsyn gives an output voltage, but with a 18 degree phase shift
right. So, it is very difficulty in the in both direction whether it is clockwise direction and
or anticlockwise directions I will get same unbalanced voltage.
But if I know want to since whether the rotor has moved in the clockwise directions or
anticlockwise directions I have to make some phase shift demodulations. That is I am
telling now if the shaft moves in opposite direction microsyn gives an output voltage of
the same amplitude with 180 degree phase shift. That means if the position is same as in
the clockwise direction if the anticlockwise direction, which is amplitude of the signal
will remain same, but with an 180 degree phase shift right. The phase sensitive
demodulations of the output voltages is necessary if a direction sensitive output is
required in a phase sensitive demodulation will be necessary here right output is
required.
Now, specification let us look at excitation voltage is 5 to 50 volt. Then we have this
typical excitation voltage that is we have excitation we have an output already we have
discussed in excitation frequency its lies between 50 hertz 5 kilo hertz. Now, typical
range depends on the size and all those things we can use this up to 50 hertz to 5 kilo
hertz the sensitivity 0.2 to 5 volt per degree. You can see the sensitivity is enough to get
the same number that is I do not meant the further amplifications whatever you get from
the microsyn the signal is enough high and it is enough high. So, that I can use directly
that signal to make my angle measurement for; obviously, it is and it is not very difficult
I mean to amplify the signal if the signal is even small. Suppose you if it is 0.2 volt; that
means, 200 milli voltage it is not very we can get an amplifier simple the single amplifier
or devices cannot divided the signals.
And I can get a large resolutions of the I am mean I can measure a smaller angular I
mean it is not that we measure a fraction of that I mean degree. But we can do it we can
have a large output voltage. Obviously the sensitivity is quite high nonlinearity is 0.5
percent of full scale scale for plus minus 7 degree of rotation. So, plus minus 7 degree of
rotation this is 0.5 percent of the full scale readings I will get and 1percent of full scale
for ten degree of rotation, because it is a non-linear devices it is a relatively seen that the
2 coils will be there. So; obviously, we will get non-linear device, because the basically
principle dip depends an as well same as synchro only difference is that there is no
winding on the rotor otherwise it is same. So, I will get a sinusoidal signal again I am
giving a sinusoidal signals; obviously, it will be a now null voltage, because I thought I
am saying that the you will see that the voltage 1 and 3 will exactly cancels.
What is that and you have seen in the microsyn that is the soft iron code no winding
nothing to the core. So, the reluctance we are very in this positions reluctance between
these and this and reluctance between this and this will aid each other this will also aid
each other, but it will opposes. So, it will exactly same, so I will get a 0 output voltage
clear, but what will happen t moves then I will get a non-linear output voltage. So, non-
linear output voltage why you simple that between 1 and 3, what will happen if it moves
the reluctance between 1 and because this move is iron ore has moved this core has
moved.
So, reluctance between 1 and 3 will increase and reluctance between this 2 and 4 will
decrease. But though I am saying that at these positions we have exactly the these 2
voltages; that means, when it will in this position reluctance will between this. And this
these 2 coils with this position of the coil reluctance between these 2 coils is exactly
same. Though it is in the figure it will appears to be exactly same, but 2 windings you
know that 2 magnetic surface is very extremely difficult to make exactly identical. So,
there will be always some mismatch. So, though it is an our position; that means, this 2
voltages and this 2 voltage should nullify each other.
But you will get a non zero voltage why that is actually I want to that is to be specified
that is called the null voltage of a microsyn right. Null voltage less than the output signal
generated by, so it is 0.01 degree of rotation now; obviously, it is quite small as you can
see that is actually we are not interested in measurement of such a small angle. But it is
less than the output signal generated by the 0.01 degree of rotation this is extremely
small this null volt. That means, this voltage will be equivalent to a voltage which is we
will get from the microsyn when it is rotated by 0.01 degree clear.
The advantages as a I told you earlier this microsyn has the advantage over the synchro
permanent synchro. But the application areas is very narrow as I told you I mean
compare to synchro it is nothing and very small angular displacement can be measured.
Because there is no sleep dreams, because slip rings will make an missions it will always
allow you to make the problem that is a it will a use additional I mean resistance to the
rotary I mean systems. So that smaller angular measurement extremely difficult very
small angular displacements can be measured with this type of system right. And no slip
rings are required for the rotor excitation in contrast with synchro that is the great
advantage of thesse particular devices.
But especially as I told you it is very slow I mean low torque devices compared to
synchro it is not that popular just some manufacturer company we have used. Only
advantage that I have prominent these are the 2 prominent advantage what we have in the
microsyn micro synchro right. With this I come to the end of the lesson 30 of industrial
instrumentation. Welcome to the lesson 31 of industrial instrumentation. In this lesson
and the subsequent lesson; that means, lesson 31 and 32, we will discuss basically the
dissolved oxygen sensors. So, was not that important I am some time bad, but due to the
evolution of the biotechnology in a very big way.
So, this is to be considered very thoroughly and also I, its application, because the newer
the bio deters are coming up in the markets. And lots of new product are coming like I
mean medicines then antibody which is a basically the I mean the growth of sales. So,
and in this particular I mean situations we must consider the dissolved oxygen sensors.
That is one of the new sensors, which can which can measure the dissolved oxygen
concentration as that the partial pressure of the oxygen in a liquid medium. And this will
give you lot of other informations which is not exactly measurable which usually we
have to estimate.
So, that is a reason this is also this I mean this dissolved oxygen sensors we brought
under the industrial instrumentations. And it has also I mean I should say that the it is
also used as a I mean measurements of the dissolved oxygen in environments likes the in
the water treatment plans. And the water discharged where it is in the I mean, because
there is some safety level of the water I mean dissolved oxygen concentration which is
essential for the living animals like fish and others as we and all those things.
The current output of the electrode is propositional to the oxygen flux at the cathode
surface. Now, NFA Pm will be equal to multiplied by partial derivative of P with respect
to equal to 0; that means, at the cathode surface.
Where N, F, A and Pm are the number of electrons per mole of the oxygen reduced and
Faraday’s constant, it is FA is a Faraday’s constant is a surface area of the cathode,
which is very easy to calculate. Because it is a pi R square or pi D squire by 4 and the
oxygen permeability Pm is the oxygen permeability of the membrane respectively right.
A is the number of electrons per mole of the oxygen reduced A is the Faraday’s constant
A is a surface area of the electrode cathode and Pm is the oxygen permeability of the
membrane respectively. The permeability Pm is related to the diffusivity of the oxygen
by the following expression what is that expression Pm equal to Dm into Sm.
What is a Sm? Let us look at where Sm is the oxygen solubility of the membrane and
from equation 5 and 6 the current output, it the current output. Because actually that is
only measurement that is reason, we are going for the permanent membrane based
electrode membrane covered electrode. Otherwise, we can make the chemical analysis
we need this current output for continuous measurements or continuous control of the
partial pressure of oxygen or dissolved oxygen concentration either in the bioreactors or
in the waste water treatment plant. As function of time will be expressed as follows this
is equal to you see the current output it equal to NFA Pm by Tm multiplied by Po 1 plus
2 summation n equal to 1 infinity minus 1 to the power n exponential minus n square pi
square Dmt upon Tm square this is equation number 8.
In IIT Kharagpur say in the bioprocess institution lab and we found that we got a
consistently good results though the result in consistent obviously, but when you
measure with the other sensors. So, there is slight options that which is quite obvious. So,
this galvanic electrode actually it worked very nice in our laboratory right. So, with this I
come to the end of lesson 31 of industrial instrumentation. Basically, we have considered
the dissolved oxygen sensing here and basic more details of the sensors 4 different will I
will more details of this sensing will come in the lesson number 32.
Prof. A. Barua
Lecture - 31
Dissolved Oxygen Sensors – I
Welcome to the lesson 31 of industrial instrumentation. In this lesson and the subsequent
lesson; that means lesson 31 and 32 we will discuss basically the dissolved oxygen
sensors. Dissolved oxygen sensors was not that important I mean some time bad, but due
to the evolutions of the biotechnology in a very big way. So, this is to be considered very
thoroughly. And also it has a tremendous application, because the newer the biodetectors
are coming up in the markets. And lots of new product are coming like I mean medicines
then antibody which is a basically I mean the growth of cells.
So, and in this particular I mean situations we must consider the dissolved oxygen
sensors. That is one of the only sensors, which can measure the dissolved oxygen
concentrations and the partial pressure of the oxygen in a liquid medium? And this will
give you a lot of other informations which is not exactly measurable which usually we
have to estimate. So, that is a reason this is also this I mean this dissolved oxygen sensors
we brought under the industrial instrumentations.
And it has also I mean I should say that it is also used as a I mean measurements of the
dissolved oxygen’s in a environments likes the in the water treatment plans. And the
water discharged where it is in the, because there is some safety level of the water. I
mean dissolved oxygen concentration, which is essential for the living animals like fish
and others and all those things that is why that is the reason we have taken this. Now, let
us look at the contents of this lesson 31 dissolved oxygen sensors.
The contents are in this particular lessons we will basically we will see this is a probe.
Actually the signal conditioning circuits all those things will be discussed later on. So,
basically like a ph probe we will discuss that is of oxygen probes. We have anode and
cathode we will see all this in details we will discuss at various construction. What is the
theory behind this? What is the, I mean chemical equations inside this, all these things
will be discussed typical constructions dimensions all those things will be discussed.
Polarographic electrode; there are 2 types electrodes we will find one is a polarographic
electrodes another is galvanic electrode one need power supply other do not need. So, we
will consider the one which has which usually not I mean does not need any power
supply or a bias that is called the galvanic electrode. Then basically we will consider
theory of operation the detailed constructions of these electrodes will be discussed in the
subsequent lessons; that means in the lesson 32.
At the end of the lesson the viewer will know the working principles of the dissolved
oxygen sensor. Electrodes of the dissolved oxygen probe; what are the different
electrodes? 1 layer and 3 layer model we will find that there are 2 types of model
available. One is called 1 layer model of the dissolved oxygen sensor another is a 3 layer
model of the dissolved oxygen sensors. Then we will discuss the reaction at anode and at
cathode of the electrodes. So, there are chemical reactions which actually occur in the
both anode and cathode these also will be discussed in details.

Introductions, let us see the dissolved oxygen sensors or electrodes have been widely
used both in research and industry. In the academics also we are using this probe to
measure the dissolved oxygen concentration and also industries also where they are
producing some biotechnical product, biotechnological product like medicines. And all
those things like your antibody then antibiotics all those things you will need this type of
sensor. Now, compared with the chemical analysis, now other way you can measure the
dissolved oxygen concentrations or partial pressure is a chemical analysis. That means,
you take out a sample and you take the make the chemical analysis.
There is lot of limitations of that type of I mean chemical analysis, because what will
happen? That you have to make offline you would not get it is not very easy to get
electrical output. So, if you want to control the dissolved oxygen concentrations inside
the liquids you cannot do it. So, in that case this I mean the membrane based electrode
which is called the, or membrane covered electrode offer the several advantages. That is
the reason you can compare with the chemical analysis the measurements of dissolved
oxygen’s in water by membrane covered electrode offers several advantages.
They are as follows what are those? Simplicity the design is very simple less interference
by other solutes in the water it does not matter with liquids. Because if you have
chemical analysis if you want to measure the partial pressure of oxygen or the dissolved
oxygen concentration. So, they in turn if we have some I mean other solutes that will
also react with your sample. So, that will create problem and you would not get the exact
value. In-situ measurement with lower time constant, you known in-situ measurement is
basically new terms in the instrumentation that is basically means on the site
measurements. In their wire reactors you will find that there are various parameters
which usually we measure we cannot measure. But those also when you write I mean
straight equations you will find that those parameters are to be incorporated. One of
those is like concentrations I mean dissolved oxygen I mean concentration is 1.
Then we have a concentration of the product, concentration of the subtract all those
things will be unmeasurable. It is you cannot measure those quantity only you can online
measurement is not possible you have to take out and there is there are few in-situ
sensors have been reported, but it is not much of use. So, only parameters which you can
measure are the dissolved oxygens concentrations by the electrical output. So, it is an in-
situ measurement that is on the site you can immediately measure you do not have to
make a chemical analysis of your sample. Continuous measurement is possible whenever
you can make the continuous, because it is the electrical output.
So, the continuous measurement also I mean will be there also accordingly if the
continuous measurement you have then. Obviously, you can do you can make the control
also. So, that you can precisely control the dissolve oxygen I mean in a particular
medium or liquid. Real time control that is I just I have said the real time control of
dissolved oxygen concentration in bioreactors or waste water treatment plant is next this
I was talking what you see there are many places the water are discharged. Now, if you
discharge those water as it is in a I mean environment that will create, because the plants
and animals marine animals. They need some sort of oxygen in a in a when they survive
want to survive in a particular liquid.
There is a particular concentration that is the reason you find that if that concentration
goes down the fish and other animals will die. Now, once you discharge that waste
water. So, in an environment; that means, in a swamp or something like that. So, upon
you must treat that you must know that what are the dissolved oxygen concentrations or
partial pressure of oxygen’s in that particular liquid. In that case also we need some in-
situ measurement. So, that we can take care and make some artificial I mean increase of
dissolved oxygen’s in that particular liquid of that waste water plant. And bioreactor as I
told you it is an, I mean vital parameter which is only measurable part not only that it is
only measurable parameters in the bioreactors.
Bioreactors actually I should define as it is basically it is basically a medium or reactors

where the biological cells are grown. That is the reason we are calling instead of calling
simple reactors you know out of chemical reactors we are calling bioreactors. Because
cells are grown it does not matter mater whether it is animal cell or simple yeast these are
all bioreactors. Now, dissolved oxygen sensors have been developed for different areas
to meet the requirements of the specific applications. There are various applications and
accordingly the people need different types of dissolved oxygen concentration
measurement. The examples are the steam sterilizable DO 2 probe suitable for
bioreactors. Now, you know the bioreactors if I take an example of the bioreactor it looks
like this if I take a white page it looks like this.
I have a reactor here this is a star tank reactors I have a starrer. It consists of blades 3 or 4
5 blades. So, it will rotate to look like this there are several blades. Now, this is called
impeller of the bioreactor, this is a tank in which we have medium. So, tank will look
like this 1. So, tank is liquid is there and it is continuously rotating either in clockwise or
anticlockwise directions. If it is there, so what we find that there will be some eddies will
be formed which will look like this. Like this 1 some swirl will be formed this is eddies,
this is tank and this is swirl. Now, it is continuously rotating and if you increase the
rotation. So, what will happen? This is called CSTR, CSTR is the continuous star tank
reactor. This is called the continuous start tank reactors we can see here this is called the
continuous start tank.
Reactors are various type you can have a bubble column reactors you can have a CSTR it
is a continuous start tank reactors. Now, the basic principle is nothing like if you increase
the blades blade speed by, because this impeller is rotated by a motor. If you increase the
motor speed what will happen? That you see the oxygens is very difficult one thing you
must know that it is very difficult to dissolve oxygens in a liquid. Now, only place is
where the oxygens will be will be dissolved is through this through this top surface. To
the top surface and top surface whenever these oxygens are getting in contact it dissolves
and it is this top surface is getting saturated with oxygen.
So, this top surface is to be refreshed; that means, this lower surface of the water will go
to the top like this 1. So, that it will get more and more oxygen. Now, if we if we
increase the speed of this impeller what will happen in this CSTR continuous star tank
reactor? There is a frequency of this water, which is coming from the top from bottom to
above the refreshing I mean of this water top surface will be faster and faster. If it faster
then what will happen? That more and more oxygen will be dissolved if you lower the
speed. So, the dissolved oxygen concentration also will be also, will decrease. This is a
basic principles of the bioreactor, because of cell which will go inside the reactor
depends on the oxygen concentrations.
It is does not end here, because these dissolved oxygen concentration is a very vital
parameters in any bioreactor. Because that you know that the cell growth by if I look at
the cell growth that whenever if I continuously measure this dissolved oxygen
concentration. So, you will find that initial stage what will happen that? Initial stage
there will be good consumptions of the dissolved oxygen. So, the cell concentration that;
that means, the dissolved oxygen concentration will fall slowly fall, because it is more
and more cell is getting. Now, after some time we will find that the this concentrations is
getting saturated that, That means, at during that time the cells are no more no more
consuming any consuming any oxygen from the liquids.
So, the health of the cell; that means, the growth of the cell can be predicted also from
the dissolved oxygen sensor the output of it might be the current or voltage. Now, after
some time we will find that the concentrations are getting increased; that means, during
that time thus it is cell is totally saturated. There are no more cell will be grown that is
the time we should take out the cell from this biological cells from the reactors. So, I can
see that the dissolved oxygen plays a key role not only to maintain the growth of the cell.
But also to tell that whether the entire reaction has total I mean is complete also the time
when you should take out the cell from the reactor itself. So, these will this is very
important and this is the only parameter as I told you I mean in the bioreactors which is
measurable.
Now, this I mean if you refresh these things I mean continuously what will happen? This
I mean it will be getting newer and newer surface on also you can do one thing. That
means, if there is if there is oscillation of this liquid columns we will find at the
dissolved oxygen also will oscillation of the liquid columns. We will find that; that
means, suppose if you have a liquid here in the two columns, we will find if you may by
force if you can make the oscillations. So, that the liquid the dissolved oxygen will be
also getting the oxygen’s will be dissolved in the liquid through the falling stream
through this one. There is another way there are also bubble column reactor then bubble
column reactor, what will happen you know if I take a new page.
The oxygens are passed through a pipe and it is going out rightly this 1 through in the
liquid. So, here what will happen? Again you will find that the oxygens are getting
dissolved through this bubble column? If there is a various constant I mean various
method of desiccating I mean getting oxygen dissolved. But you have to measure that is
either through to get a continuous measurement either you have to make through a
galvanic sensor or through a polarographic sensor.

So, that is called as to now, this reactor is you know you have to know basically it will be
sterilized. So, this DO 2 probe there are some DO 2 probe which is suitable for making
sterilizable; that means, this probe is to be sterilized how will you sterilize? You will I
mean at the, you have to be when you heat the entire reactor along with the along with
the probe DO 2 probe at 120 degree centigrade which by superheated steam you have to
do. That means from the boiler you have to pass that team. So, that the oxygen will be is
a sterilized.
Oxygen micro electrode for DO 2 measurements in human tissues then a fast responding
sensor for respiratory gas analysis, because in many cases the patients need to feed the
oxygen. So, that during that time; that means, you must know that, because pure oxygen
is very difficult there is some problem form of impurity is always there. So, how much is
impurity also you can know from the dissolved oxygen. Measurement of trace of oxygen
in boiler feed water because boiler feed water is also necessary in a thermal power plant
the feed water is how much oxygen is there. So, that also can be online measurement is
possible through this DO 2 probe.
Three broad areas I should say where the measurement of DO 2 concentration is

essential are biochemical engineering then microbiology biochemical engineering.
Obviously, includes the bioreactor microbiology and environmental engineering; that
means, waste water treatment plant and all those things. So, there also we need to
measure the dissolved oxygen concentration. So, the partial pressure of the oxygen in the
waste water which is getting discharged to the environment.
Now, principle of operation there we have polarographic electrode as I told you first will
discuss the polarographic electrode then will discuss the galvanic electrode principle is
basically same one need bias other do not need. When an electrode of noble metal such
as platinum or gold is made 0.6 to 0.8 volt negative with respect to the reference
electrode made of calomel or silver chloride in a neutral potassium chloride solution. The
dissolved oxygen is reduced at the surface of the cathode. This is a basic principles of the
a dissolved oxygen sensor it does not matter I mean whether it is polarographic or
galvanic.
This phenomenon can be observed from a current voltage diagrams called a polarogram
of the electrode. This is called the polarogram we will see the polarogram very soon. As
shown in figure 1 a, the current increases initially with an increase in negative bias. Then
it reaches a region where the current becomes essentially constant. It will be very clear
from the next slide. In this saturation region or the plateau region of the polarogram the
reaction of the oxygen at the cathode is so fast that the rate of oxygen reaction is limited
by the diffusion of the oxygen to the cathode surface.

This is our Polarographic electrode you see this is a polarogram sorry polarogram you
see here. This is you see this is a negative bias we are giving as we increase a negative
bias you will find that the output relative current output will increase. So, at that some
position we will find there is saturation. And after there is a sudden increase that is a
different thing we will find there is saturation or a plateau region. So, a percentage of
oxygen, so the partial pressure of oxygen it is increased as you can see.
So, the, it is 1.5 as you increase the bias you are getting the saturation then it increases. It
is a with the concentrations of the oxygen is 7 percent and if you increase the bias
initially it will make a linear then it will be at saturated like this 1. Now, in the
subsequent usage this is a calibration curve. So, I have plotted here the bias voltage here
bias voltage is 0.7 volt I have plotted the percentage oxygens on the x axis and I am
getting a relative current. So, it is almost linear curve as you can see this is called the
calibration curve.
When the negative bias voltage is further increased, that the current output of the
electrode increases rapidly due to the other reactions let us go back.
You see as shown in figure the current increases initially with an increase in the negative
bias is not it as you increase the negative bias increase the negative bias current increases
then it reaches a region where the current becomes essentially constant. In this saturation
region the plateau region of the polarogram the reaction of oxygen at the cathode is, so
fast that the rate of reaction is limited by the diffusion of oxygen to the cathode surcfae.
That is a reason I am getting a plateau region or a saturation region.
This you see as you increase in the bias it slowly increases then it is, so fast that is why I
am getting a saturation region here same as you increase a bias for a particular
concentration negative bias it increase. Then it is gets to a plateau region after that is a
slight there is a drastic increase of the current. When the negative bias voltage is further
increased the current output of the electrode increases rapidly due to other reaction
mainly the reduction of the water to the hydrogen. So, that is a different thing we are not
concerned with that region.
Now, if a fixed voltage in the saturation region of the polarogram is applied to the
cathode. Then the current output of the electrode can be plotted for different dissolved
oxygen concentration that is figure 1 shows such a calibration curve. What is it saying
that? The fixed voltage in the saturation region of the polarogram is applied to the
cathode fixed voltage. I have given in the previous curve you see we have given a fixed
voltage of 0.7 volt right cathode.
Then the current output of the electrode can be plotted for the current dissolved for the
different dissolved oxygen concentration. Current outputs are plotted on the x axis
dissolved oxygen. Now, percentage partial pressure of the oxygen in the figure b shows
that which is called the calibration curve of a polarographic electrode for dissolved
oxygen measurement. It has to be noted that the current is proportional not to the actual
concentration, but the partial pressure it does not matter say the partial pressure is the
high the concentration also will be high.
But due to the activity or equivalent partial pressure of the dissolved oxygen which is
often referred to the oxygen tension. A fixed voltage between 0.6 to 8 volt is usually
applied as the bias voltage or polarization voltage when using silver chloride as the
reference electrode.
When the cathode comma anode and the electrolytes are separated from the measuring
medium with a plastic membrane which is permeable to gas, but not to the most of the
ions and when most of the mass transfer resistance is confined in the membrane the
electrode system can measure the oxygen tensions in the various liquids. So, what is the
main, what is the key a plastic membrane which is permeable to gas, but not to the most
of the ions.
So, the gas will be permeable through this medium and through this membrane and when
most of the mass transfer resistance is confined in the membrane the electrode system
can measure oxygen tensions in various liquids. This is the basic operating principle of
the membrane covered polarographic DO 2 sensor. This is the basic requirement or basic
operating principles of the membrane covered polarographic DO 2 sensor in both the
cases we will find we may use the membrane we will be clear after some time.
A polarographic electrode is shown in figure 2 for polarographic electrodes the reaction

are as follows. I will show the figure 2 after sometime reaction at cathode O 2 plus 2 H 2
O electron gun plus 2 electron is equal to hydrogen peroxide plus hydroxide H 2 O 2 plus
2 electrons it is 2 hydroxide reaction at anode AgCl it will be AgCl plus e 1 Cl is
negative discharge. So, that it will release 1 Cl and will get AgCl with 1 electron.
Overall reaction 4 Ag plus O 2 plus 2 H 2 O plus 4 chlorine equal to 4 AgCl plus 4

hydroxide iron. The reaction produces alkalinity in the medium of quite; obviously, you
can see 4 OH. So, it will make it alkaline and a small amount of hydrogen peroxide is
also produced that we have seen in the previous.
Now, this is our Polarographic electrode you can see this electrode it is basically a
cylindrical in nature. You see I have an ammeter which is basically a micro ammeters we
have a, we have a cathode there is a membrane this is a membrane you can see this is our
membrane. You can see that this is our this is a membrane permeable membrane the gas
can be permeable most of the irons will be blocked here, because this entire thing please
note will be dipped in the reactor itself or in the waste water. This is a tank or the reactor
tank or in the waste water this entire thing is be dipped entire assembly. So, this whole
assembly this is our probe this is our total probe. You see electrolyte is here we have put
anode here insulation is here there is a membrane and there is a cathode.
This is the diagram and we have given a bias we have seen that if we increase bias;
obviously, that I mean output will increase with the concentration. Now, for a fixed bias
what will happen if I; obviously, I will for different increased concentrations. I will get
the higher relative current outputs, which will be utilized for the measurement or for the
control of the I mean dissolved oxygen partial pressures in the liquid. Now, if I draw the
actual diagram it will looks like, because you see this electrode as I told you it should be
combined in the tank and this entire electrode. This entire electrode actually will find that
it will be in the form of it will be in the form it is a very handy things and it will be in
basically cylindrical in-situ let us look at the diagram.
Let me take a white page it will look like this one see it will look like this you see that a
cylindrical in shape this entire thing. So, it will come down here sorry it will now, we
can take a new page and see it will come down here and we have a probe here and go
down there is an anode cathode this is cathode. So, entire thing like this one, so there is
another one it looks like this one the membrane is here membrane is here. We need as
better supports and all these things that is I am not discussing because it cannot hang in a
air as you know you need a support and this will work as a your anode and this will work
as a cathode of your systems. And we have a glass cylinder the entire thing we put on a
glass cylinder it looks like this. It is coming like this I mean a glass cylinder like this one
it will look like this.
There are various dimensions I mean typically in electrical surface we made it like this
one and we put Teflon here now, it is Teflon is very extensively used which is 100
micron thicknesses. This is this diameter is 2 centimeter sorry this diameter is 2
centimeter and this diameter is 1.4 centimeter 4 centimeter like this 1. And this is quite
small this diameter is 0.75 centimeter. These all the dimensions it will go to anode and
this will also, will get a signal here. I am not showing the bias, because the galvanic also
has a same type of forms and this can be made this can be made or anode can be made of
lead this lead and cathode can be made of silver. These are polarographic electrodes
insulation we have membrane here where I can use Teflon here.
This is a galvanic electrode. So, this is we have shown that is an ammeter there is no bias
electrolyte is here anode insulation membrane and cathode.
Galvanic electrode the galvanic electrode is shown in figure 3. The galvanic electrode is
different from the polarographic type in that it does not require external voltage source
for the reduction of oxygen at the cathode that is the basic difference.
When a basic metal or base metal such as zinc lead or cadmium is used as the anode and
a noble metal such as silver and gold is used. And the cathode the voltage generated by
the electrode pair is sufficient for a spontaneous reduction of oxygen at the cathode
surface. In IIT Kharagpur actually in our Bioprocess Instrumentation Laboratory, we
have developed a DO 2 senses, which is giving quite good response where we are using
lead as an anode and silver as a cathode.
The electrode reaction of the silver-lead galvanic probe is as follows. Reaction at cathode
an O 2 plus 2 H 2 O plus 4 e 4 electrons equal to 4 hydroxyl ions. Reaction at anode Pb
to it is I mean positively charged then and 2 electros, so look like this. Overall reactions
will be O 2 plus 2 Pb plus 2 H 2 O equal to 2 Pb O H 2.
As shown above the oxygen is reduced via 4 electron reaction this reaction is called 4
electron reaction. And unlike the Polarographic probe these electrolyte does not
participate in reaction, but the anode is gradually oxidized, that is one problem with this
unlike the polarographic probe the electrolyte does not participate in the reaction, but the
anode is gradually oxidized. Therefore, the life of the probe depends on the exposed
surface area of the anode.
Whether the polarization voltage is applied internally in the case of galvanic or

externally which is applied in the polarographic the operating principle of the electrode
remains the same. Operating principle same whether you applying a galvanic I mean I
mean polarographic in the case we are giving it externally in the case of galvanic it
generates internally. There are some gases, which reduce in the presence of 0.6 to 1 volt
in the test medium. Examples are halogens which is chlorine bromine and iodine and
oxides of the nitrogen it will lead to erroneous output in the measurement systems. So,
you have to be very careful if you have the presence of this type of gas in the liquid
where we are interested to measure the dissolved oxygen.
Theory of operation the basic principle of measurement for the membrane covered D O 2
probes can be summarized as follows. If the oxygen diffusion is controlled by the
membrane covering the cathode, the current output of the probe is proportional to the
oxygen activity or the partial pressure in the liquid medium basic principle is same.
The following assumptions enable a mathematical analysis of the pressure profile of

oxygen in the liquid and the current output of the dissolved oxygen sensor. What are
those assumptions? Let us look at the cathode is well polished and the membrane is
tightly fit over the cathode surface. So, that the thickness of electrolyte layer between the
membrane and the cathode is negligible. So, that we are assuming there is no there is
absolutely no layer of the electrolyte in between the cathode and the membrane usually
which is deformed membrane. The liquid around the probes will start, so it is mostly in
the in the case of I mean whether you are measuring I mean it is it is a well start liquid.
Whether you are using in the bioreactors or the waste water I mean it is a well start; well
start and agitated. So, that the partial pressure of oxygen and the membrane surface is
same as that of the bulk liquid, because we will need these assumptions otherwise very
difficult to calculate the establish the basic the current equations.
So, the oxygen at the membrane surface is the same as that of the bulk liquid. At present
we consider that the oxygen diffusions occur only in one direction that is perpendicular
to the cathode surface. We are assuming in the one direction only that is perpendicular to
the cathode surface. It is, so called 1 layer model, but it can be modified to include the
effects of other layers as will be discussed later.
Now, suppose the electrode is immersed in a well-agitated liquid and at time zero the
oxygen partial pressures in the liquid is changed from the 0 to Po. According to Fick’s
second law the unsteady state diffusion in the membrane is described as delta p del p by
del t. That means partial derivative of p with respective to t equal to D m D subscript m
del square p by del x square double derivative of partial derivative of p with respect to x
this is equation number 1.
Where Dm is the oxygen diffusivity in the membrane and the Teflon membrane x is the
distance of cathode surface it will be clear from the figure 4. What is actually fourth the
initial and the boundary conditions are p equal to 0 at t equal to 0, initially there is no
partial pressure P equal to 0 at x equal to 0 at the cathode surface. There is no p equal to
p naught at x equal to Tm, T is a thickness of membrane, please note T is thickness of
membrane. So, we start at the 0 and we end it T equal to Tm x equal to Tm.
So, this is a 1 layer model. So, we have a as we have shown that is we are considering
the perpendicular the cathode surface this is our cathode surface. So, we are considering
at X equal to Tm P note this a thickness of the membrane, because you remember that
the reactors are looks like this, I mean sorry the membrane I mean a membrane here is
not it.
So, we have the cathode here, so perpendicular to cathode surface. So, as we are going
like this 1 this directions that is the this is x equal to 0 and this is x equal to this x equal 0
and this x equal to Tm; that means, thickness of the membrane. So, we are considering
that this needs to be inserted in a tank or liquid medium X equal to Tm. So, X we are
from the cathode surface we are getting this is our thickness of our membrane this is
again we are coming to the liquid this is a 1 layer model of the electrode. So, we have a 2
layer model or 3 layer models also we have electrolytes. So, we are assuming there is
absolutely almost there is no electrolytes in between this cathode and the membrane
surface that is actually we are making the assumptions.

Where Tm is the membrane thickness and the first boundary condition, which is equation
3 assumes very fast reaction at the cathode surface. And the solution of the equation 1
with boundary conditions will be given by P by p naught x by Tm summation n equal to
1 to infinite 2 by n pi into multiply by 1. I mean 1 to the minus n sin of n pi x by Tm
exponential minus n s of n square pi square Dm t by Tm square.
The current output of the electrode is propositional to the oxygen flux at the cathode
surface. Now, NFAP m will be equal to a multiplied by a partial derivative of p with
respect to x at x equal to 0; that means, at the cathode surface.

Where N F A and Pm are the number of electrons per mole of the oxygen reduced and
Faraday's constant it is the F. F is the Faraday’s constant is the surface area of the
cathode which is very easy to calculate. Because this is pi r square or pi D square by 4
and the oxygen permeability Pm is the oxygen permeability of the membrane
respectively. N is the number of electrons per mole of oxygen reduced F is the Faraday’s
constant A is a surface area of the electrode cathode and Pm is the oxygen permeability
of the membrane respectively. The permeability Pm is related to the diffusivity of the
oxygen by the following expression what is the expression? Pm equal to Dm into Sm
what is the Sm? Let us look at…
Where Sm is the oxygen solubility of the membrane and from equation 5 and 6 the
current output, it current output, because actually that is only measurement that is a
reason we are going for the membrane based electrode. I mean powered electrode
otherwise we can make the we need this current output for continuous measurement or
continuous control of the partial pressure of oxygen or dissolved oxygen concentration
either in the bioreactors or in the waste water treatment plant. As function of time will be
expressed as follows this is equal to you see the current output I t equal to N F A Pm by
Tm multiplied by Po 1 plus 2 summation n equal to 1 to infinity minus 1 to the power n
exponential minus n square pi square Dm t of a Tm square. This is equation number 8.
The pressure profile and the current under steady state conditions can be obtained
nothing, but what will happen this steady state condition this time will go this time will
be 0, because a steady state conditions. Because it is a negative exponential power then
what will happen? If the as the time grows, so this will be larger and larger, because it is
1 of this. So, this will become this. So, it will be 0 at the time goes. So, only this portions
will remain that is the steady state current output. And the pressure profile and the
current under steady state conditions can be obtained from the equation 5 and 8 let us
look at what is 8 and 5?
So, we are combining 5 and 8. So, we are getting that p by p naught equal to x by Tm.
And I as steady state current you see that I t is the total current or transient current I
should say not total current transient current equal to N F A Pm by Tm into P naught
equation number 10. At steady state the pressure profile in the membrane is linear and
electrode current is proportional to the oxygen partial pressure of the bulk liquids. So,
equation 10 forms the basis for DO 2 probe measurements. What is equation 10? This is
equation 10. So, this is this is for all DO 2 measurements or DO 2 probe.
Another important consideration in the time response or the time constants of the probe,
how we can make the time constant lower and lower? Because you see it if the time
constant is large though the bioprocess usually I mean we do not need very fast
measurements a typically we can make 5 to 6 minutes most of the reaction are lying.
Because you see the reactions in the bioprocess usually is quite long I mean sometimes it
30 hours I mean 40 hours.
So, you do not needs like a I mean very fast sampling I mean you do not have to make
the measurement very fast. You can measure in the 6 minutes 10 minutes, but still that I
mean you must consider the time constant of the response time of your DO 2 probe.
These important issues are important parameters of a probe. According to equation 8 the
probe response depends on the probe constant K defined as follows. That K equal to bi
square into Dm by Tm square equation number 11.
A large K, which means a thin membrane and at a high Dm divisibility large results in a
low value of time constant quit. Obviously, a large K which means a thin membrane is
thin and diffusivity is high for the membrane is high result in a low value of time
constant. So, this will make the time constant low and the consequence of the probe
response will be very fast.
However, the assumptions of the membrane controlled diffusions will be more valid in
these conditions. It is very contradictory, so way we are reducing the time constant time
constant also always should be lower for any instrumentation system for any sensors
lower is better. Thus a compromise has no has to be made for an optimum performance
of the DO 2 probe. Thus, a compromise is to be has to be made for an optimum
performance of the DO 2 probe.
Limitations of single layer electrode model in reality the assumptions one and 2 made
earlier are not entirely satisfactory. So, a made some assumptions at the beginning to
establish or to build the theory are not entirely satisfactory often there exists a finite
thickness of electrode layer between the cathodes. We have assumed that the there is no
electrolyte layer in between cathode and the membrane, but there is finite layer of
electrolyte in between cathode and membrane. And the membrane because of the
roughness of the cathode surface though we clean that the cathode if the cathode surface
are membrane is absolutely plane. So, there is no chance of I mean liquid electrolyte
coming in between, but if there is slight roughness.
So, a layer of liquid or a layer of electrolyte will come in between membrane and the
cathode. So, that is to be taken care in the in the single layer model we are not taken we
have assumed that that is zero thickness of the electrolyte is zero in this that region. Also
a stagnant liquid film always exists outside the membrane even at very high liquid
velocity. Even though we had a very good start I mean, because in the starting you are
doing you see in the reactors. Now, usually the there will be a liquid I mean because it is
inside the liquids it will be shorted I mean. So, always there is a stagnant liquid film
always exists outside the membrane and may not very high liquid velocity. That means
liquid a liquid which is contact the membrane we are consuming assuming that it is a
liquid film exists which is stagnant what is that look at very carefully it will look like this
1.
So, we are way you see the probe is like this one we are way probe here membrane. So,
we are way cathode here and this we are inserting insight the liquid is not it is our
membrane this is our cathode this is connected like this one. First of all, we are assuming
that this cathode there is a layer. That we which we have assumed that the layer of thin
liquid d is I mean it was electrolyte does not present in the in the in the case of in the
case of 1 layer model that is not true. If it is there is a slight roughness electrolyte will
come in between. Secondly the film even though it is rotating I mean this we have a
agitators either in the continuous CST; that means, continuous start time by reactors or
your other type of reactors what will happen? That there is a continuous I mean moment
are agitations of the liquid even though we will have a thin layer in contact with this one
will be stationary or stagnant thin layer of liquid in contact with the membrane.
A more realistic model of the electrode is shown in figure 5, which is called basically 3
layer model. All the 3 layers namely the electrolyte, the membrane and the liquid film
have been considered all the 3 layers have been considered.
You see they are we have considered this is our electrolyte in between this is our cathode
this is our cathode. I can take this one this is our cathode we have a membrane after that
we have a I mean a a definite thickness of the electrolyte. Then we have a this is a
membrane thickness Tm; this is a membrane thickness and this is our liquid film
thickness. You see it is slowly increasing. So, like this one it is increasing you see here
after that we are getting the main liquid bulk liquid. But here this is a film thickness
which is in stagnant with the with the membrane itself. So, may be larged though this is
very, very small in the order of excuse me in the order of micro meters. But just to
explain to the students we have I mean elaborated it like this one.
The effect of different layers on the electrode behavior can be estimated by using the one
layer model. At steady state the oxygen flux J through the each layer at figure 5 becomes
identical. But J equal to K naught into P naught Klm P naught minus Pm Km multiplied
by Pm minus Pe which is equal to we can write Ke into Pe which is equation number 12.

Where K 0 is the overall mass transfer coefficient and small k's represent individual mass
transfer coefficient corresponding to the liquid film Klm the membrane Km and the
electrolyte Ke respectively. The overall mass transfer I mean resistance one by K 0 is
then expressed as the sum of the individual resistances, which can be given like this one.
1 by K 0 1 by Klm is a liquid film the master transfer of coefficient 1 by Km membrane
mass transfer coefficients and 1 by Ke electrolyte right.
Equation 13 can be rewritten by using the oxygen permeability and the thickness of each
layer as follows. We converting layers 1 by K 0 equal to Tl by Pl Tm by Pm Te by Pe,
where Tl Te Pl Tl already we have defined it is a thickness of the membrane Tl Te Pl.
And Pe are the liquid film thickness this the liquid film thickness this the liquid film
thickness this is the electrolyte thickness. The oxygen permeability of the liquid film
liquid film only and that of the layer itself this is of the electrolyte itself this Pe is clear.
A completely stagnant liquid film was assumed here although it is more accurate to use
the convective mass transfer coefficient. This is typically more complicated other
convective mass transfer coefficient and the condition for membrane controlled diffusion
becomes that Tm by Pm will much, much greater than Tl by Pl plus Te by Pe equation
number 15.
This means that a relatively thick membrane with low oxygen permeability is required
which contradicts the requirement for a lower time constants of the probe. That justify I
mean you can be just this can be justified, because in the most of the, I mean this
dissolved oxygen measurements. We do not need very fast response for I mean controls I
when if you can measure only 6 minutes or every 10 minutes that will suffice. Same
thing with the waste water plant treatment plants also we do not need that faster even if
we measure. If we have a sensor, which can measure the dissolved oxygen I mean partial
pressure of the oxygen or dissolved oxygen concentrations very frequently every second
that does not make any sense. Because you need to control if you want to control that
will increase the dissolved oxygen concentrations in the particular liquid. The
mechanisms, which we have to follow that we increasing the starrer filed or any other
oscillations or anything it will take a large time.
So, there is no use of measurements of very fast I mean. So, that is I mean little large a
time constant will be justified if you have a proper model in that sense. So, that is most
important this means that a relatively a thick membrane with a low oxygen permeability
is required which contradicts the requirement. Because if you make the thick membrane
low oxygen permeability; obviously, what will happen that constants of the system will
increase. For a given cathode geometry the resistance of the electrolyte is almost fixed
for the resistance of the electrolyte is fixed that is not a very problem right for a
particular geometry of the cathode. Also since the also since the electrolyte is contained
inside the membrane it does not affect the measurement.
Therefore the condition for accurate measurement of dissolved oxygen becomes the Tm
by Pm plus T by P much, much getter than Tl by Pl which is equation number 16. When
the individual resistances are taken into account the steady state current output can be
written as when the individual resistances are taken into account. Then steady state
current output can be written as Is steady state current NFA Pm by T bar into P naught
this equations are already we have written we have just modified according to this
expressions right.
Where T bar is defined as the T bar equal to Tm plus Pm by Pl into Tl plus Pm by Pe

into Te that is equation number 18. In this case the probe constant k is modified as
follows we write K equal to bi square into Dm diffusivity of the membrane often T
subscript small T dash square with this equation number time 19 what is known t Tt, let
that will be clear from the next expression?
Where T subscript small T bar equal to Tm plus under the square root Dm multiplied by
Tl over Dl plus under the square root Dm of Te multiplied by De equation number 20.
Equations 17 and 19 and show that the steady state current decreases and the probe
response time increases when there is a significant mass transfer resistance in the liquid
film around the membrane which is quite obvious, right.
Normally probes are operated such that the effect of liquid film resistance is negligible. It
is achieved by using membranes of low oxygen permeability and by the well agitation of
the liquid around the probe. So, there is I mean all this compromise you have to make. In
fact, actually we in the IIT Kharagpur, we have developed a DO 2 sense now DO 2
sensing if you look at. So, actually it looks like this if you if we look at the complete
instrumentation diagram.
So, it will look like this I have a DO 2 probe sorry I have a DO 2 probe I am getting a
current. Current output is steady state current this current is converted to the current
voltage converter current to voltage converter in the near way a voltage amplifier
amplifier. Now, it is easier to make then we again we can convert to the difficult current
because these amplifications will make us to the proper voltage because most of the
convertors. So, we have a way V to I convertor here because once it is current to its
current to voltage this can be difficulty. I will get the voltage usually typically the current
voltage is 0 to all the convertor available for 0 to 5 volt which respect which will give
you 4 to 20 mille ampere.
So, this can be done here itself; that means, 0 to 5 volts this will give you now I am
getting a current output which is 4 to 20 mille ampere which is used for the
measurements as you know industry that 4 to 20 mille ampere is a standard. So, you will
get like this one Moreover you see that we have used a I mean a plot what we found that
we basically made we made in a this measurements you made in a bioreactors. In IIT
Kharagpur in the bioprocess instrumentation lab and we found that we got a consistory
good results though the result in consistent; obviously, but when you measure with the
other sensors. So, the galvanic is slight absent that which is quite obvious.
So, this galvanic letters actually it worked very nice in our laboratory right. So, with this
I come to the end of lesson 31 of industrial instrumentations. Basically we have
considered the dissolved oxygen sensing here and basic more details of the sensor 4
different will more details of the sensing will come in the lesson number 32. Welcome to
the lesson 32 of industrial instrumentation. This is a continuations of lesson 31 actually
the we will consider here the dissolved oxygen sensors. We discussed dissolved basic
principles of the dissolved oxygen sensors in details in the lesson 31. This particular
lesson will consider the different types of electrodes its calibration type of membranes.
These are the very important things for designing any D O 2 probe and type of is I mean
type of materials which is used for making the for making the electrodes specially the
electrolytes.
And what should be the suitable nature of the membrane also that is also very important.
Because in some cases we have to sterilize the dissolved oxygen sensors when it is used
in the bioreactor or you will put in the auto clamp. So, the temperature minimum
temperature there will be 120 degree centigrade. So, in that type of situations we will
discuss we need some sort of some sort of auto cleaving and with the or some sort of
sterilizations. And these particular lessons we will discuss all those in details and we as I
told you earlier that is it is a continuations of 31. Let us look at the contents of this lesson
32 dissolved oxygen sensors 2.

Contents: construction of electrodes; all the electrodes what is there, what is anode, what
is cathode, what type of materials used for making anode and cathode, what type of
electrolytes we are using where the membrane lies? So, all these things we will discuss.
Prof. A. Barua
Lecture – 32
Dissolved Oxygen Sensors - II
Welcome to the lesson 32 of industrial instrumentation, this is a continuations of lesson

31 actually, the we will consider here the dissolved oxygen sensor. We discussed
dissolve of basic principle of the dissolve oxygen sensors in details, in the lesson 31.
This particular lesson, we will consider the different types of electrodes, it is calibration
type of membrane, these are the very important things for designing any D O 2 probe and
type. I mean, type of materials which is used for making the for making the electrode
specially, the electrolytes and what should be the suitable nature of the membrane also
that is also very important.
Because in some cases, you have to sterilize the dissolve oxygen sensors, when it is used
in the bioreactor or you will put in the autoclave. So, the minimum temperature, they are
be 120 centigrade. So, in that type of situations, we will discuss we need some sort of
some sort of autoclaving and with a or some, some sort of sterilizations. And at this
particular lessons, we will discuss all those in details it will be a as I told you earlier, that
is it is a continuations of the lesson 31, let us look at the contents of this lesson.

Lesson 32 dissolved oxygen sensor 2 contents. Construction of electrodes, how the
electrodes? What is there? What is anode? What is cathode? What type of material used
for the making anode and cathode? What type of electrolytes we are using where the
membrane lies? So, all this things we will discuss, then Clark-type electrodes, we will
basically, consider 4 different electrodes. So, that we will discuss, 1 by 1 what are the
advantages? What are the disadvantages? What are the relative I mean, relative
properties of this particular electrode, that we will discuss. Then Mancy, Mancy
electrodes, this all this name actually, given by some inventors. So, those are most
widely for bioreactor, so that we will discuss in this particular lesson. Mackereth
electrode, then you have Borkowski and Johnson electrode. So, these are the 4 different
electrodes we will discuss in this particular lesson.
At the end of the lesson, the viewer will know, different electrodes and their construction
4 different electrodes, how they are, what are their construction? Instrumentation
necessary, because you see, that it is usually, the current, which you will get from the
dissolved oxygen’s is very, very low. So, what type of precautions? We will take all
those things; we will discuss in this particular what type of amplifier? We will use all
these things; will be discussed in this particular lesson, then calibrations of electrode,
because calibration is most important. So, there are I mean, we can calibrate, it in 3
different conditions, either in the partial pressure dissolve oxygen concentration, all this
things will be discussed in this particular lesson.
Design of electrodes the membrane covered D O 2 electrodes, basically consist of a

cathode an anode and electrolyte it is basically, it is a probe like a p H probe. So, it has
same thing, we have a D O 2 elect, we have a cathode, we have a anode. And the
electrolyte one thing, we are missing here, we are talking about there is a membrane
anyway, it is a part of the. So, electrolyte is most important as I told you, so there is the 3
parts one be cathode, one be anode and one be electrolyte also, there will be membrane.
In designing D O 2, D O 2 means, dissolve oxygen concentration. So, dissolve oxygen as
you know actually, we are measuring partial measure that is converting into the
concentrations.
In designing D O 2 probes the following requirements are generally, considered what are
those requirements? The calibration has to be stable over a long period usually; you
know that in bio reactions. It takes a long time some of the reaction takes over, several
days to complete the reaction no reactions will be less than 72 hours, some even longer.
So, the stability of the calibration; that means, whatever the reading it is giving there are
many problems, you will find there is a depositions on the cathode those things are there.
So, it will change the calibration. So, we have to avoid that type of I mean, change of
calibration otherwise entire output will be wrong.
So, this to very careful, so the stable of calibrations is more important. So, calibration has
to be stable over a long period number 1. The current output of the probe has to be
sufficiently large and linear with D O 2, D O 2 mean, dissolve oxygen concentrations or
partial pressures of the oxygen. Whatever way you say right, it should be linear that is
most important otherwise if it is non-linear. So, it is very difficult to calibrate. So, it is
linear is that is I mean, basic principle or the basic desirable characteristics of any
instrumentation system or instrumentation sensors.
So, we will consider this thing then we have the effect of liquid flow on the probe
performance has to be small the, if we suppose we are putting this probe on a liquid
which is moving. So, in that types I mean, cases, so the liquid flow on the, because there
is a permeable in membrane through which a liquid may come in or the the electrolyte
may go out. So, those, things is to be cleared of this needs to be think of. The time
constant of the probe should be small, this is another important thing time constants of
the probe should be small the temperature; that means, the response time should be very
fast. That is I am talking about the response time should be very fast or the time
constants of the probe should be that is desirable, because sometimes we will find it is
out of our control. So, we will try always, to make the time constants, we will see that
some, some conservation. If we particularly, design this electrode in a way, that by
which I can get a fast response time.
Usually, you know the bioreactors, we are much not much concerns with this thing you
will find that the most of the bio reaction cases. We will find the time constant is
typically, large right suppose, I have a I am there is a reactions, which is going on for
seventy 2 hours. So, you will find or 48 hours. Even if you sample the each sensors
output 10 minutes 6 minutes 10 minutes that is to be considered to be quite high also.
Because any reactions, it takes a long time it does not change that rapidly, that you have
to sample every minute. So, the time response, we should not give much emphasis on
this, but sometimes we need we have to take some action suppose I mean suddenly. The
dissolve oxygen concentration or partial pressure of the oxygen falls. Immediate action is
to be taken in type of situation; obviously, if you have a lower time con I mean time
constant of the fast response of the sensor that is always helpful for taking actions and all
those things anyway.
Temperature compensation technique is to be incorporated in instrumentation, because as

you know, that if the temperature changes. So, so many properties of the, of the probe
will change. So, that will change, because the electrolyte properties will change. So,
there is some sort of temperature compensations or the more precisely, I should that the
ambient temperature compensation should be incorporated, that is a part of
instrumentations. Then we have the probe must withstand high pressure and repeated
autoclaving, this is also problem, because you know that in a in a bioreactors. Whenever
you we are using the sensors so; obviously, one important thing, we have to make
autoclave. You have to sterilize autoclaving some sort of I should say, it is like a
pressure cooker all of you have a household like a household pressure cookers.
Which you have seen usually, the temperature of this pressure cooker I mean, inside
temperature. You have to rise it if mean, if you have to raise the temperature suppose
120 degree centigrade something, we have to increase the inside pressure, so so the
probe should withstand that high pressures as well it is; obviously, more than
atmosphere. And high temperature repeated autoclaving also is I mean, one of the
desirable properties having a good, good probe. In actual probe design, some of the
above requirements are emphasized more than the others, depending on the specific
applications; you cannot get all this things on a single probe design. We have to dispense
off some of we have throughout, some of the requirements we have to take, some of the
requirements the most which is more important for our particular applications.
Construction methods Clark type electrode let us, look at as have shown in figure 1,
which will come after sometime, that this probe is characterized by a flat disk type
cathode. And a pool of electrolyte in which a reference electrode, which is A g by A g
and C l silver chloride is vast right. So, that is the basic electrodes, we have seen we will
show that in the figure. Although the size of the cathode the membrane material and the
electrolyte differ widely, this design is most popular in commercial D O 2 probes. For
use in laboratory or in the field, this is the best most common sort I mean, this is a
electrodes that is the reason. We have put in the position number 1 the Clark type
electrode let us go to the details of this.

This you see that, the figures this, a Clark type electrode, you can see here, we have a
anode A g, A g and C l cathode. We have which is made of platinum, then you have
membrane usually, this membrane is made of Teflon, there is a O ring it is like this one
which is to be tightly put. So, that the membrane will be in contact membrane will be in
will hold in a position. And we have electrolyte throughout the medium, you see the
electrolytes are here right. So, this is our, I mean this is the 2 probe output, this is a 2
wires is coming out, which is to be which is to be connected. You can see here 2 wires
are coming out. So, through which I will measure, I will show the circuit, how we will
measure that current right? So, this is our basic Clark type electrode.
The probe malfunction is caused by A g C l deposition on the anode surface and

depositions of silver on the cathode a depositions of the silver on the cathode and
depletions of chlorine from the electrolyte or loose membrane. So, loose is to be this
membrane is to be very tight. So, that is the reason, we are using a o ring, which is will
put the membrane very, because you have seen you see here, there is we have a
membrane here . So, you have a membrane here, we have seen, so this is to be tightly so
that no electrolyte will come out, you can see here cathode and anode is there. So, caused
by A g C l depositions on the anode surface; that means on the platinum surface and
depositions of silver on the cathode, which is a platinum and depletion of C l from the
electrolyte or a loose membrane, these are the most important.
However, with proper cleanings of the electrodes membrane, replacement time to time
replacement of the membrane. And electrolyte the frequent calibrations these probes can
be used for a long period. So, this types of probes can be this is a very I mean, very I
should say very obvious, sort of I mean, any sensors. We will use, we have to be need
frequent calibrations any probe, whether it is a p H or measuring the I mean, conductivity
or ((Refer Time: 12:41)) dissolve oxygen concentration. Everything we need some this
type depositions will be there, we have seen the polarizations all those things are there.
So, it is to be, so repeated cleaning then of the electrolytes or it is supplied from some
reservoir permanent reservoir. So, that type thing; obviously, will make the, and frequent
calibrations will make I mean, make the same for a longer use, longer use over the over
the period of time.
With a 25 micron Teflon membrane 95 percent, response time of 50 to 20 seconds see

usually, obtained right. Response time means, if it is a within 50 or 20 seconds, if you a
step input suddenly, that suppose a probe is inserted or there is a change in the
concentrations of dissolve oxygen or partial pressure of the oxygen. So, if we use a 25
micron Teflon. So, 90 percent response time is usually, obtained right, of 20, 15 to 20
seconds you will obtain. But these sensors show a hysteresis the response to an increased
oxygen concentration is faster than that to a decreased oxygen concentration.
Step change if basically, response time varies depending on the direction of the step
change in oxygen concentration. Suppose I had, because it is very unusual though I
mean, it is oxygen concentration or oxygen partial does not very rapidly, but there
suppose there is a change suppose. So, that is what I am saying? That there is some sort
of hysteresis in the systems, but the sensor shows a hysteresis, that is means, that the
response to increased oxygen concentrations will be different when the this, when the
when we are decreasing the concentration. This phenomena is caused by the electrolyte
acting as a reservoir of dissolved oxygen or the accumulations and slow decompositions
of hydrogen per oxide in the vicinity of the cathode.
Mancy electrode; the galvanic probe originally, designed by the Mancy shown in figure
2 the major difference. From the Clark electrode is the elimination the electrolyte
chamber; instead a thin film of electrolyte is placed, between the cathode and the
membrane. And it will eliminate the hysteresis, hysteresis the problem; that means
increased concentrations. And decreased concentrations, it will be a I mean, giving a
problem. So, that type of problem can be eliminated in the Mancy electrode right.
So, this is our Many electrodes, we can see here plastic casting is there the membrane is
there, an A g cathode we have a P b a lead anode and plastic collar, this you can see here.
Due to relatively, large chamber point 6 centimeter cathode, employed a micro ammeter
could be directly connected to the probe. We do not need further, amplification and
improved probe stability was reported compared with the earlier polarographic probes.
But the useful probe, life may be somewhat restricted, because the available surface area
of the anode is relatively small. So, this is another problem anode surface is gradually,
oxidized with the use until the probe ceases to function. So, it will be oxidized until the
probe ceases to function.

Mackereth electrode, it looks like this it was noted that the Clark electrode and the
Mancy electrode lacked long term, stability as I told you, because of the depositions and
all those things. So, because in the case of Clark electrode, we have to clean it all those
things are there whereas, here there is a deposition in the case of Mancy electrode. So,
this type of thing will be not there in the case of Mackereth electrode, it was noted that
the Clark electrode and the Mancy electrode, lacked long term stability. And produce
small currents of the order of microampere, we have seen that the current is quite small,
Mackereth probe eliminate all this problems. Here perforated silver tubing was used as a
cathode and a massive shot of porous lead was used as anode figure 3.
Porous P b lead is anode and this is electrolyte, you can see, we have o ring and
perforated tubing cathode. So, this is our Mackereth electrode, you can see here, here O
Ring and membrane is there.
The current output was much higher than those of the other probe. So, that an ordinary
current meter can be directly connected to the probe without amplification this is very
important because; obviously, any entity will find it is corrupted with noise. So, that if
the current output is large. So, all, this problem signals noise ratio will be quite good. So,
we do not need any further amplification, the sensitivity was stable over many months of
continuous operation, this is again a very important thing in the case of electrode design.
With a 25 micron Teflon membrane, this probe giving 90 percent of its steady, state
value in 1 minute. So, it is a quite fast not that, fast compared to the Clark or electrode,
but as I told, you that response time, fast is not very stringent requirement or a very rigid
requirement for bio reaction. So, in that our application, this probe can be nicely, used
most important thing, it can be used over a longer period, without any calibrations or
recleaning.
Because of it is long term stability, this probe has been used for monitoring the dissolved
oxygen concentration in continuous cultivation, which lasts several weeks in some cases,
I said 72 hours in some cases, it is can be continue over a week. So, in that type of
situation, this electrode is very much suitable, though the response time is very. I mean,
is quite bad, I should say which does not matter, because I will not sample, I will not
measure every minutes.
I may measure every hour, because of if the with reaction going on several weeks;
obviously, just reading in half an hour of the dissolved oxygen concentration will suffice
when vertically, inserted in a air-sparged cylinder vessel. The Mackereth probe exhibited
a response free, from any effects of air bubbles, ascending through the vessel which is
very obvious in many situation. We will find whereas, other probes, because this type of
situation will arise in the air column, column bioreactor. So, bubble column, if you know
that a basic principle is something like this that it is like this.
I have, so dissolved oxygens are, because instead of suppose I have a stirred tank
reactors here. So, it is rotating and oxygens are getting dissolved, in the case of bubble
column reactors, what will happen sorry what will happen?
That I have a vessels and I am purging, oxygen through this one. So, bubble will go out,
like this one, it will go the air bubble will go. So, it will purge along this one. So, this
type of this may cause problem in other type of dissolve oxygen probe, it will not make a
problem in this probe. Let us go back when vertically, inserted in an air spurged cylinder
vessel the Mackereth probe exhibited response free from any effects of air bubbles.
Because the function of the air bubbles to increase the dissolved oxygen concentrations
or partial pressure of the oxygens in the bulk medium of the liquid where the actually my
reaction takes place right, where the other probes in this case. So, that is the most
important thing which we should consider that problem does not arise and if you use a
Mackereth electrode. So, the Mackereth probe exhibited response free from any effects,
of the air bubbles ascending through the vessel whereas, other probes with cathodes at
the tip showed interference due to bubbles touching the cathode. So, this is again
problem in other.
Since the cathode area is relatively, large a vigorous agitations of the liquid is required
for reliable operations, because since the cathode area is large so; obviously, it should
average. So, the vigorous agitations of the liquid is required for reliable operations and
oxygen consumption by the probe must be using the probe for the measuring oxygen in a
small liquid volume. So, it is; obviously, if we use a c s t r continuous stirred tank
reactor; obviously, the liquid will be vigorous only, problem this condition will not be
satisfied, if you make a very gentle I mean rotations of the blades.
So, that the a very sensitive, I mean, culture like animal cell and all those things, So, this
question arises. Otherwise for most of the conditions like the reactions, where the you we
are using c s t r continues stirred tank reactors. So, this type of problem does not, I mean
arise. So, this type of condition will always will be there, because it is rotating at a high
speed so; obviously, there is a vigorous agitations of the bulk medium or the liquid. The
probe may not be suitable for a viscous medium, if the viscous the liquid this is having
the problem, because you use the bio-reaction the viscosity does not remain same at
initial stage. When the reaction start is the viscosity, will be less as the reaction going on
for an example, very trivial things I am talking about the yeast growth. You will find that
the you will find that as the reaction goes by at the end of the day, you will find that the
liquid becomes very much viscous.
So, need for a tubing membrane makes, it difficult to fabricate the probe in the laboratory
this is another problem, which is like in the Clarke electrodes is very simple to make.

Borkowski and Johnson electrode: The original design was by Johnson, but it was
improved later by Borkowski for a longer life and better stability, as shown in figure 4.
The cathode is made from a silver spiral and a flattened, lead wire forms the anode a low
p h acetate, buffer is used as a electrolyte to prevent interference by the dissolved carbon
dioxide. This probe has been used widely in the biochemical engineering applications.
You see here, we have a glass tubing anode is there which is helix, then we have
electrolyte. We have a glass wool, there a cathode which is I mean, silver spiral sorry and
we have a membrane silicon tubing is there. So, this is our glass tubing and 2 2 wires are
coming out.
The probe can be withstand repeated steam sterilization and is capable of operating for a
several months as a linear response, with a 50 micron Teflon membrane 90 percent of
steady state value is reached in 1 minute. This is almost same as the previous electrode,
electrode a vigorous agitations of the liquid at least 60 centimeter per second for water.
Velocity rotational velocity, if you take is required for reliable measurement the probe is
not suitable of viscous liquids unless, thicker membrane is used same as the previous
electrode.

Oxygen microelectrodes the major consideration with these probes are a high special
resolution and a fast response time, because in if I in some conditions, we may need a
fast response. So, that type of that type of cases 1 minute or even 20 seconds is might not
be I mean very desirable properties, we have to reduce that. So, that we can do in this
type of oxygen then as it is name incline implies, it is a very small in size high special
resolutions. And a fast response time, when the cathode diameter becomes, less than one
micrometer even a bare metal cathode, which is figure 5 a becomes insensitive to liquid
flow and the measures local oxygen tension tensions or oxygen concentration, it is
almost the same thing.
You can see here, so 5 a b c all these are this is a bare this is glass insulation. This is a
cathode, this is a membrane, and this is like this one recessed membrane covered and
recessed 3 difference we have shown.
The probe performance is improved by covering the cathode with a membrane;

obviously, if I cover my membrane. So, it will be it is possibly, slower and by extending
the insulation such that in shown in figure 5 c diffusion gradient is confined inside the
recess. A thick platinum wire is first etched in an electrolyte solution to a fine point of
point 2 2 1 micrometer diameter. Then insulated with a thin layer of glass right, I will
repeat a thick platinum wire is first etched in an electrolyte solution to a fine point of
point 2 2 1 micrometer diameter and then insulated with a thin layer of glass.

The membrane is applied by a dip coating, polystyrene is most popular, but other
materials have been also used. Although, the, of the local oxygen measurements is
attractive, several problems exist with these microprobes extremely unstable. Useful
probe life is short moreover the instrumentation is difficult. Repeatability of the sensor is
poor each and every sensory is to be calibrated separately. So, all these problems are
there. Now, electrode metals, what type of metals? We will use in the electrodes, we will
cover 1 by 1 membrane type of electrolytes. We are using all this thing, and finally some
instrumentation.
The oxygen reduction at pre-oxidized metal electrode is a p H dependent, where as for

pre-reduced metal electrodes the reduction was p H independent. For polarographic
probes, platinum gold silver and or rhodium have been used as a cathode and silver and
silver silver chloride, silver silver dioxide. This A g 2 O please note, it is not A g 2 it is A
g 2 will be in subscript. So, it is like this, so it will be 2 O sorry 2 O. So, I can or calomel
as a reference electrode is used.
Gold is generally preferred to platinum as the cathode material, because it is less

susceptible to poisoning by noxious gases notably H 2 S hydrogen sulphide. And the
reaction at the cathode surface is less complicated and the surface aging is less
pronounced. However gold may not convenient, for application in steam-sterilizable
probes the macro-electrodes, since gold and glass cannot be fused together. So, it is a
larger I mean, I mean type of electrodes. So, it is if I want to make steam sterilizable that
is not a very suitable for this type of purposes.

Gold plating method has been used in making the microelectrodes to circum, circumvent
these problems, so that if it is a macro electrode, it is not possible, but microelectrode it
is possible. The reference electrode has a, has to maintain a stable reference voltage for a
good performance of the polarographic probe also, it has to have a large surface area to
avoid polarizations. So, the surface area should be large, silver silver chloride is
normally, used as a reference electrode. However silver silver dioxide is preferred,
because it gave better stability for their probe.
The aging effect of the probes, such as the change in calibrations or unstable probe
sensitivity is attributed to the depositions of silver ions on platinum, and the oxidations
of the catalytic surface or an excessive deposition of silver chloride on the reference
electrode.
The cathode surface can be mechanically, cleaned with soft scouring powders. So, if you
regularly, clean it with a cleaning powders and wet leather or by tooth paste. So,
excessive silver chloride deposits can be removed by washing with sodium. I mean,
ammonium hydroxide 15 percent ammonium hydroxide is possible. This I told you from
the very beginning, where we were at cleaning of the electrodes.
For galvanic probes the silver as the cathode and lead as the anode are most common.
But sliver aluminum platinum aluminium platinum lead gold gold lead pairs have also
been used although the galvanic probes suffered less from poisoning and survive
autoclaving with greater reliability. The probe life is limited by the available surface
area, of the anode due to gradual oxidations of the anode surface, this is again problem.
The useful life-time of given probe depends on the current drain, just like an electrical
battery in other words a probe can be used much longer. When it is used for monitoring
low rather, than the high oxygen concentration this is not in our hand. So, it varies
suppose in a reactions, because in many reactions in bio process we have seen that, you
see the many concentration. We cannot measure online, suppose a substance
concentration I mean, C l concentrations we cannot measure only the D O 2. We can
measure and that will give you the, suppose when the D O 2 is initially, is when it the
reactions start. Its starts we find that the D O 2 is not growing that fast; that means, the
whatever the D O 2 that we are supplying that is consumed by the cells.
At the end of the reactions, after few I mean, several hours or might be several days, we
will find that no more oxygens are getting consumed due to concentration is increased.
But coming to a steady state; that means, you can see that whatever oxygen, which we
are supplying, it is getting consumed and after that is if it is if it is further. So, due to
again, we will find that the it will indicate that the cell reaction is completing, it is time
to take out the cell from our or take out the product. So, due to concentration
measurement is very important right. So, I mean, and it is varying, so that I mean, it is
not in our hand, but in some condition. Where we have just monitor not much of control
that is what is the value of the D O 2 in what are the safe level of D O 2 in the oxygen.
Suppose in the marine, I mean animals will need the measure the D O 2 concentrations
there. So, in that type of situations, we can say that the we can either low monitoring or
high monitoring, but it is not in our hand.
Electrolytes since the electrode reactions, occurs in the electrolyte solution the
composition comma p H and the volume of the electrolytes are directly related to the
probe stability. For polarographic probes the electrolyte takes part in the reaction, so a
refill of electrolyte is necessary in the regular intervals. This is a very important thing is
very difficult so that what we have to do? A, we have to use a reservoir, we will discuss
in the next slide.
In general pH of the electrolyte, does not affect oxygen reduction on a clean metal
surface, but an adverse effect was observed for an oxidized metal cathode.
The solubility of the electrode metals in the electrolyte solution has to be low for probe
stability, sometimes the polarization voltage may change the depend, change depending
on the electrolyte concentration. In this case the change in electrolyte concentration due
to evaporation or diffusion of water through the membrane affects probe stability.
Because any liquid you will find, when you will make the reactions and mostly it is in
water. So, water may diffuse through the membrane. So, that will again change the
properties of the electrolyte.
Various electrolyte solution used various electrolyte solution used by different

investigators for polarographic probes. Potassium chloride is the most common
electrolyte, which gives a constant calibration well over 48 hours, which is quite for
many of the applications for galvanic probes, preferred potassium hydroxide than the
potassium chloride. Because potassium chloride gave a high residual current and it did
not maintain a clean anode surface. The solubility of silver is reported to be high in K C l
solution.
It was found that only K C l and K 2 H P O 4 gave a linear response up to a high oxygen
concentrations and also where steam sterilizable right. So, they gained steam sterilize
sterilizable is a good property and also autoclaving is a good property of the system. That
is also always, we have to think about, because the contamination is a great problem in
may all the reactions. We will find especially, when you are using the bioreactors, if we
are using the field that is does not matter, because many reaction does not allow a little I
mean, contamination that is to be, we have to think that in mind. Rather serious problem
with the electrolyte is a loss of solvent through the membrane by diffusions or
evaporation several methods, have been used to minimize this effect.

One method; by having a large electrolyte reservoir as I told you which can effectively
supply solvent to the electrolyte film where the solvent loss occurs. In this case the path
between the reservoir and the electrolyte film has to be as long as possible to minimize,
the residual current. Another method is to have electrolyte in the form of gel or paste and
added advantage is that the residual current becomes smaller, because the oxygen
permeability is normally, lower in more viscous medium.
Recent development has been in the use of a solid electrolyte instead of the conventional
liquid forms, this all will eliminate through the previous problems, which we have
discussed.
Membrane: an ideal membrane for use in D O 2 probes must have a relatively, low
oxygen permeability, high oxygen diffusivity. The permeability has to be low to ensure
the membrane control of the oxygen diffusion whereas, the high diffusivity gives a fast
probe response, so you have to make a compromise.
The membrane has to be mechanically, strong and chemically, inert right; that means,
mechanically strong means; it can, should withstand the high pressure. It can should I
mean, in a position to withstand high temperature at least 120 degree centigrade. Little
more than 1 atmosphere, it is to be I mean, because if you do for autoclaving. So, this I
mean, high temperature high pressure, conditions have to be satisfied. Since the current
output is directly related to the thickness and the oxygen permeability, of the membrane.
The probe sensitivity is directly, affected by the change in the membrane properties.
Membrane swelling or change in the apparent permeability, results in the change of
probe sensitivity, sometimes membrane swells, if it lies over a long time in a liquid. So,
that will change the apparent oxygen, this is not a problem. We have to recalibrate or we
have to change the membrane that is it there is no other solution.
Other important factors are C O 2 permeability, and water permeability the water
permeability, of the membrane has to be low to prevent loss of water from the electrolyte
solution, which causes an increase in the electrolyte concentrations. And early failure of
the probe low C O 2 permeability of the membrane is also desirable for probes to be used
in aerobic cultures. And in blood sometimes, under the means, of oxygen concentrations
and that is also important.
Now, the material, what are the different materials used for the making the membrane?
Really I will just I will say, that Teflon is the mostly, widely used material for making
the membrane Teflon. Polyethylene have been most popular as the membrane material,
but silicon polystyrene and Mylar have also been used, but still I should say that Teflon
is the most widely, used membrane for making D O 2 probe. The polypropylene is better
than the Teflon in several aspect. It has a lower oxygen, permeability a lower carbon
dioxide permeability and yet a higher oxygen diffusivity. However Teflon seems to be
more popular in a steam sterilizable probes due to the higher heat resistance that I told
you earlier. So, this is the property which must have.

An added advantage of the teflon is, it is extremely, low water permeability, so this is
another advantage of the teflon. Polystyrene is popular for microprobes, because it sticks
well on the glass insulation and has a relatively, low oxygen permeability. Polystyrene
and polyethylene are not suitable for steam sterilizable probes, because of low heat
resistance. So, the problem is there on one side is a low oxygen permeability and the
other side it has a low heat resistance.
Instrumentation: let us look at… Unless the current output is very large, some sort of
current amplifications is required for D O 2 probes. So, we know various types of current
amplification that, single condition is we can do, it has improved a lot. So, that is not a
problem, operational amplifiers, which provide multi-range amplifications and current
suppression are normally, used for this purpose.
Often, existing current amplifiers, voltmeters or strip chart recorders are directly,
connected to D O 2 probes. Obviously, one thing where may not mentioned, that you see
any D O 2 probe must be any D O 2 probes, must be appended by a; obviously, by the
followed. I should say followed by a by a current transmitter, because the current amount
of current, you will get there is directly from the probe is not sufficient to be transmitted
this is not sufficient, so we have to use some current transmitter. So, that it will go to the
computer right. We can monitor their itself, whatever the current we will get that
sufficient for monitoring, but for control purpose.
If I have to send to the computer, I have to use some current transmitter there itself. So,
any probe must be with a followed by current transmitter for polarographic probes a bias
voltage of 0.6 to 8 volt depending on the type of electrodes has to be externally applied.
That we have seen always, we need a bias voltage, which usually lies between 0.6 to 0.8
volt. A mercury battery is preferred as the voltage source, because the voltage discharge
characteristics are superior to other batteries.
For galvanic probes, usually a resistor is connected in series with the probe and the
voltage drop across the resistor is monitored with a voltmeter or a potentiometric
recorder.
Now, general design considerations if I look at one of the important considerations in D

O 2 probe design is to minimize the residual current which is defined as the current
output of the probe at 0 oxygen level that is when there is no level. But you may
immediately ask the question how will you know that is a 0 oxygen oxygen
concentration in the bulk liquid is 0. We will show that thing how you can make it 0
sometimes it is called the background dark or zero or offset current. I should say the
background current I should say the dark current that is more appropriate name.
There are 4 major sources which contribute to the residual current: electrochemically
active impurities in the electrolyte, electrical leakage, incorrect polarization voltage, back
diffusions of oxygen.
The effect of reducible or oxidizable impurities in the electrolyte is normally of short-

term duration, because they are scavenged by the cathode at the early stage of
operation. Electrical leakage between the anode and cathode through the insulating
material is not normally significant for leakage for large cathode, but becomes a problem
for a small cathodes.
For example, the insulation requirement of a the microelectrode has to be better than ten
to the power 7 into the power 10 ohm. That means, you can see it is around coming
around 70 gig ohm and have a residual current of less than 10 to the power 11 ampere.
That means, 0.1 picoampere. So, it is quite small this is why glass is used as the
insulating material for small cathodes it will give you very less residual current.

Epoxy resins, unless specially selected for their water resistance, often absorb water and
increase current by forming an extraneous conducting path between the cathode and the
anode. This is again problem.
Because usually use resins to I mean to hold this different cathode anode with the
systems. The correct choice and the stability of the polarizing voltage is important for
stable performance of the probe. Now, figure 6 next figure shows the 3 polarographic
curves which is called polarogram also. If the polarizing voltage is chosen between the
points marked B and C and maintained in this region the probe current will depend only
on the oxygen concentration.
Let us look at you see here it is it is between B and C if it is if it is between B and C it is

depends on the, it is B and C. So, depends on the oxygen concentrations we are talking
about partial pressures of this, right.
However, when the polarizing voltage is high larger than the indicated by the point D,
there will be a residual current due to electrochemical gel reduction of water. If the
polarizing voltage is less than that marked A then the calibration of the current versus
oxygen concentration will become non-linear. Again this is non-linear region. So, if you
come to the plateau region only. So, that is our desire, so that is the reason we have
marked between B and C. So, it is totally because it will depend in remain in the plateau
region. So, depends only on the concentrations of the oxygen or the partial pressure of
oxygen.
Since the shape of the polarographic curve changes depending on the electrode metal the
electrolytic concentration and the size of the cathode the optimum polarizing voltage has
to be selected for each probe separately. In general the large cathodes have wide and well
developed plateaus in their polarogram whereas; small cathodes have narrow and often
very poorly defined plateaus.

The fourth contributor to the residual current namely black diffusion of oxygen back
diffusion of the oxygen from the internal electrolyte is most pronounced for the small
cathodes. Mechanical designs which favor increased path length between the cathode,
and the electrolyte reservoir as well decreasing the thickness of the electrolyte film
between the membrane and the cathode minimize the effect.
Mechanically well-designed probes with the above consideration in mind have very low
residual current and are capable of measuring dissolved oxygen down to parts per billion
which is called ppb instead of part per million we are calling part per billion range.
Galvanic probes have been preferred in the bioreactor applications, but the disadvantage
is the expendable nature of the anode, because it is bioreactor we have to make
autoclaving all those thing that is the reason galvanic electrodes are most preferred.
Currently available designs of the steam-sterilizable galvanic probe could be improved
considerably by further decreasing the cathode area increasing the anode area and
increasing the path between the electrolyte reservoirs and the cathode. This is the last one
is most difficult one. A well designed polarographic probe was shown to withstand
repeated steam sterilizations with a high reliability.
Now, operations of the electrodes, let us look at calibration this is most important thing
D O 2 probes can be calibrated in three ways saturations in partial pressures of oxygen
and actual concentration.
Since the reliability of measurements depends on the calibration it is vital that the
calibration done as accurate as accurate as possible right. So, the calibration will show
the some equation by which we can make the calibrations. In all calibrations good
temperature control; that means, 0.1 degree centigrade plus minus one of the test medium
is required, because the probe sensitivity changes substantially with temperature, right.
So, some sort of I mean in the, we should take care of this temperature compensation
now it is in the computers we can do this I mean very nicely. So, that the output voltage
which is the output which is coming out of the computer can be taken care of all this
temperature compensations or I mean temperature compensation if we cannot do in the
probe itself that is not a problem.
Calibration based on percentage saturation. Gas phase calibration is rapid and

convenient. The probe output current in nitrogen is set at 0 percent saturation while that
in the air that in the air is set at 100 percent. Oxygen can be used instead of air this
method is simple, but recommended only for a rough calibrations.
A solution with zero oxygen content can be prepared either by stripping oxygen with a
nitrogen or by using chemicals such as sodium sulfite or sodium di dithionite, because by
this the all oxygens will go out. These chemicals react rapidly with the dissolved oxygen
to do a to provide anaerobic conditions. Additions of two volumes of saturated sodium
sulfite to 100 volumes of liquid is sufficient for oxygen removal. So, that liquid now is
totally oxygen free. So, I can make my calibrations that for my zero oxygen
concentrations what is my output the, which is supposed to be 0 if it is not that is it is
offset current or dark current.

Calibration based on partial pressure D O 2 probes can known to measure the activity of
the equivalent partial pressure of the dissolved oxygen, but not the actual concentrations.
We actually calibrate in terms of actual concentration, but actually it is measuring the
partial pressure. The calibration based on percentage saturation can be directly converted
to partial pressures when the barometric pressure is known right.
When the water is equilibrated with air at temperature T, the partial pressure of oxygen,
which is written p O 2 is expressed as a following equation which is p O 2 equal to p B
minus p H 2 O multiplied by constant 0.2095. What are the legends? Let us look at
where p B is the temperature corrected barometric pressure p H 2 is the vapour pressure
of water at a given temperature and 0.2095 is the fraction of oxygen in atmospheric air.
Calibration based on concentration the calibration based on partial pressures can be

converted to a concentrations. As I told you if the solubility of the oxygen in the liquid is
known if we know that thing how can I write.
The solubility is often expressed as the Bunsen coefficient a, which is defined as the
milliliter of O 2 absorbed milliliter of oxygen absorbed by 1 milliliter of solvent at 0
degree centigrade. And one atmospheric pressure of O 2 right this is my expressions a
equal to Bunsen coefficient equal to V g 273.15 upon V s by T where V g V s and T are
volumes of the gas absorbed and the volume of the absorbing solvent and absolute
temperature respectively.
For a sparingly soluble gas such as oxygen which is very difficult to dissolve in the bulk
medium a pressure can be converted to the concentrations by using Henry’s law which is
given by C equal to p knot by H. Where H is the Henry’s laws constant and the
conversions of a to H is as follows. Is H equal to 22 4.14 at 760 millimeter of H g upon
100 a plus. Please note there are some precautions we have to take. So, that I will discuss
now those are like this if the probes are dead. You know that this can be rejuvenated by
dissolving away the oxide layer on the anode surface with a 20 percents of C H 3 C O O
H or 20 percent of H C l they are very standard; that means, if you have a I mean some
sort of it is not reacting.
So, clean it at the anode surface clean the oxide layer of the anode surface should be
clean with a 20 percent H C l, which is easily available in laboratory. Now, you see the
electrode content in electrolyte when the chloride content electrolyte is used the chloride
ion concentration will fall as it is consumed by the anode reactions and when we replace
by the hydroxyl ion generated by oxygen reductions at the cathode right. Now, another
important thing is you see that the current here is very, very extremely small. We have
seen that this is current is coming around order of 10 to the power minus 10 10 to the
power minus eleven ampere. So, almost like a I should say ten to the power I mean 0.1
picoampere which is extremely small this type of small current right. Now, so special
amplifier is required this amplifier will not do alone, because there is it will introduce a
lot of noise.
So, the common mode noise I mean we have to use some differential technique by which
we can I mean eliminate this type of problem. Because amount of current there is some
inherent noise in the systems you will find thermal noise. Because, so that type of things
because when you will transmit the signal over the even the probe length whatever the
small it when it is going to the meter itself we will find it has some length. However,
small it may be, because you cannot immediately, you cannot connect a transmitter just
over a probe itself a certain amount of length is to be I mean you have to certain amount
of layer length of the wire is to be connected.
So, that wire also will give some noise. So, that type of things can be eliminated if you
use a instrumentation amplifier and the input of this current transmitter. So, that the
common mode noise will go away, so the only differential mode signal will be amplified.
So, this is to be I mean we have to keep in mind. So, this I mean these are the most
important thing we should think while using the D O 2 sensors. But the some precautions
you have to sense depending on the what are the requirements accordingly we will
choose on the pressure will come P knot equal to A b 1 by A b 2 I 1 by I 2 P 1 minus P 2
right. So, this actually will give you the direct this you see the output pressure is again
the function of P 1 minus P 2 which is actually calibrated in terms of the flow. So, with
this I come to the end of the flapper nozzle system.
Prof. A. Barua
Lecture - 33
Flapper-Nozzle
Welcome to the lesson 33 of industrial instrumentation. In this particular lesson, we will

study the flapper nozzle or rather flapper nozzle system. Flapper nozzle is a very
important I mean item in most of the pneumatic systems or pneumatic based
instrumentation. Because as you know that even though we are as many sensors or all
those things are converted to the electronic domain. But still in some cases we need these
pneumatic systems for the safety reason as well as for the large control in the large
control for all those things.
And especially in the petrochemical industries, or the hydrocarbon industries where the
electronic systems or electrical system is forbidden to use, because there is a voltage
rating. It should be not it should be above 45 volt and that system also we should be we
need this flapper nozzle systems Now, flapper nozzle system the advantage is that you
see that it has a it has a it can be used as a displacement sensor. It can be used as it can be
used as differential pressure transmitter and so on, we will discuss all this things in
details. Let us look at the content.

Introductions then we will go to the principal operation then we will go to displacement
measurements and displacement sensors rather we will see that how the flapper nozzles
can be utilized to make a displacement sensor we have. A reverse acting pneumatic
relay, because I mean I will see that to make to keep this pneumatic sensors I mean
flapper nozzle systems in the linear region. We will find that we should make the
movement of the flapper very small. So, to do that thing we need a pneumatic relay we
will discuss this in details it is called reverse acting pneumatic relay then you will also
see the current to pressure convertor or I to P convertor using a flapper nozzle systems.
How we can make I to P to convertor? Because in many point the process always we
need to do I to P convertors many a cases we will find whether the 4 to 20 milliampere of
current you have to convert 3 to 15 PSI of pressure.
So, in that type of situations we need this I to P convertor. Then differential pressure
transmitter this also we will discuss in details, because differential pressure transmitter as
we know we need it in many process. And though we have now electronic system, but
still this differential pneumatic differential pressure transmitter is still used in many
process industries. Because as you we have studied in during the flow meter that the flow
measurements the flow is calibrated in terms of pressure differential pressure. That is to
be transmitted we can transmit four to 20 milliampere, but we can transmit to directly 3
to 15 PSI pressure also. So, we discussed these thing how we can make with the flapper
nozzle system I mean a differential pressure transmitters.

Basic introduction flapper nozzle is one of the most versatile and versatile device. It
finds it is application as displacement measuring devices in differential pressure
transmission and as an IP converter. The basic principle which is the which is the
movement of flapper depending on air pressure coming from the nozzle remains the
same as in each application. In all applications we will find there is a there is a nozzle
and through which the continuous air is bleeding through that nozzles and we have a
flapper. This nozzle’s position is fixed, but the flapper position is I mean variable. Now,
we will find that if the flapper position varies we are getting some I mean ((Refer Time:
04:42)) of pressures or back pressures. So, which can be utilized which can be calibrated
in terms of pressure currents and so on, application as a displacement measuring devices.
Let us look at the displacement measuring devices how the flapper. Now, to know the
displacement measuring devices let us first look at what is this flapper nozzle system
looks like? A flapper nozzle system is used for the measurement of small displacement.
A small movement of the flapper can result in a large change in the output pressure and
hence it is treated as a displacement to pressure converter. Basically I am getting
pressure in all the flapper nozzle system. Please note the output is pressure whether it is I
am measuring the displacement or velocity or current. Output is in terms of pressure 3 to
15 PSI or is whatever in kg if you say it does not matter. So, it is it is output is pressure.
So, that it is treated as you look at the last line and it is treated as a displacement to
pressure converter
Such transducers are designed for use on both hydraulic and pneumatic power supplies.
It is used as a pneumatic and hydraulic power supplies. The pneumatic system is
operated at an air pressure range of 15 to 30 psi. This is I am talking about the supply
pressure, but we will find out the pressure with controlled output which we will get from
the or the range of the output which we will get from the pneumatic system our flapper
nozzle system is 3 to 15 PSI.
Flapper nozzle system let us look at them. This is the basic flapper nozzle system you
can look at very carefully. You see here that we have one, this is our nozzle. This is
small in diameter it is a circular in nature. This is a nozzle and these are the flappers
which move in these directions, which moves in this direction as you can see it is
written. So, it moves in this direction it moves in this direction. You can see here now,
see there is a continuous flow of pressure, which is called the supply pressure of air and
there is a fixed supply restriction. These restrictions will make a constant pressured
which is constant Ps I mean it will make a restriction, so that you will get a constant
bleeding of air through this nozzle. Now, you see that if the if the flapper goes out if the
flapper goes in these direction if the flapper moves in these directions we will find the
more air will leak through this nozzle.
This is the nozzle, which is in circular in shape this is the nozzle I mean the more air
will. So, this will reduce the pressure P naught and obviously, please note that Ps will not
change; that means, the supply pressure will not change. But this back pressures or pilot
pressure will change this Po will change it has some time by the, it might be different
from the supply pressure temperature Ts it does not matter. Now, if it moves in this
direction the flapper moves in this direction as in if I use it as a displacement sensor. If it
is displacement sensor; obviously, I will put the displacement to the flapper is not a there
is a hinge here. So, it moves in this direction. So, either it will go in this directions or it
will go in this direction this is a neutral positions what we have in the flapper.
Now, you see that when it moves in this directions. So, it will allow less air to leak
through this flapper, because there is a continues bleeding, because supply pressure Ps is
coming. So, this will lead to the increase over the pressure Po or output pressure. This
output pressures will be calibrated in terms of not the temperature output pressure will be
calibrated in terms of the displacement. This is the basic principles of the all flapper
nozzle systems you will find that whether it is a temperature current it does not I mean
current it does not matter you will find it everywhere. This is the basic principles this is
the output Po which is to be calibrated in terms of the process parameters we actually we
are measuring here. The process parameters we have taken as a displacement. Now, let
us look at the ambient, I mean what are the legends we are talking about?
P atmospheric is the atmospheric pressure. Then Xi is the displacement of the flapper if

you look at Xi is a displacement of the flapper. Gu is a outlet mass flow we are
considering here mass flow also from the volumetric flow the mass flow mass flow of
air. Gs is a inlet mass flow inlet flow from the supply pressure which is coming from.
Then we have a Ps and Ts is the pressure of the supply air, and it is temperatures
pressure of the supply air and it is temperature. Then Po is the pressure output who also
weights as the temperature of T naught that we have not mentioned in the legend. So, it
is always bleeding to the atmospheric pressure.

The output pressure of the nozzle back chamber is measured by a pressure sensor. See
you can use many pressure sensors we have studied, so many pressure sensors. So, I can
use to measure the pressure by any gauges like diaphragm gauges this that we have
studied so many. So, we can use utilizing I mean any principles I mean principles we can
make a diaphragm is one of the most suitable diaphragm gauge also can be used pressure
sensors. The flapper nozzle system shown in the figure; that means, figure 1 consists of a
chamber of small volume connected to a constant pressure source which is Ps on one
side and vented or bleeded on the other side to the atmosphere through a nozzle. This is a
basic principle if you say what is the flapper nozzle system? You must write this line.
That means, it is saying that the flapper nozzle which is consists of the chamber of small
volume connected to a constant pressure source on one side which is supply pressure and
vented or bleeded on the other side to the atmosphere through a nozzle.
The flapper in front of the nozzle is used to regulate the rate of air flow bleeding out
through the nozzle to the atmosphere. As I told you we have explained that the how the
movement of the flapper will control the regulate the rate of the air flow or air I mean
which is bleeding to the atmosphere. When the flapper is held tightly against the nozzle
no air leaks out and ultimately the output pressure Po reaches the value of the supply
pressure. Obviously, is it not it when I tightly which that situation never arises in a
flapper nozzle system. That means if the flapper is suppose I have a flapper here. So, I
have my, so if this is the flapper and this is a nozzle through which the air is
continuously I have a supply pressure on this side suppose this is I am flapper is now
moved and we have controlled we have totally stopped it bleeding. Then what will
happen to the Po output pressure? Output pressure will be equal to the supply pressure is
not it because nothing is vented to the atmosphere. So, the upload output pressures Po
will be equal to the supply pressure.
When the flapper is held too far away, it has no effect or control on the output flow rate
and the flow rate is governed by the system size and configuration. That is also it is more
important this will tell you the range total range of the flapper nozzle system; that means,
total range of the displacement sensors, which we are going to design. What I am saying
if it goes away from I mean far away from the nozzle itself. So, that suppose I have a
nozzle, so flapper goes away this 1. So, that time there is no control; that means, if I
move like this 1 there is no control over the bleeding of the air, because ultimately by the
movement of the flapper I have to regulate or control the air which is coming out from
the nozzle.
If I cannot then; obviously, there is no control. So, whatever where ever position you put
it, so it has no output pressure will not change. That time continuous bleeding of the air
through the atmosphere. So, it depends on the system geometry or system size and
configuration. As the system is designed for measurement of the displacement the supply
side restrictor and the nozzle are so designed that the output pressure never falls to the
atmospheric pressure even with a flapper held far away it is to be done like that. So, that
the you have to regulate the movement of the flapper in such a way, so the output
pressure will never come to the atmospheric pressure.
Now, if this is the flapper nozzle characteristics. If you look at we have plotted here on
the X axis the P naught by Ps normalized. You see this is normalized I mean I mean we
have normalized pressures. We have plotted here in the Y axis and the X axis we have
plotted. What we have plotted? We plotted the displacement of the flapper from the
normal positions. If you look at you see the graphs looks like this. Ps is constant if the Po
increases then what will happen; that means, the distance from the flapper distance of the
flapper from the nozzle become getting decreased and decreased. This is side we are
avoiding you see this position; that means, it is very close to the tight positions or
stopping the bleeding of air.
So, you see it is coming horizontal after that is when it is far away when the flapper is far
away you see the Xi is greater and greater. So, it has a constant value of the Po by Ps. So,
we will be more interested in this linear region, if you look at we will be more interested
in this linear region of the flapper do to the calibrations later on. So, you will find that
this pressures with this pressures the Po by Pa will come this will corresponds to a
pressure of three to fifteen PSI for a some predicted displacements of the flapper. The
above shows small change in the displacement, if you look at the graph the plot; that
means fall or the slope the above shows a small change in the displacement produces
large output change small change in the displacement produces a large output change.
Supply side restrictor we have load there are two type restrictors. Where the supply side
restrictors we have a load side restrictors load side; that means, on the flapper side; that
means. And supply side means which is as restrictors and from the supply of the air
which is coming in to the coming in to the flapper nozzle system. You see here
estimations of the parameters from the experimental flow. You see here we have mass
flow rate we have plotted Gs there is sonic flow or chocked flow and this subsonic flow
you see up to this. And after this it is falling down and Ps is a supply pressure Po is some
meet position of the flapper systems where the pressure is coming out. Now, what are the
legends? Let us look at.

Now, load side restrictor looks like this one. You see here for different position of the Xi
for different position of the Xi. If I if I change the Gu, if I change the Po how my Gu will
change. So, I plotted like this in this particular graph.
This is supply side restrictor same things again we have plotted slope is it is I am sorry, it
will be slope Pknx Gu, we have plotted and Xi is the displacement So, what is this
legend? Let us look at let us look 1 by 1.

In the neighbourhood of the operating point we can do a linear approximation; that
means, at the linear point; that means. If I look at, that means if I sorry if I look at that
means, the plot is I can take some other pages some other colour like this you see. So,
that restriction is line. So, I can approximate this positions and I can take this is a linear
region this actual operating point of this will corresponds to the 3 to 15 PSI of pressure
as a controlled output of the flapper nozzle system. In this neighborhood of the operating
point we can do a linear approximation. We can assume that the system is linear system.
Now, we are writing Gs equal to Gs Po equal to Ge. So, what are the legends? I will
explain plus derivative of Gs with respect to P o at P o equal to P o o multiplied by P o
minus P o o, which can be written as G.

So, plus Ksf Po, where now the legends we are explaining which corresponds to 3 4
slides which I have shown G. So, is the value of Gs at equilibrium operating point. Po is
the value of Po at equilibrium operating point. P o p is the small change in Po from P o o
so; there is a small change from the equilibrium point, because now, there is a
displacement. So, that is point I am the output pressures we are writing P o p. Ksf is the
value of dGs by derivative of Gs supply side I mean I mean restricted d p o at operating
point.
So, we can write Gus equal to Gu Po Xi equal to Guo plus Knp P o p plus Knx Xi p. Gus
is the value of Gu at equilibrium operating point and Xip the small departure from the
equilibrium operating point.
Therefore I can immediately write that Gu equal to Guo plus del of Gu plus d p o d p o. I
mean dell of I mean Gu with respect to Po at Xi 0 at P 0 0 multiplied by Po minus P o o
plus del of Gu with respect to Xi. One side we are making the derivative with respect to
this is our partial derivative with respect to the output pressures and other side with
respect to the displacement. So, that is we are multiplying the relative displacement or
change of displacement Xi minus Xi naught similarly here the change of pressure Po
minus P o o. What is Gu Gs already we have explained.

Now, we know that, that in a flapper nozzle system mass in which is mass in means mass
of air flow which is coming in minus mass out. Obviously, the additional mass stored
inside the flapper nozzle inside the nozzle inside nozzle system I should say. If I write
that is; obviously, I can immediately write the equations which is Gso plus Ksf P o p dt
minus Guo plus Knp plus multiplied P o p plus Knx Xip all multiplied by dt.
Now, from the Gauss theory we know that poV equal to MpRT naught so that we can
write immediately V by RT naught equal to dMp. Let us assume that the Gso equal to
Gu; that means, at the initially that the supply which is come Gso the mass flow which is
coming in and going out mass flow rate is same. See if it is same then; obviously, we can
assume that V by RT naught derivative of P o p with respect to dt t time Knp minus Ksf
P o p plus Knx into Xip equal to 0.
Let we assume some introduce let us some constant K equal to minus Knx upon Knp
minus Ksf. And tau time constant equal to V by RT naught Knp minus Ksf or; obviously,
if I make little manipulations. So, it will give you tau s plus 1 P o p equal to Kxip. So,
this will lead to a first or a differential equations or the I mean if you look at the response
P o p with respect to Xip will be equal to K upon tau s plus 1.
Thus, we see these above it a first order system. So, I can say that the flapper nozzle
system is a we can consider it as a first order system it is also a reverse kind of
transducers. That means, that if I forcibly if I change, because if there is a spring action
see if I change the pressure, so obviously, change of I mean bleeding pressure there
which is coming out. Obviously, the I will get a displacement there that is the reason we
call it a reverse kind of transducer.
It is very fast acting amplifier with good resolutions I mean that is if you consider the
pneumatic amplifiers. So, it is a fast acting amplifier with good resolutions sensitivity is
also quite high that we have already seen. And from the graph also we can predict that it
is has a good resolutions used in precision measurement; obviously, for very small
displacements I can use this flapper nozzle system. Because we can we have seen that we
are getting a large change in the small displacement we are getting large change in the
normalized pressure output.
Construction nozzle diameter is typically 1 by 32 in inch, which is 0.8 millimeter.

Volume is 1 inch cube which is around 16 cc it is quite small. Supply orifice diameter 1
by 64 inch or 0.4 millimeter, which is called fixed side restriction. Value of K is typically
8000 psi per inch, because this will give you a static sensitivity. So, if I draw the
electrical equivalent circuit of a flapper nozzle system we can represent it by a simple Rc
circuit. So, the analysis will be easier if I can represent by a simple Rc circuit, it is very
simple I mean very well resistance with that the…
If you look at the load side restrictor you will find that there is a very well resistance
which can I can replace by a potentiometer or a rheostat. And at the supply pressure
restrictors we can represent by a simple resistors and the volume we can represent
volume of the total volume of the flapper nozzle system, we can replace by a we can
capacitor. So, keeping all these mind we got the following electrical representations of
our flapper nozzle system which looks like this.
You see this is out flapper nozzle systems. This Ri and C Ri is the input side; that means,
supply side restrictor representing. And this is Ro is the load side restrictors I am giving
Vi I am Vo this is a electrical representations of our flapper nozzle system.
Where is Ro? Is a variable restriction Ri is the fixed restriction. What is the fixed
restriction? That is at the input the side the restrictions, which we are getting that is the
fixed restrictions. Then C is the capacity of the systems that is the volume, so these
actually makes our flapper nozzle system. The transfer function are easily we can find
the transfer function of the electrical system is given by V naught by Vi into s R naught
Ri plus Ro 1 upon 1 plus Sc multiply by Ro into Ri upon Ro plus Ri.
You see here let us I mean we can redraw the equation the flapper nozzle systems. We
have just redrawn we have change the position it does not matter it can be flapper can be
this side this can be that side. I mean the nozzle can be the supply air from this side or
that can be that side it does not matter much. You see the same the circuit I mean the
flapper nozzle diagram, because we refer to this that is the reason why I have redrawn
this in the figure 7.

Now, let us how will I compute the static sensitivity of the flapper nozzle systems. Let us
redraw flapper nozzle system as I told you in the figure 7. That just we have redrawn the
figure it is same as the figure 7 there is no difference. For static sensitivity calculation the
flow through the restriction is assumed to be incompressible flow through the restrictor is
incompressible in first we have assume. In many field dynamics we thought we have
seen that we have assumed like this 1.
Now flow through the fixed restriction is given by f 1 C d 1 pi d 1 square by 4 root over
Ps minus Po which is equation number 1. Where did I get this equation very simple I got
this equation from our basic orifice I mean orifice principles Bernoulli using the
Bernoulli’s principles we will get this 1 let us look at what is d 1 what is? You see d 1 is
the supply side restrictor orifice diameter and d 2 is the load side restricted I mean
restriction diameter which is circular in nature. So, Cd 1 is the flow coefficients on the
supply side and Cd 2 is the flow coefficient of at the load side of the flapper nozzle side.
If I have that thing; obviously, equation looks like this.
So, Cd 1 is the discharge coefficient of fixed restrictions; that means, supply side
restrictions Ps Po you know there is a output pressures and Ps is the our supply pressure.
As nozzle the discharges air to the atmosphere the flow through the nozzle is as the
nozzle flow discharges the air to the atmosphere again. The flow rate instead of
volumetric flow rate I am talking about is mass flow rate I am talking the volumetric
flow rate through the nozzle is look like, if 2 equal to Cd 2 pi d 2 into X dot into P
naught. This equation this side is slightly different. We have seen that this is instead of
we have taken the circumference multiplied by the x naught. the displacement that we
are taking at the bleeding volume. Let us look at a Cd 2 is the discharge coefficient of the
nozzle x dot x bar is the distance of the flapper from the nozzle. So, distance of the
flapper of the nozzle multiplied by this pi d 2 will give you the flow. If I have that thing
so; obviously, this is again I mean is coming in the meter square or something like that.
Now, I assumed that the condition of the flow continuity. What is mean by the flow
equation continuity? That means, if 1 equal to f 2 and moreover we assume that the 2
discharge coefficients will be same. Even though there is a lot of an assumption for
simplifications of our calculation, let us calculate Cd 1. That means Cd 1 equal to 2
discharge coefficients are same and also assume there is flow continuity is there. Then
equation 1 and 2 if I then combine this equation 1 and 2 we will get that 16 d square into
x bar square by d 1 to the power 4 equal to Po equal to Ps minus Po. Which will give us
Po minus Ps 1 upon 1 by 16 d 2 x bar d 1 square whole square. Let us look at how did I
get this thing let us take a blank page or let me take this.

It will look like this, you see f 1 equal to Cd, because we have assumed that Cd 1 equal
to Cd 2. We have assumed like this; ((Refer Time: 29:09)) that means, Cd 1 equal to Cd
2 and f 1 from the supply side equal to Cd equal to we write Cd this will be equal to Cd
pi d 1 square by 4 root over Ps obviously, will be more than Po, so Ps minus Po. And the
nozzle side flow rates we can write Cd again multiplied by pi d 2 2 pi r. So, instead of 2
pi r I am writing pi d 2 in to x bar x bar is the movement of the flapper from the nozzle
into Po. Since we know that f 1 equal to f 2, so we will get 16 d 2 square x bar square
into Po divided by d 1 to the power 4 Ps minus Po equal to 1 .
So, this will lead to 16 d 2 square x bar square Po by d 1 to the power 4 Ps minus Po. So,
I can write Ps minus Po by Po equal to 16 d 2 square x bar square Po divided by d 1 to
the power 4. So, this will lead to Ps by Po equal to 1 plus 16 d 2 square x bar square
upon d 1 to the power 4. So, I can write that Po by Ps I can take a new page.
Po by Ps equal to 1 upon 1 plus 16 d square d 2 here I write x bar here I write by d 1

square to the power whole square this is our equations. This is actually we have
explained in our ultimate equations; that means, normalized output how the normalized
output is relates to the displacement.
So, this is our expression 1 upon 1 plus 16 d d 2 plus I am sorry I think that will be 16 d
2 because if I take 16 inside. So, it will be it will be here, it will be 16 if I see it will be
here if I take 16. So, this will be…
Anyway let me write down this equation this will be let me take pen. So, this will be like
this 1 P naught by Ps 1 upon 1 plus 16, we will take outside. Then I will get d 2 x bar by
d 1 square to the power whole square. This is my final equation anyway as you have
seen.
Now, assumptions made for arriving at equation one and two I mean how there are what
are the assumptions basic assumptions. We have taken the velocity of approach is
neglected for small x bar the area of the nozzle outlet to the atmosphere is taken as pi d 2
into x bar instead of pi d 2 square by 2 pi d 2 square by 4. We have taken pi d 2 into x
bar because we have seen that it is variable. That is the reason we have taken like this 1
where as in the supply side restrictor we have taken pi d 1 square by 4 that is the simple
orifice formula. Now, we have taken the normalized response curve shown in figure 8.
What is the figure 8? Let us look at these are normalized response curve.
Let us look at very carefully this is our plot Pn normalized output by input P naught by
Ps X n X n bar. This is also we have plotted in the x axis which is d 2 x bar by d 1
square. This graph is very important this will control the entire position you see here if I
come down so; obviously, what will happen? The if I come to the close to the to the
nozzle when the flapper moves close to the nozzle; that means, you will find that output
pressure is increases which is supposed to be. This is a normalized response curve from
this curve we will do, so many other things.
The response slope dPn by dx n is approximately linear for short xn ad pn ranges what is
that? You see here, so dPn by Xn; that means, this slope which is supposed to be actually
if I say that it is this slope is actually if I take a thicker pen this slope is equal to actually
that dPn by dxn. So, this slope is actually will give you the response of the flapper nozzle
system approximate linear for short xn and Pn ranges for I mean small ranges we will
find it is linear. This very important we need a linear relationship from the beginning we
have seen that many all instrumentation cases we needed a linear relationship The
maximum slope for sensitivity is obtained by equating the d square pn which 1 basic I
mean basic elementary of our calculus. We know d square Pn by dx is x n square is equal
to 0.
Such that Xn bar equal to 0.144 that is the value for which we will get the maximum. So,
which gives dPn by dxn bar equal to 2 point minus 59 and this need this minus sign
signifies that the slope is negative. That we have seen from the graph the slope is
negative and Pn equal to 0.75 negative sign indicates the slope is negative.
When Xn bar is sufficiently large, Pn becomes constant. That you have seen when it is
coming, it is the constant right in Xn bar is how does it look? You see we have seen.
How does it look it will look like this 1 see here if I take a blank page. So, the slope will
look like.
So, if I have take a this 1 that already we have explained we can I mean if we do not take
that much. So, it will look like this 1. So, the pressure is Xn is faraway ((Refer Time:
38:02)) is faraway you will find that it is becoming constant. So, it is a pressure is; that
means, Po by Ps is constant here. If I take different pen, so right. So, it is coming like
this slope is looks like this 1. So, Po by Ps is constant to this value right when Xn is large
clear. Is to avoid this zero sensitivity portion that is lower and upper limits or bounds are
fixed to Pn equal to 0.75 normalized output is 0.75 and Pn equal to 0.15.
Because I cannot take Pn equal to; obviously, zero I mean I mean it is Pn 0 means it is
not possible. That means that these I am talking about um. So, I have taken a I mean
range of Pn equal to 0.75 to Pn equal to 0.15. For supply pressure of 20 psi we will find
that this will lead to Po to lie between 3 to 15 psi for Pn I can I cannot make it 0. So, if I
make it 0 there is a problem. So, that because we need a continues bleeding of air
through the nozzle. So, I cannot make it 0.
So, Pn I make the Ps 20 Ps over the supply pressure of 20 Psi this will be 2 3 to 15 Psi.
This is the key reason we will find in all the pneumatic system always we are getting this
value many a times we calling the industry standards is 3 to 15 Psi of the control pressure
where it came from the actually. This is the reason why it is I mean this is the point
where it actually came from, so 3 to 15 Psi, because why I want to make it within this in
range, because this will make take the linear portion of our graph. So, the flapper nozzle
system will be linear in this region that is the reason we are limiting this pressure.
The upper limit is fixed mainly by the pipe thickness the lower is not kept zero to enable
positive check on the control pressure on the low side as also to avoid the leakage in any
leakage from outside into the system. There should be always some bleeding. So, that is
the reason why we are saying there. So, there is no if there is not bleeding there is a
chance of the output pressure will come output atmospheric air will come inside the
system. So, that you do not allow. So, that is the reason here always leakage to always
there will be some bleeding of the air from the nozzle to the atmosphere.

Flapper nozzle with air relay this is another important thing in order for the flapper
nozzle to work satisfactorily the pressure changes must be small. We have seen that the
pressure change must be small I mean very small. This is by using an air relay an
excellent thing you see the how it works?
A control pressure of 3 to 15 Psi obtained from the relay by a process pressure change of
only 3 to 5 Psi and this involves only one-sixth of the original movement of the flapper.
So, I will strictly I am bringing the I mean, because I am saying that 3 to 15 Psi, but I am
saying that if I can even reduce the movement of the flapper to a smaller to this value
then what will happen I will get a less pressure. We will find that the pressure changes in
this type of situation say will be 3 to 5 Psi, but with the relay I will make it 3 to 15 Psi
which is a industry standard. How it looks like? Let us see here the flapper position is
directly proportional to the pressure change. In other words, the system is strictly
operated in the linear region the schematic shown in the relay is…
You see this is the relay position this is from the process there is a bellow you can see
here there is a bellow. So, if the pressure increases this flapper will move further here in
this direction. Let me take it this bellow will move in this direction the flapper, because
this is a fixed end of the below I mean the bellow the flapper will move in this direction.
Vertically it will move in this direction I mean horizontal If it moves in what will
happen? You see the more air will leak out more air will leak this will reduce the
pressure here. If we reduce the pressure this will what will happen? You see that
ultimately it will reduce the inside pressure inside this diaphragm gauge.
So, this diaphragm will come down if the diaphragm will come down what will happen?
You see diaphragm will come down. So, this ball will come down. So, allowing the more
pressure I mean the allowing the. So, allowing the more pressure to indicate so; that
means, the more I mean signal to the remote station. So, this is the basic principle like
other way if I if the flapper moves in this direction this will increase the flapper. So, it
will move this in this direction. So, it will move in this direction the supply less air will
come to the gauges it will show less, right.
Let us look at the air relay is reverse acting type. With the flapper moving away the
larger process when the process pressure is large from the nozzle the pressure Pi falls
when the diaphragm comes down along with the ball d. Pressure Pi this is the pressure
Pi, because it comes down the pressure Pi falls along with the ball d. So, it is closed the
path b again if I if I go that back it closed the path b. So, this path will go because this
will now valve since the pressure increases this will move out. So, this path there is a no
chance of air to come close the path b more and the passage between a and c is more
open. What is look at sorry passage between a and c is more open.
So, this will come down if the valve is coming down sorry if the valve is coming down
then what will happen? So, there is a less path. So, that less if the valve is I mean coming
down. So, what will happen? So, there is a less path to come here, but the more path a
and c will be more and more open. And the passage between a and c is more open giving
a larger pressure indication in gauge g 2. So, when the Pi falls when the g 1 is showing
some I mean lower pressure g 2 will show the larger pressure.
The readings in gauges g 1 and g 2 are always inversely related. It is the other way also
when the flapper is coming close when the system pressure is less then what will
happen? Flapper will come close then the valve will move the g 2 pressure will be high
pressure showing by g sorry pressure showing with the g 1 will be high. So, what will
happen? That the diaphragm will go up if the diaphragm goes up; obviously, it will show
the less I mean less it will restrict the air passage to the gauge g 2. So, that is the reason
we are telling that the g 1 and g 2 will be always inversely related.
So, g 1 and g 2 will be g 1 and g 2 will be this is g 1 we can see here this is g 1 sorry g 1
and g 2 will be inversely related this is g 1 and g 2 will be inversely related. With the
flapper moving away that is already we have explained. Now, as the output pressure Po
is diverted derived sorry derived from the supply pressure Ps the highest reading is
emitted by this factor. The air relay gives us amplified output for a small change in
pressure signal. Air relay will gives amplified output for a small change in pressure
signal that is the reason from 3 to 5 psi is I am getting a signal from 3 to 15 psi.
Now, current to pressure convertor I to P convertor as I told you this is very widely used
in industry. Current to pressure transducers are used primarily in the process control to
change a 4 to 20 million ampere of electronic signal from a electronic controller into 3 to
15 psi for a pneumatic controller. An effective I and P most incorporate the following
features.
What are the following features; it most provide air to the receiver quickly accurately
and in sufficient quantity. It must be able to exhaust air quickly when the signal
decreases sorry It must it should be consumes it should consume a minimum amount of
supply air for operation. It should be easy to repair and must be sufficiently rugged to
withstand difficult environmental conditions including variations dirty supply of air. That
is not very important usually we can filter the air temperature extremes and corrosive
conditions.
The flapper nozzle system fulfills the above all criteria that is the reason we are using it
because it is not very much effected by the temperature that is the most important thing
in the flapper nozzle system. Now, traditional flapper nozzle system is shown in figure
ten you see here.

So, this is the traditional flapper nozzle system. You see the current is coming through
the electromagnetic coil here current is coming here current is coming here right. So,
there is a bellows you see the, what will happen the pressure it will if it comes down? So,
this will increase the pilot pressure here or back pressure. So, the flapper now the
position of the flapper depends on the position of the flapper depends on the current. I
want to make I to P convertor ultimately the position of the flapper will be corresponds
to the pilot pressure this is the pilot pressure should be a function of the current. But the
current actually will control the position of the flapper right. So, this there is a hinge you
see here at the center point there is a hinge here the flapper can go this way or that; that
means, it can go this way or that way like this one. So, if it go moves like this and;
obviously, what will happen you see this is a flapper position will change clear.
The input current is four to 20 milliampere is applied to a coil-armature arrangement that

acts on a beam. Beam is now the armature or flapper whatever you call we are no more
calling it flapper we are calling it a beam. The beam positions itself against a nozzle . So,
positions itself a nozzle that has air flowing through it. The gap between the flapper and
the nozzle determines the back pressure also called the pilot pressure that builds up in the
nozzle. So, this is the pilot pressures the in the nozzles we have seen what is that pilot
pressures you see here the pilot pressure. So, this gap will control the pilot pressure the
same thing we have seen that it can be supplied to a relay to get a higher. I mean that 3 to
15 psi of pressures also you have a feedback relay also bellows also to neutralize.
The gap between the flapper and nozzle determines the back pressure also called the
pilot pressure and builds up in the nozzle right. You see the reason of the feedback,
because it is working like a spring, because if comes very close, if the bridge comes,
because I have a supply pressure, if it comes to very close to this one. So, this will
increase the back pressure because pilot pressures and this pressure is same. So, this will
increase the more pressures in the bellow. So, it will make the make the system. So, that
it will restrict the nozzle to come very close because it will work as a spring there right
that is the reason we have used a bellows here also. So, the gap between the flapper and
nozzle determines the back pressure also called the pilot pressure that builds up in
nozzle.
A bellows is sometimes connected a bellow I am sorry is sometimes connected to the

nozzle area to balance the forces on the armature-flapper. As I told you it is just like a
spring instead of spring I am using the bellows there which will control the position of
the forces on the flapper nozzle system armature-flapper or armature is a instead of
flapper. We are calling it armature or bar whatever you call it basic principle is same the
nozzle remains fixed only the armature is moving. The pilot pressure for linear operation
is channeled to a pneumatic relay or booster the pilot pressure or back pressure which is
coming that is going to relay for boosting operations right.
That means, it will be with the pressure is less three to fifteen psi will make a small
movements where the more linear region this will lead to after the once it goes to the
relay. So, I will get a three to fifteen psi of pressure. The pilot pressure the booster or
relay act as an amplifier translating the low pilot pressure into higher output pressures
and capacity. Because the capacity is also important, because if the capacity of this
system might be very, very small in this case. We have capacities we can use both of the
capacity and range I can go to the a booster circuit or relay circuit.
Mechanical I to P convertor; that means, current to pressure convertor that use a flapper
nozzle and which do not use electronic feedback sensors face some difficulty in dealing
with the environmental factors what are those factors let us look at. The flapper is
susceptible to vibration, and traditionally has forced users to mount the I P and current to
pressure convertors separately on a pipe or rack. This requires additional tubing to carry I
to P output signal to the valve the additional cost nullifies the benefit from mounting the
I to P convertor together in a common location.
The dead time and lag time introduced into the loop by longer output signal tubing has a
significant impact on the loop performance. Because whenever we are using some tubing
in pneumatic systems it is not very I mean predominant in the case of electrical systems.
But a pneumatic systems when using more tubing; obviously, this will increase the the
lag in the system as well the dead time in the system. Traditional I to P converters are
also adversely affected by fluctuations in air supply downstream tubing leaks
temperature changes and aging of the magnetic coil within the I to P convertor. Periodic
calibration checks are required in order to maintain the output of the I to P convertor
within the desired range.

Dirty supply air is the major cause of I to P downtime mechanical I P’s do not have
electronic feedback to compensate for partial plugging. But designing the nozzle opening
to be at least point 0 4 centimeter; that means, 0.015 inch in diameter this problem can be
avoided.
Now, there is new type of I to P convertors has been introduced let us look at that I to P
convertors thought there is basic principle is again the flapper nozzle system. But with
the help of peiso-electric systems they are used. Several new concepts have been
introduced these new concepts have change the nature of the pilot stage and have
incorporated sensor-based electronics systems. It includes some tradeoffs like between
efficiency and cost or improvement of one factor, which leads to the degradation of the
other one such important device piezo-ceramic bender nozzle. So, piezo-ceramic bender
nozzle is fine that piezo is you know that there are I mean we can your I mean natural
systems we have an natural piezo electric systems. We have a synthetic we have a piezo
ceramic also this are some advantage of a piezo-ceramic systems. So, let us look at that.
Now, piezo ceramic bender nozzles you will find that its device is shown in figure 11.
The unit does not use the coil to move the flapper which are used in the I mean
conventional I to P convertor. There is a coil actually which energizes and which current
is coming to that coil and it is giving the movement of the flapper. But instead the
flapper itself is made of layers of different materials which are laminated together these
different materials flex or bend when the voltage is applied across them, right.
The 4 to 20 milliampere of current input signal to the I to P is first converted to voltage

in the range of 20 to 30 volts dc right. Now, this voltage is actually given to the flapper
this design tends to be more stable in variations vibration. Then the typical flapper
armature and some of the drawbacks of this new concepts are…
You see that is will come later on, let us see this one this is our figure you say you see
here. So, we have a lead zirconate this is our ceramic piezo electric systems. So, now
first the 3 to 4 to 20 milliampere of currents we are converting to 20 to 30 volt dc is
applied. As you know that the piezo electric system is reversible systems if you apply the
pressures I will get a voltage if you apply the voltage I will get change in I mean
pressure. So, this since it is lead zirconate and brass we are using. So, it will try to bend
and this bends will be sensed by this you see this lead zirconate and brass. And this will
make the, this will work as a flapper that time and it will come to the either to the close
to the nozzle at pay off.
So, this will give you this booster this output this is the pilot pressure which is sensitive
to the, which is this pilot pressure or booster pressures or booster control. This pilot
pressure will be a function of this dc voltage we have applied to the flapper nozzle at this
ceramic. And this voltage again its I mean controlled to this function of the current
which is coming right. So, this will lead to you see the whatever the pilot pressures or
back pressures I am getting here that is the proportional or that is a function of exactly
the current which you are getting. So, many other I mean point of the I mean those we
have eliminated here in this because there is no coil as such. So, directly applying and
due to this bimetal sort of this is ceramic and brass; obviously, this will bend in one
directions or the other. So, when the voltage is applied due to piezo electric effect, so
obviously, this will make the this flapper close to the nozzle or not or away from the
nozzle.
The bender does not have a very good memory; that means that it has a creep which
memory means it is called in the systems and will tend to locate in a different position
for the same input signal. That means, creep is cumulative cumulant and eventually it
exceeds the adjustment range of the calibration mechanism.

An electronic feedback sensor can be combined with the piezo-ceramic bender to
compensate for the creep temporarily for the… But the feedback circuit typically uses
such more power available from the input signal having little to energize the bender.
The bender cannot balance against the force of the nozzle air unless the nozzle is kept
relatively small thus large nozzles must be traded for improved bender control. Plugging
of small nozzles is a chief reason for I to P field failure.
Force balanced differential pressure transmitter; you see this we have seen the 2
pressures are coming this side. You can see here two pressures I mean P 1 sorry this P 1
and P 2 is coming here from the process and which is coming for the from the flow. So,
this actually from this I am getting a supply pressure which is function Po after all
function of k P 1 minus P 2. Because always in the flow measurement we have seen that
in all differential pressure are based I mean flow meter. So, this type of output pressure
this will be this P 1 minus P 2 or Po will be calibrated in terms of flow. So, that we can
use a total pneumatic systems you can see this is you see there is another. This is the
feedback bellows we have used this is the bellows please note feedback bellows we have
used.
To block diagram this if I draw the block diagram it looks like this one. So, all the
significance is P 1 and P 2 the output pressure. And the output pressure is coming here 2
pressures on the upstream and downstream is coming here which is combined. So, we
are giving the relation like this one.
Say 1 Ab 1 Ab 2, the area of the bellows Kb the bellows linkage compliance Km the
flapper nozzle again.
And in the above circuit P 1 is always greater than P 2 then the upper bellow exerts more
forces. Then the lower one always it will be greater because upstream flow is higher than
the. Thus the flapper moves away from the nozzle and the pressure inside the tube A
falls. The feedback bellow cannot hold the flapper in such a position and brings it down.
So, the output resist pressure will come P naught equal to Ab 1 by Ab 2 I 1 by I 2 P 1
minus P 2 right. So, this actually will give you the direct this you see the output pressure
is again the function of P 1 minus P 2 which is actually calibrated in terms of the flow.
So, with this I come to the end of the flapper nozzle system.
Prof. A. Barua
Lecture - 34
Smart Sensors
Welcome to the lesson 34 of industrial instrumentation. In this lesson, we will study

smart sensor. Now, is the talk of today I mean when you call the smart sensors and
suddenly I mean there are sensors. We sometimes we call sensors sometimes we call
transducers as a whole sometimes we call instrument. And there was intelligent
instruments or intelligent sensor suddenly we started to call it a smart sensor right.
Obviously, there is, are differentiation between the ordinary sensors smart sensor and the
intelligent sensor. So, in this particular lesson we will look at all those and we more
discuss in more details I mean when the sensors can be called as a smart sensor. There
are some definitions even though that is not widely expected, but the some people who
actually developed some kind of smart sensors. So, well those are discussed they have
developed some norm. So, they have developed some terminology I mean on the smart
sensors. So, in this lesson we will discuss all this in details. Let us look at the contents of
these lesson 34 smart contents.

Contents; introduction to integrated smart and intelligent sensors there are integrated
sensors smart sensors. That means, if we have some sensors if the signal condition is
circuitry itself on the chip itself we call it smart sensor I mean we can call it integrated
sensor not necessary intelligent one right. So, it starts from there then the working
principle of intelligent sensors some examples of smart sensors these are basically we
will cover in this particular lesson right.
At the end of the lesson, the viewer will know what is meant by smart sensors, difference
between the integrated and smart sensors and some different fields of application of
smart sensors? I am sorry that there is some error it will be at the end of the lesson the
viewer will know. So, this is should note there. So, it should be wiped out like this. So, at
the end of the lesson the viewer will know it will like this. Now, integrated smart and
intelligent sensor. Let us look at all these in general then we can classify one by one,
right
The successful applications of many types of solid state sensors that means
semiconductor sensors solid-states. We are using that is semiconductor sensors has
stimulated the market, but led to a demand for either a lower unit cost or an enhanced
functionality to improve the market value. You see that if you say some smart sensors
and with a huge value nobody will buy. Suppose if you are telling about some
temperatures sensors which is based of semiconductor based which have integrated I
mean circuitry inside. So, that is a smart sensors with some intelligence functions, but the
cost if the cost is extremely high.
So, nobody will buy that type of sensors. So, they will rather use a thermistors and with
signal conditioning they will make it outside right. So, the cost factor is very important
while you are choosing the sensors or designing the sensor that is the reason we have
been calling. The successful applications of many types of solid-state sensors has
stimulated to the market, but led to a demand for either lower unit cost or an enhanced
functionality to improve the market value. See until and unless the functionality is over
suppose it can discard some value which is coming of the range that type of thing it has
some decision making functionality.
So, then only if the even if the cost is higher the people can buy or industry can buy
industry can install that type of sensor otherwise not. Both of this may be achieved
through a higher level of device integrations. Now, all these sensors we are talking about
nowadays it is SOC basically system on chip right which is or I mean entire sensors. It is
single conditioning circuits everything should be on the chip. Then only we can which is
called basically a SOC or system on chip then only we can say that I mean it is a smart or
intelligent sensors. If we had some intelligence and. So, large scale device integration is
necessary to make this type of sensors.
In addition more difficult sensing problems are now being studied which require a higher
level of processing power or more intelligence than is achievable from today’s sensors.
Suppose a simple thermocouples or a thermistor for making a measurements of some sort
of I mean the functions which we will get. Suppose a nonlinearity can be removed if
there is some signal which is of some other sort of a self test systems inside the sensors.
So, that type of things I can say it is intelligence smart.
A sensor output or there is some checking that the sensor is working or not which is very
common in the case of thermistor and the thermocouple. You know some sometimes
thermocouple is out of order I mean because the junctions, because it is a dissimilar
junctions. Sometimes the junctions are out or cut and thermistors sometime fails all those
things these things can be. Suppose if you have a built in self test; that means, some
testing algorithms or testing mechanism by which I can check the health of the sensor.
Then only I can say it has some sort of intelligent this is one form of intelligence sensors
or smart sensors I should tell. In order to define more carefully the different types of
intelligent sensor. It is necessary to subdivide the processor in to signal preprocessor
which is basically a converter and the main processing unit that is a microprocessors.
Obviously, any if I need some computations nowadays. So, I should have some form of I
mean CPU or central processing unit is necessary. So, one of the easiest way to use that
thing is either microprocessors or a micro controllers. You know 8 bit microprocessors
we know that very common form of microprocessors which has quite good. I mean
computational power which can which can I mean make some computations it can make
some decisions that type of things is possible. And more interesting thing is you see most
of the sensors what we are talking about is a is basically a analog I mean in nature. The
signals which is coming out from the sensors whether it is a thermocouple thermistors or
any RTD’s team basically it is a analog in sense. That means the output is continuous
whether it is a variations of the resistance or the variations of the current it does not
matter right. But if I now incorporate that thing in a digital form that means, if I want to
incorporate on a microprocessor; obviously, that comes that the mixed signal circuits
right.
Now, signal preprocessor or converter carries out low level tasks such as these are
basically low level tasks. Because I have a amplification filtering and analog to digital
conversions like this one, what are this? Let us look at…
That means suppose I have a signal which is coming from the sensor itself it is
generating some signal it might be low in level voltage level. So, I need an
amplifications, so I put an amplifier here. Then I may need a filter also if it is AC signals
I want to unbound it I mean undesired frequency band is to be discarded. So, I am using
a filter then I am using a A to D convertor right. Now, this for this can be a data
acquisition card also data acquisition systems or DAS. So, this is the entire systems this
should be already incorporated in a single chip right. Now, signal preprocessors or
converter carries out low level tasks such as amplification. So, this is a preprocessing it is
a filtering it is making A to D conversions; that means, analog to digital conversions.
Or then figure 1 to 3 illustrates the 3 levels of integration which make up sensor system;
that means, both the normal sensors intelligent sensors smart sensors. So, this 3 things
are I mean we have discussed in the we will in the following slides 3 slides what are
those actually means actually we want to meant right. There are different differing views
at present over the precise definition of the smart sensor. As I have told you it is a very
new thing and which is coming up there is no precise definition I mean why should I call
it a smart sensor? So, there are different views from one one differs from the other. So,
this is you see a typical sensor system without any intelligence without any smart test.
So, I have a, you see I have a non-electrical output. Let me take a another type of pen. I
have a non-electrical output see it might be pneumatic signals also this signal might be
pneumatic is not it? This I am signaled through pneumoelectric converters I am
converting to electrical signals right. Suppose in the flow measurements we have seen in
details in the flow measurements that is in pneumatic signals which is coming from the
differential suppose differential sensing devices. Now, we want to convert that to 4 to 20
milliampere. Electrical signal is coming now signal processors because this is basically
this output will be voltage.
Now, signal processor you see it is a entire unit right I am talking about just now I which
I have drawn; that means, it has amplifier it has filter that type of thing. Now, processing
unit not necessarily it can be inside the computer itself. So, this current usually this will
be 4 to 20 milliampere of current this will come. Then I have a processing unit
processing unit what will do it will convert to the digital domain it will take some
decisions I mean some it will display and all this things. So, there is no as such there is in
the entire scheme if you see there is no sort of any sort of intelligence or any sort of
smartness in this particular sensor scheme. These are basically sensors or I should say
not necessarily conventional sensors many modern sensor modern day senor might be.
Suppose a very extreme brand new a micro Do 2 sensors; that means, where I want to
measure the oxygen contents in human tissue which is very sophisticated sort of very
small in size I want to know that type of things. It is a very new type of devices, but not
necessarily that will be intelligent or smartness smart sensor. In that type of situation the
entire things I can define any sensors I should say can be defined by this particular
diagrams. These particular diagrams I can I mean sense any particular I mean any typical
sensors right. Now, this is integrated sensors in a sensor system.
Entire thing as a whole I can I am calling it sensor system because while I am calling it
sensors only. That is actually I am getting some measurement right on the particular
parameters might be pressure temperature flow. Now, when I am making the excuse me
when I am making the single conditioning when I am making a single conditioning that
times I can say that it is a as a whole sensor system right. So, here you see the sensors we
have non-electrical output input, because not necessarily from the all the sensors we are
getting electrical input right that might be non-electrical input. So, I have integrated
sensors I have a sensors preprocessor then electrical signals and processing unit and
electrical outputs what I am getting right.
So, there are preprocessors and sensors you see in a one. So, this is some for sort of
integrated sensors you can look at right. So, I have sensor unit I have a preprocessing
unit before though it is a current. Now which is sent to the processing unit or might be
the CPU of a computers through the data acquisition. And because at the data acquisition
card usually this is as I told you earlier it is a 4 to 20 milliampere of current. This is to be
converted in the voltage domain and processed through the data acquisition card this
signal will go the computers and the computer will process and I will get a electrical
output right. So, as such it is I should say it is integrated sensors; that means, the sensors
and the preprocessors on a single chip right. Now, a smart sensor or integrated an
intelligent sensors looks like this.
You see the sensor preprocessor processing unit, are all in single chain. So, this type of
sensors I am calling it smart sensors. Obviously this processing unit has some decision
making I mean functionality that there should be there otherwise I should not say a smart
sensor right. It can discard suppose a signal say I mean some data is coming which is out
of band. So, the system should be in a position to discard that signal not shown in the
electrical output right. That type of decision making algorithms or root functionality
should be there in the, but as a whole this is a smart and integrated sensors.
One proposes that the device on to which at least one sensing element and signal
processing circuit has been integrated is a smart sensor. Not necessarily, because when
the, it started people as started to define. Like that that one proposes that a device on to
which at least one sensing element and signal processing unit circuit has been integrated
is smart sensor where this definition is little vague I should tell. Since it is a newer
devices; obviously, you can understand that the, this is just developed definitions is not
very I mean well accepted I should say right.
The short coming with this definition is that the most sensors would be called smart even
when they possesses possess such a low level of intelligence that to call them intelligent
is rather ambiguous. Because this I cannot say I mean it is just a I mean a sensors which
is signal conditioning something I should not say intelligent. Because it is not a
intelligence it is a part of signal processing I should not say it is a intelligent sensor right.
So, if this definitions is not well accepted if you see you can say integrated sensor
obviously, but not necessarily you can say intelligent sensor.
Instead we will use the term integrated sensor to describe this type of low level smart
sensor where most of the preprocessor is integrated as shown in the figure 2. We have
already shown in the figure 2 that this all the preprocessing unit rather the preprocessing
units are integrated within the sensor itself. That mostly this type of things; obviously,
comes in the case of semiconductor sensors right instead we will use the term integrated
sensor to describe this type of low level smart sensor where most of this preprocessors
are indicated as shown in figure 2. We have shown already in figure 2 we can go back
this is the figure 2 right some sort of preprocessing is done you can see here some sort of
preprocessing is done right fine right. The term smart is reserved to denote the
integration in part or full of the main processing unit which adds the intelligence. That
means say when I am calling it smart; that means, it will be either part or fully
preprocessed. So, a smart sensors already we have shown in the figure 3, right.
There is little confusions in this practical definition because all smart sensor must be
integrated and intelligent. While any sensor that has significant intelligence which has
not been integrated can be called an intelligent sensor. So, smart sensor must be must
integrated sensors, but a smart sensor; if they are some sort of intelligence. But it is not
integrated I we just simply call it integrated sensor or intelligent sensors I am sorry. The
only definition that remains is that of an intelligent sensor the definition proposed by
Brekenbridge and the Husson considers artificial intelligence and run as follows.

They are telling the sensors itself has a data processing unit with automatic calibration
automatic compensation function in which the sensor detects and discards abnormal
values or exceptional values right. That I told you several times it incorporates an
algorithm which is capable of being altered. And has a certain degree of memory
function and further desirable characteristics are that the sensor is coupled to other
sensors adapts to change in environmental condition and as a discrimination function. All
this thing should be I mean this should other sensor there is a combination protocol all
those thing then we can call it intelligent sensor.
Not necessarily; that means, suppose I have a DMM a digital multi meters it can
nowadays digital multi meters in auto ranging auto zeroing. All those facilities are there,
but I should not call it a smart sensor or like that I should rather call it a not also
intelligent sensors. I can call it only the integrated sensors right because it cannot discard
the values which are coming I mean out of range and all those thing which I have just
talked about. You see here in the algorithm which is capable of being altered has a
certain degree of memory functions compensations abnormal values. Or exceptional
values can be discarded only then I can call it a intelligent sensor.
This definition is rather long, but does not incorporate the essential functions required to
define the intelligence in a sensor. What are the how will I call a sensor intelligence that
is will define in particular this definitions right. An intelligent sensor must possess one or
more of the following 3 features an intelligent sensor must possesses one or more of the
following features 3 features what are those? Perform a logical function, number 2;
communicate with one or more other devices. Perform a logical function communicate
with one or more other devices and make a decision using a crisp or fuzzy data.
That means too hot or hot too cold like this fuzzy data it can take some decisions right
only. Then I will call it intelligent sensor and also it should have communication facility
with the other sensors also not necessarily only one sensors. I should if there are ten
sensors you should have a communication facility and perform a logical functions that
should be all; that means, some form of arithmetic computation should be there.
Obviously I need a microprocessor or micro controller for doing this then only I should
say a intelligent smart sensor.
This definition distinguishes between an integrated sensor and a hybrid intelligent

sensors. It might be it is a hybrid in the sense that it is some form of intelligence is there
or may not be there. Now, logical function what are those logical functions I talked
about, because I said already that it is a should be a logical function perform a logical
functions. So, what is that logical functions? In order for an intelligent sensor to perform
a logical function it clearly requires some type of processing unit. Any logical function
as you know it is a whether it is a I mean adder subtracter. So, I need a some form of
logical I mean if you want to do I need processing. What is the simplest processing unit
available? It is a microprocessors not necessarily very large I mean very highly
computational processors.
Most of the applications we have seen in the instrumentations; that means, it is a 8 bit
microprocessors will solve our purpose. So, either microprocessors like Intel 8085 which
is a 8 bit processor as you know Pentium is a 32 bit microprocessors. When we call it 8
bit microprocessor 32 bit microprocessor please note actually we are referring to the
resistor size. Because all the, I mean in all processing in all this logical functions and all
this when you perform we are have to shift a load all the manipulations actually we do in
a resistor itself. There are several resistors depending on the complexity of the
microprocessors and these resistors. Actually the if you look the resistor size we can tell
what type of bits I mean how many bits?
When we said the 8085 Intel 8085 is a 8 bit microprocessors we can look at the resistor
size we will find all the resistors are 8 bit. Whereas, when I call the Pentium processor is
a 32 bit processors if you look at the resistor size of the pentium. Number I do not
consider there are several numbers there are or the number of resistor differs from the
8085 8086 and a and the pentium processors. But this size of the resistor will tell whether
it is 8 bit processor or a 32 bit processor. But most of the instrumentation applications we
will find that this 8 bit processor will suffice right. And if you need some peripheral
chips in most of the cases we need some peripheral chips we can go for a micro
controller micro controller is.
Difference between the microprocessor and the micro controller is that it has some in
built chip which is necessary because in 8085 a microprocessor a processing unit cannot
work until and unless it has some peripheral chips. Some sort of RAM to a minimum
RAM is necessary 2 kilo RAM is necessary 2 kilo byte RAM is necessary some ROM is
necessary. This is a minimum thing is necessary in the microprocessors I may need some
counter or timer or counter. These all those things are incorporated in a 8 bit
microprocessors like 8051 or 8091 this are basically Intel based microprocessor. This is a
8 bit microprocessors it has 120 byte of memory RAM very small, but most of the
applications you will find that is enough right.
So, that is the logical function in order to for an intelligent sensor to perform a logical
function it clearly requires some type of processing unit. The processing unit it most is
most likely to be a microprocessor, but could be a some type of programmable logic
devices. PLC also we will do, but that will reduce the cost, but nowadays as you know
microprocessor cost is not very much. So, instead of programmable logic controller you
can use it, but the microprocessors in most of the cases will work very fine. Because we
know how to programs and all this things microprocessor has some language that will
simplify the process.
The integration of the processing unit with the sensor then requires a significant
electronic design effort. So, this basically most of the sensors I talked about the
intelligent smart sensors is basically electronic sensors or a semiconductor sensors and
effort perhaps through the use of application specific integrated circuit or ASIC which is
called applications specific integrated circuit technology. So, lot of integrations are
necessary, if you want to make a intelligent or smart sensors.
An intelligent sensor may be able to communicate with it is operator. So, provide

valuable information about the problems etcetera. Alternatively an intelligent sensor may
communicate with the another device. So, modify it is own behavior right. Intelligent
sensor may be able to communicate with it is operator and. So, provide valuable
information about the problem etcetera. Alternatively an intelligent sensor may
communicate with the another device. So, modify it is own behavior.
A type of intelligence can in it is simplest form provide a warning of abnormal operating

conditions or more cleverly provide a feedback control mechanism. That I told you; that
means, suppose a thermocouple is not working. So, a temperature sensor is not out of
order. So, intelligent sensor should be even in the position to tell that the sensor is not
actually working right. That means a some sort of test mechanism should be there in the
inside sensor itself or in the integrated chip itself. Intelligent sensor may provide a good
level of control such as via digital controllers that is possible if you use a feedback
systems and all those things. An intelligent sensor may have some form of high level
adaptive control strategy that permits the control parameters to be automatically updated
with time. So, you know the controller parameters those who are studying control you
know. So, you can update automatically. So, that type of situation should be there. It will
have some I mean it will predict something or it will get some those data from as the
time changes. Then only it will update those controller values then only I will call it
intelligent sensors.
The implementation of a sensor which can warn it is user or adapt to environmental

conditions requires some decision making capability right. Now, it is very difficult
suppose I have a Wheatstone Bridge, we have studied that suppose I have I am
measuring the I mean load by a strain gauge right. Now, as you know the strain gauge is
basically resistive based sensors. Now, if the temperature changes environmental
temperature changes usually what we do? We connect the, that with a strain gauge in a
Wheatstone bridge either a single strain gauge or 4 active strain gauge it does not matter.
Now, if you have a single strain gauge as the temperature changes environmental
temperature changes there is a change of resistant, but I am no I do not want that.
Because the unbalanced voltage of the strain gauge of the Wheatstone bridge will be
different. What I want that the output will be only it should be dependent on the
unbalanced voltage of the Wheatstone bridge and basically that depends on the change of
the resistance in the strain gauge itself right. So, simply how we will do we use an an
ambient temperature we use another strain gauges. So, for ambient temperature
compensation, but that I should not say intelligence sensors. Just it is one to one
correspondence that if that the both the resistor changes due to temperature change that
will be nullified. Some sort of the decision making system algorithm should be there in
all the intelligent sensors. Though I am saying to adapt a environmental condition
requires some decision making capability there should be some decision making
capability there.
Traditionally sensors use a parametric data to make a decision. For example, the signal
from a diode can be used to provide a overload protection to a device which exceeds it is
normal operating temperature. Diode can be used as you know I mean it is a used in
many places we have seen for overload protections, because in the op-amp also, as you
know in the op- amp the output is, if you accidentally connect the op-amp output to the
ground. It is not that op-amp output will op-amp will burnt out there is a short circuit
protection in the op amp right. Because with a diode and as you know that in a series
regulators we are also protecting the junction which is they are costly transistors series
transistor by diodes also. Because if you bias the diodes always you will get across the
diode point seven volt, right. So, that can be utilized to protect your circuit right. So,
traditionally sensors use parametric data to make a decision for example, the signal from
a diode can be used to provide an overload protection to a device which exceeds it is
normal operating temperature.
However, more intelligent sensors of the future may use a nonparametric methods such
as artificial neural networks and or expert systems which relies upon fuzzy data. Expert
system as you know is a program and is based on an IBS program it has some form of
form of decision making algorithm. But as you know this when you are this is all
basically a program it is not necessary it is a software when you want to I mean connect
it to a hardware devices to a sensors. So, I have to make it some microprocessors or some
micro controller where this decision making algorithm like neural network. Neural
networks also as you know it is very it is not very cost effective. If you want it is cheap
over it will extremely high if you want to realize a neural networks in the wafer.
So, if you want to I mean use the neural networks synthetically or artificially neural
networks; obviously, I need some form of computations or some form of CPU. So,
microprocessor are best thing which you can give that the definition of an intelligent
sensor used here includes any sensor system that contains a discrete microprocessor unit
right. Either a discrete microprocessor unit, because you see that is a SOC or mixed
signal circuits. Because mixed signal circuit is a circuit where both you have a digitals
and analog on the same chip right.
Now, it is very talk of today as you know this mobile phone and all this things are
basically a mixed signal circuits, but it is quite expensive also. So, if I take out this
microprocessor unit I mean then we can call finally, it as a intelligent sensor, but in
future; obviously, we will try to make the microprocessor units with digital circuits some
analog sensors also at the same chip. So, that it will lead to a a mix signal smart sensor.
Consequently there are a large number of instruments that are classified as intelligent
sensors.
A few is examples of intelligent sensors and their applications are listed in table one
examples of intelligent sensor system radiant. The sensor class the sensor class principle
intelligence and monitors the special Fourier transform of the retro reflected light from a
surface is basically machine tool control.
Sensor class mechanical monitors the generation of acoustic noise by cracks in a metal
the non-destructive testing. Sensor class is mechanical effect or sensors for handling
objects for robotics it is for robotics. Chemical monitors air intake and fuel combustions
in engine control right.

Sensor class chemical monitors air condition for comfort and safety in buildings.
Magnetic monitors eddy currents to measure the proximity defects of plate thickness is
industrial automations. Now, integrations of signal processing.
The integration of the preprocessors and the perhaps the processor itself on to a single
chip can bring many advantages. Why we are calling it I mean willing to go for
intelligent sensors it looks like this. The most strong argument for the fabrication of an
integrated sensor is to enhance the signal to noise ratio. Obviously, if you want to
transmit this signal to the outside world or to 2 separate chip signal to noise ratio will not
be very good. But on the same chip if you may figure obvious a signal to noise ratio or it
is an R which is called which is usually expressed in db 20 log 10.
So, it is; obviously, better. The electrical power output from a micro sensor is often low
and susceptible to the stray capacitance inductance and noise. You see these
microsensors are very, very small in natures and one of the one of the greatest problem is
the capacitance. So, if you want to connect to an external world by wire. And all those
things that will be in the capacitance of the wire itself will play a picture, because those
values of the sensors of the capacitance sensor is extremely small.
In this cases the direct transmission of the output signal down a long interconnecting
wire may be impractical, right. If you want to do, because it is a microsensors that is say
it has a very small value of the capacitance change might be. If you want to transmit that
signal to some bridges for measurement of the capacitance change or accordingly some
voltage change it will be very impractical. This is particularly true when for example, a
capacitive pickup is used in a micro sensor.
Capacitive pickups is used for many as you know for the one talk that is our capacitor
microphone is there capacitance are used for level sensing. And so many thing
capacitance are used for the differential pressure measurements in all this, but if the
change of the differential pressure is very small. So, capacitance change also will be
very, very small usually these capacitances are order of femtofarads. So, any I mean
stray capacitance which is coming of the same order that will totally I mean change the
value of the output voltage. The capacitance of a lateral resonant silicon structure may be
only a few femtofarads.
Thus on chip MOSFET circuitry is highly desirable in order to remove the effect of the
high input capacitance of transmission cables and subsequent instrumentation. So;
obviously on a chip itself, if you can do it; that means, MOSFET circuits remove the
effects of this all this interconnecting problem or the interfering of the parasitic
capacitance will be immediately removed. Clearly this will improve the response
sensitivity and resolutions of the microsensors.
Resolutions means the smallest detectable values which can detected as output smallest
change of the parameters which I am going to measure which will make a detectable
output right. So; obviously, this will because if the entire thing is I mean I the entire
thing is camouflaged by the noise itself . So, that is I would not get the output. There are
other reasons for integrating the signal processing unit with the sensor such as providing
an enhanced functionality at a lower cost size and weight. Cost will not be that low
obviously, but the size and the weight; obviously will be much, much low.
Now, self calibrating microsensors this is another important thing for the smart sensors
we will look at. Most sensors have a at least 2 parameters that need to be set during the
manufacturing process this is offset and the sensitivity or gain. Usually we do any
sensors if it is not simply passive type of thing, suppose a ammeter or the voltmeter.
Then; obviously, we offset if there offset off set might be the zero offset also usually we
do we set it before taking a reading and adjust the gain also. If it is a active sensors; that
means, it has some signal conditioning circuitry along with the sensor. So, that type of
cases we can adjust the gain. A gain is basically as you know we have defined in the first
class it is also sensitivity. In the mass production of a sensor the process of calibration
individual sensor is expensive and undesirable, but often essential right. So, individual
calibration is not possible, but it is sometimes it is very essential specially in the case of
semi conductor sensors.
This cost can reoccur during the lifetime of the sensor because of the parametric drift,
because this if the parametric drift. From there if you have; obviously, and specially it is
very difficult in the case of as I mean integrated sensors where you do not have much
provision to change any I mean parameters right. So, you have to either you have to
through it out that is the only thing, but you must know whether you have sensors are
perfectly working or not. So, some sort of testing unit should be there inbuilt. So, that by
which I can know the health of the sensor itself, whether the calibrations which I have
done at the initial stage of the sensors that is now still valid or not otherwise entire
erroneous I can throw it out.
But the thing is first of all I must know whether that calibration whether I which I did are
still valid or not. So, maintenance work is carried out to recalibrate manually the offset or
gain right. So, it must be laser trimming that type of thing which is very difficult to do
consequently there is considerable need for a self calibrating sensors which can carry
out. It is own calibration and this is particularly true when a high level of precision is
required when a high level of precision is required. So, this is true even though that is are
quite costly.
The conventional calibration of a sensor may involve the laser trimming on integrated
resistors. Integrated resistors to get a typical values. So, I can laser seam it to change the
values. So, that the, suppose due to drift suppose the resistance value changes. So, I can
make laser trimming to bring it to the value. So, that my calibration will remain valid this
means that the resistor film and the patterning process must be comfortable with the IC
and the micro sensor technology. Typically resistor films like chromium and nickel
chromiums are vaporized from the substrate by a YAG lasers in controlled cuts or trims.
So, that I will get the precise value of the resistance.

Now, schematic representation of the signal processing in a smart capacitive pressure
sensors. That you see what I have a sensor unit here which is variable I have a
capacitance to voltage conversions basically it will be capacitance to volt. Because all
capacitors ultimately if you measure you put one bridge I will get a unbalanced voltage.
So, this is a bridge circuit, so capacitance of voltage conversion I have a signal
conditioning. Now, I have V reference voltage you see here see any sensor. So, in this I
will create a smart sensor this entire thing I should say this entire thing I should say a
smart sensors. This I should say a smart sensors only if I have the capability of changing
this calibrations of this all this 3 units. So, for that reason there should be some sort of
intelligence for the measurements of the output voltage. From time to time I should
check accordingly I will change this parameters.
So, that the whatever the output desirable output for a particular change in capacitance
will remain same as before only that time and that time I could call it a smart sensor. So,
that is the thing reference voltage some sort of you see I am changing here offset and
gain of the signal conditioning circuit. If there is already due to drift this that it has
changed I will change the offset I will change the gain. So, I must know whether my
calibration is still valid or not right. Figure 4 shows a block diagram of the signal
processing that takes place in the capacitive sensor and it is associated electronic
circuitry. That is we have just shown let me see it again schematic representation of the
signal processing in a smart capacitive pressure sensors. You know the pressure sensors
also we have seen that thing; that means in the differential pressure, because why we
have we have seen that we have in that I mean we have a differential pressure sensors.
Let me take a black pen. I have a still diaphragm this you have already discussed right
still let me just make a recapitulations. So, this pressure is P one this pressure is P 2 if the
P one is greater than P 2 the still diaphragm will move like this one, is not it? And this
inner side is coated with some metal preferably gold and that type of things inner side it
is basically is section of a section of a sphere this is a section of a sphere. So, this inner
coating and this plate will make one capacitance. So, if there is a change of capacitance.
So, there is a change of the capacitance differential capacitance because this at the
middle it will be C.
Now, when it is moved like this one. So, this is if I say that this is C 1 and this is C 2, so
C 1 will be previously C. Now, so if the D is increased, so what will happen if D is
increased capacitance will be decreased say delta C and C 2 will be C plus delta C is not
it? So, this is our actually you see that capacitance to voltage. Now, if I put on a this in a
Wheatstone bridge this 2 sensors in a Wheatstone bridge. Then what will happen? I will
get a unbalanced voltage. So, this is a capacitance to voltage conversions. So, and also,
from pressure to capacitance from, then capacitance to the voltage, so from pressure
differential pressure to capacitance, change from capacitance; change to voltage change.
This capacitance this is the capacitance change to voltage change this one. This is one is
basic sensor which I am giving pressures I am getting a signal which is you see the 2
signals are coming because there are differential in nature. So, the I am getting a signal
which is corresponds to the measure. So, I am getting a change in this capacitance which
corresponds to pressure right.
The capacitive pressure my pressure micro sensor has an on-chip CMOS switch-
capacitor circuitry to perform the accurate capacitance-to-voltage conversion, because
you know that this can be done with a switch capacitor circuits also. Accurate
capacitance to voltage conversion signal conditioning and automatic compensations for
the device temperature using a relative voltage. The advantage of the switch capacitor
circuit it is performance depends totally on the ratio of the capacitance rather than the
absolute value of the capacitor that is the reason switch capacitor circuits are preferred.
Though there are some limitations of the switch capacitor circuit like parasitic
capacitance. But we have some configurations where I can use the switch capacitors in a
parasitic insensitive structure right. Basically function is something like this that I have a
I can take a blank page.
If I have a switch a capacitor like this sorry if I have a capacitor like this one and I have a
switch which is toggling between this 2. So, this resistance this capacitance can be
replaced by a resistors. So, the resistors can be simulated by a capacitance and 2 more
switches clear. So, this will give you the resistor simulation. The advantage of this now
this is a parasitic sensitive I can have a I mean in parasitic insensitive if I have a structure
like this one if I have a structure like this one this is parasitic insensitive structure right.
We can have parasitic in I mean I mean I mean we can have the switch capacitor
integrator also I am not going to that details about those things. That is you see the
CMOS switch capacitor circuit to perform the accurate capacitance to voltage conversion
signal conditioning and the automatic compensation for the device temperature using a
reference voltage V reference. By changing the V reference I can change the automatic
temperature compensations and all those things.
The temperature of this smart sensor d alpha V is related to the reference voltage V
reference by the relations V reference V naught 1 plus alpha v T minus T naught and
alpha v is programmable with a 3 bit resolutions. Now, self testing microsensors.
The ability of a sensor to test the functionality is highly desirable right whether it is
working properly. As you know I do not know whether I mean that when you switch off
when you switch on a microprocessor or you switch on a Pentium processor usually the
Pentium tests all the, I mean logic conditions by a built-in self test systems. Which will
tell you which will it will load all the accumulators with all zeros if it does not load then
you know there is some problem in the chip itself. So, it do it usually do it very fast with
a nanosecond. So, you would not be it is not realizable to you, but actually this is they
are doing this is called the built in self test. Recent developments in the in the field of
smart sensors are leading to sensors with some limited diagnostic capability. This is
basically an ability of a sensor to determine whether it is a functioning normally or not.
A complete failure would usually be detected by the user as the output either current or
voltage falls below the operating specifications. So, this is another saying I mean we
have we should have some mechanism when it falls below some values. So, we can say
there is a failure. There are various scheme of determining this type of failure of the
sensors it is a different testing scheme of the sensors or functionality testing of the
sensors. Suppose when you buy a sensor; obviously, we check something for this voltage
whether it is showing this current or for particular temperature it is showing the proper
value of the current. If it is 4 to 20 milliampere output all those things these are I mean
parametric testing right functionality testing. In many cases a sensor can fall fail to
perform adequately, but provide a reasonable output many cases I may fail to perform,
but provide a reasonable output right? In these cases more sophisticated quality
assurance is necessary, because it may not it is giving some output for particular cases or
it may not get the correct output if it fails. I mean it may not get for some other value of
the sensor input, right.
For instance a noise characteristics power spectral density may be related to the physical
property of the sensing element that is changing and affecting it is performance can thus
be used to provide diagnostic information. So, diagnostic sometimes you know that I
mean one of the thing that in a integrated circuits; that means, the bias current changes.
Where the bias the current of the circuit changes abruptly then I can say there is some
problem in the circuit this is some sort of diagnostic informations which you can get. We
can continuously monitor this bias current of the circuit and predict whether the, your
circuit is healthy or not. A significant level of self testing has been reported on a smart
micro accelerometers in which the basic mechanical performance can be tested routinely
and thus diagnosed for the faults if there is some faults. So, I can detect it.
Now, built in self test will be incorporated in future microprocessor. Built in self test; as
I told you microprocessor Intel from 80386 onwards till Pentium processor they have a
BIST or built in self test where you have the on chip devices which you check the health
of your or test of your devices. Here in the case of micro I mean sensors it will there
should be some mechanism by which or built in systems on the same chip by which I can
check the whether the, my sensor is or not. Multi-sensing smart sensors often improve
their performance through the use of other sensors to monitor undesirable dependent
variables multi-sensing is another form it is a redundancy basically. So, if the sensor is
not giving correct output if the nine sensors giving correct outputs. So, I can say this that
this particular sensor is not correct. With all the nine sensors are giving somewhat I mean
similar input output of this; obviously, I can say the sensor number 10 which is not
giving some abnormal output this is some sort of redundancy.
For example, the temperature sensitivity of a micro sensor could be compensated by

integrations of a thermo diode near the sensing element. A smart sensor can also
eliminate spurious rather than the systematic erroneous signal. For instance an array of
identical sensors can be employed and coupled to a microprocessor which calculates the
average sensor output or perhaps discards any anomalous reading in a voting logic that
we can also incorporate. In a if you have microprocessor computation mechanisms I can
do it with the proper program.

The former approach has been adopted to make a smart pH sensor with 10 identical
sensing element. Integrated sensors which may also have some higher level of processing
capability or intelligence. For example, the use of an array of dissimilar chemical
microsensors and artificial neural processors which can be incorporated by a
microprocessor has led to the development of intelligent electronic nose. Electronic nose
means it can smell. So, this basically will give you the, I mean some form of intelligence.
Communications the read out electronics may be accommodated in a smart sensors. The
simplest form of a read out circuitry is often an current or voltage output. For example,
the 4 to 20 milliampere current loop is common standard and provides good immunity to
noise. You know the voltage in most of the, I mean the instrumentation cases we will
find always we convert this voltage to current because it is immune to noise. So, the
noise immunity is very important while you are communicating from one sensor to other
communicating from the sensor to the, your PCs or computers. From the sensor to the
transmitter usually all the transmitter transmit current not the voltage right because
voltage will be easily corrupted by noise.
A second approach is to use a frequency modulation because frequency modulation will

be totally different. At the out dissipating end also will be I will receive frequency. So,
there is a chance of, because even the amplitude is getting I mean corrupted or I mean or
it is totally camouflaged by the noise it does not matters. Because at the output at the
receiving end I will look at the frequency and keep them high if I have a demodulator
there I can get the actual signals right. For example, resonant microsensors produce an
oscillatory output signal which can be either counted or converted to a voltage by a
circuit integrated on to the sensor chip itself right. Moreover the signal can then be
converted to a digital signal on the chip and the digital output interfaced to a bus
systems. Because these all you see that is I am talking about a mix signal circuit; that
means, they and the same processors I mean everything should be there. That means, the
digital circuit is there analog sensors will be there.
These functions are commonly integrated onto a smart sensor although the reasons for
doing. So, are often scientific rather than the hard commercial decisions. It is not very I
mean it is acceptable by the industry and people if you make that type until and unless
you are well justified. Because the cost is a great factor for making such type of sensors
you have to look at the market potentials also. One important considerations when
deciding whether to fabricate an integrated sensor is the compatibility of the materials
processing required in particular the temperature range over which the technology bases
operate. Which is shown in the figure 4 sorry it will be figure five rather I should tell. So,
it is not a figure 4. So, this will be this will be figure five rather this will be figure 5, I am
sorry.
Relative cost of the smart sensor IC technologies as a function of the processing

temperatures right. Nowadays, you know you need to always I mean a very wide
temperatures I mean devices some sometimes some people needs however, if the sensor
can work in a range of minus 50 to 150 degree centigrade or minus 190 to kinetic
temperature that type of things. So, this all demands I mean various wide temperature
functions. So, relative cost of the smart sensor technology as a function on the processing
temperature.

Bulk CMOS is relatively inexpensive, but it is limited to a low temperature range and
maybe unsuitable when some inorganic active layers are required. In contrast silicon on
insulator SOI can withstand a higher processing temperature, but it is a much more
expensive process. Now, applications of smart sensors.
The field of radiations microsensors has had some notable successes cameras with on
chip amplifications and temperature compensation are on the market. Smart optical
sensors are relatively easy to make, because the silicon sensor technologies compatible
with the IC processing.
CCD images image sensors with integrated image intensifiers have also enjoyed some
commercial success such array sensors are; however, very expensive. Smart mechanical
sensors such as pressure sensors were initially fabricated at a relatively high cost and in
low volumes. However, the interest of the automotive industry in the low cost
microaccelerometers for the air bags has changed this situation now you can make it.
Even though the cost is quite high, but it does not matter because automative industry
needs that type of air bag which can deceleration which can sense the high speed
deceleration. Because whenever there is a there is an accident or there is a I mean there is
a crash between crash between the cars with another car or anything or any other hard
object. So, there is a tremendous amount of deceleration. So, there should be some
sensors which will decelerate and activate the air bags. So, that the persons who are
driving who are sitting in the front seats or sitting or driver as well as co-drivers. So, they
can save their life.
Smart silicon Hall Effect devices have been fabricated include a built-in offset for the
null voltage and internal temperature compensation. Perhaps the most difficult, but
potentially most rewarding development in the field of smart sensors will be in the
chemical and biological sensing.
Traditionally chemical sensors suffer from the various problems such as drift in any
offset parameters degradation of sensitivity due to poisoning interference from humidity
and other chemical species. The design and the integration of the intelligent processing
unit algorithm biochemical sensors micros could correct these deficiency and would
create large markets in fields such as environmental monitoring and mechanical
diagnostics. With this I come to the end of the smart sensors.
Thank you.
Preview of Next Lecture; lecture number 35; Chromatography. Welcome to the lesson 35
of industrial instrumentation. We will start to discuss the chromatography in general
actually we call it gas chromatography though it is more popularly known. But it is
actually it is a chromatography because the chromatography is both for the gas and the
liquid. So, in this lessons and subsequent lesson; that means, lesson 36. We will consider
the gas chromatography we will consider the chromatography. Let us look at the contents
of lesson 35.
Contents; we will basic have a introduction to the systems what the actually
chromatography look like. What is the packing material is used and what is actually
where it is used? All those things we will discuss. Now, the chromatography is basically
the separations of the fluids, because in many industrial situations as well as in the
chemistries we need this type of situation when there is a mixture. So, we want to know
it is a very difficult to know the content of the percentage content of the mixture of the 2
gases.
Thank you.
Prof. A. Barua
Lecture - 35
Chromatography – I
Welcome to the lesson 35 of industrial instrumentation. We will start to discuss the

chromatography. In general actually we call it gas chromatography though it is more
popularly known, but it is actually it is a chromatography, because the chromatography is
both for the gas and the liquid. So, in this lesson and subsequent lesson that is lesson 36,
we will consider the gas chromatography; we will consider the chromatography. Let us
look at the contents of the lesson 35.
We will basic have introduction to the systems what they actually chromatography look
like? What is the packing material is used and what is actually where it is used to? All
those things we will discuss. Now the chromatography is basically the separations of the
fluids. Because in many industrial situations as well as in the chemistries we need this
type of situation when there is a mixture. So, we want to know it is very difficult to know
the content of the percentage content of the mixture of the 2 gases.
So, in that case we can apply the chromatographic method to separate the 2 gases. And
find the percentage content of the individual gases or individual components when it is
more than 2 and 3 components present in the gas. So, basically those things we will
discuss how it works actually what is the illusions? We are talking about and how
actually we will know from the peak of that the percentage of the individual components
of the fluids I mean exists in the gas. So, those things we will know by in this particular
lesson. Introduction then you has basic chromatography. We will discuss the basic
chromatography here.
At the end of the lesson, the viewer will know: what is the packed bed column and open
tubular column? Because we will see in the chromatography, we will find that the basic
principle is something like this. That we will send the mixture of the gas or fluid through
the through a column and it initially absorbs the both the thing absorbs or adsorbs the
both the both the mixture. I mean the mixture, but we will see that depending on the
characteristics of the individual component. It will elute that mean it will relieve or it
will release the gas or fluid with the difference of time. So, that by that actually we can
know we know we can measure the contents or we can measure the concentration of the
gas.
So, this is the basic I mean philosophy of the we have different detectors. All those
things will be discussed these lessons as well in the lesson 36. So, we need that I mean
the column react I mean those tubular column or packed bed column, so through which
the mixture will be passed. So, we will discuss what is the packed bed column and open
tubular column? Then we have what are the different partition forces that are present
between the solutes? Terms like partition coefficient retention times selectivity factors all
these things we will know in this particular lesson.
How does the plate height we will see, what is the height plate height and the various
other factors affect the column efficiency? Because efficiency column is very important
not only to know the contents of the individual components of the gas, because it
depends on the material of the column. So, efficiency of the column it will also
distinguish between the 2 contents 2 components very nicely. Otherwise it is very
difficult to find the percentage contents of the individual components in the mixture.
Quantitative analysis of the chromatogram how we can make quantitative analysis of this
gas chromatography or chromatogram? The plot which will get from this
chromatography we call it chromatogram.
Introduction; chromatography is a method used to separate or to analyze complex

mixtures. The components to be separated are distributed between 2 mutually immiscible
phases; stationary phases and mobile phase which are brought into contact. So, there are
2 phases these terms we will very frequently referred throughout this lesson. That means,
in stationary phase and in the mobile phase. The stationary phase forms the bed and the
mobile phase percolates through it. So, we have a bed this is the packed bed column and
we are talking about we are, so that stationery phase will make that bed and the mobile
phase will percolates through it.
Why we use chromatography? Why do you need this gas chromatography or

chromatography in general? Let us look at. It can separate complex mixtures with great
precision. This is most important thing. I have 2 mixtures or 2 or 3 gases in a particular
mixtures. I want to precisely know; that means, first of all you have to separate. If you
cannot separate precisely you cannot measure the contents of the individual gas. So, the
precision in separation is also very important in the case of gas chromatography. So, that
is the reason I am telling it. It can separate complex mixtures with great precision.
Even very similar components like proteins varying only by a single amino acid
composition can be separated. This is most important it will I mean in many cases we
need these type of separations. So, different components with different chemical
compound with even with the single amino acids differences can be separated. This is a
major contribution since before the chromatography separating these biological
compounds was extremely tedious and time consuming and unreliable. Chromatography
is quite reliable ((Refer Time 06:50)) that since.
Chromatography can be used to separate very delicate products. Since the conditions
under which it is performed are not typically very severe very delicate products also.
That means, it is temperature is not very severe; that means, it will have that is the
reasons I am telling that it is not very typically not very severe. That means, temperatures
moistures everything is not very extreme. So, a normal room temperature you can do it.
So, that is the reason very delicate compound very delicate I mean complex of the gases
all those things can be nicely separated in this particular system. Chromatography is one
of the most important analytical techniques especially in the molecular analytical
chemistry. It is coupling with atomic spectroscopy has extended it is application to
element analysis. So, with the advent with the advancement I mean in the spectroscopy
atomic spectroscopy we can find we can identify the element analysis also. It can be used
in the element analysis.
The information obtained by the gas chromatography is not only useful for the
researchers in organic chemistry or biochemist who wants to know what the material he
or she has synthesized in the laboratory or separated from the living tissue, but also to the
industrial scientists and engineers. That is the reason we are discussing in this course of
industrial instrument. It has tremendous application industrial instrumentations where we
need to separate where we need 2 compositions of the gases. Especially composition of
the gases is important to know I mean it is very important I mean parameters in many
situations of his own or competitor’s products, because suppose I mean an unknown
products is coming. So, we do not know what is the composition? So, with the gas
chromatography we also known not only the percentage, but also know that what is the
different composition of the gas is present?
Basic chromatography equipment’s what is the equipment? Let us see this is the basic
equipments. We have a solvent tank you can see here. You see we have a solvent tank
here we have pump. So, through a pump we are putting it and this is a feed injection it is
coming here and it is through a column it is going. Now, we will see that how this
column, because it will retain there for sometime after that it will dilute it will release1
by 1 the gases. So, that will be detected by the recorder and that will be automatically
recorded in a recorder strip chart recorder or a xy recorder. So, so that type of cases we
can find that. So, from the peak we can know the percentage composition of the gas.

Column what is column? The column is where the actual separation takes place.
Separation takes place separations takes place the usually the initially in the column all
the gas or the mixture will go inside. So, it will retain depending on the properties. So, it
will elute 1 after another the gas one after another. It is usually a glass or metal tube of
sufficient strength to withstand the pressures that may be applied across it. It contains the
stationary phase and the mobile phase, which runs through the column and is adsorbed or
absorbed on the stationary phase. Adsorptions or the absorptions, so it contains the
stationary phase and the mobile phase, which runs through the column and is adsorbed or
absorbed depending on that stationery phase.
There are many mainly two types of columns. We have the packed bed column and open
tubular column. So, let us discuss 1 by 1. It is comprised of a stationary phase, which is
in granular form and packed bed into the column as a homogenous bed. The column is
completely filled by the stationery phase. Open tubular column here the stationary phase
is a thin film or layer on the column wall. And there is a passageway through the centre
of the column. Through the centre of the column there is a passage through which the
feed will pass through.
The mobile and stationary phase this is the 2 things we are repeatedly telling. The mobile
phase is comprised of a solvent into which the sample is injected. So, mobile phase is a
sample is a comprised of a solvent into which the sample is injected. We have seen in the
in the figure in the through a pump a solvent is coming and the sample is injected
through which is called the feed which is injected through this 1. So, this is the mobile
phase the sample and the solvent flow through the column together. So, with these both
the sample and the solvent flow through the column together.
Hence the mobile phase is often called a carrier fluid. So, sometimes we are calling it
carrier fluid. The stationary phases are the material in the column for which the
components of the sample to be separated have varying affinities. So, it should have a
varying affinities otherwise you cannot I mean separate that is the most important thing
in the chromatography. So, that is the reason third ((Refer Time: 11:56)) I am explaining
the stationary phase is a material in the column for which the components of the sample
to be separated having varying affinities.
Depending on the materials, which comprise the mobile phase 2 general types of
chromatographic process exist. Depending on the material, which comprised the mobile
phase 2 general types of types of chromatography process exists and they are gas
chromatography it depending on the mobile phase. Here the mobile phase is a gas
generally the gas is inert in nature and the stationary phase is adsorbent solid or liquid
distributed over a surface of a porous inert surface inert support. So, this porosity is very
small in order of Armstrong please note. This is important this porosity size of the
porosity or the also. We will I make the affinity of the different affinity to the different
gases or components of the different gases which mixture of the gases.
And the liquid chromatography here the mobile phase is generally a low viscosity liquid
which flows through the stationary phase bed. And this bed may be an immiscible liquid
coated onto the porous support a thin film of liquid phase bonded to the surface of
adsorbent solid or a solid of a controlled pore size. Note chromatography types are
classified according to the mobile phase materials. Since for different material different
operation techniques and equipment are used.

Operation let us look at steps in the chromatography. So, you see here we are I mean just
change the directions. So, that from the feed injections we have solvent here we are feed
injections it can be other way also. In this case it may come through gravity, so other
case we have to use to the pump. So, basic operations in this after this it will go to a
detector after this it will go to a sorry it after this it will go to a recorder, so this is a
detector same thing solvent oven. So, it is if oven is necessary suppose in many some
situations we have to have to use a control temperature, so in that type of situation we
need an oven. So, that is the reason we put in the oven, but it is most of the cases we do
not temperature is not very I mean crucial in that cases we can use the normal room
temperature.
Feed injection the feed mixture of various components is injected into the mobile phase.
Obviously, it is a mixture individual component, if it is a single gas, I mean single
component; obviously; we are not using going to use a chromatography or gas
chromatography. Number 2; the mobile phase flows through the system under gravity or
capillary action or action of the pump. So, there are 3 possibilities either it will through
gravity it will move or through capillary actions it will move or the action of the pump it
will move.
Separation in the column the different components of mixture feed have different rates of
migration as the mixture is carried by the mobile phase over the stationary phase. These
differential rates provide the separation between the various components. Repeated
sorption and desorption takes place I mean I do not know whether it desorption is in
dictionary. But it is opposite to desorption’s takes place during the movement of the
mixture over the adsorptive material determine the rates.

Elusion from the column you see. After the sample is displaced from the stationary phase
the different components in the sample will elute from the column at different times.
This is the important thing in gas chromatography. Sample will elute from the column at
different times if it is not in different times we cannot separate. Since, it is I mean
coming at different times we can separate we can know the percentage we can know the
individual concentrations of the gas component.
So, this is most important terms in on gas chromatography elute. The smaller the affinity
attraction a molecule has for the stationary phase the shorter the time spent in the
column. So; obviously, we will find that there will be lesser affinity I mean affinity will
be different I mean if you measure the affinity. So, it will be affinity 1 has a larger
affinity than other. So, this will make our separation possible otherwise the separation
was impossible.
Thus, the least retarded component elutes first. The least retarded component elutes first
and most strongly retained component elutes last. It will come at the end and the most
least retarded component will come at the first. Separation is achieved when one
component is retarded sufficiently to prevent overlap with the zone of an adjacent solute
as sample components elute from the column. This is most important thing if there is a
overlapping there is a problem. So, separation is achieved when the 1 component is
retarded sufficiently.
So, this is the word you see sufficiently to prevent overlap with the zone of an adjacent
solute. As a sample component elute from the column we will see later on it will come
like this 1 peak like this 1. However, if the curves looks like this it is very difficult to it is
very difficult to I mean separate. So, in that case, so that is the reason we have to
sufficiently to prevent overlap. So, there is a sufficient gap or time with the zone of an
adjacent solute as sample component elutes from the column.
Detection the different components emerging from the column are collected. And then
the concentrations and the characteristics this is the most important. The concentration
and the characteristics of each component are confirmed by using a same some analytical
procedures like infrared IR spectroscopy or nuclear magnetic resonance NMR
techniques. So, we can identify the element we can find the concentrations. So, the
concentrations and the characteristics of the gas which is coming out which elutes;
obviously, with the time difference when it is coming out I can measure it I can detect
the elements also.
Now, for an example a we considers a mixture of component xyz 3 gases are there. 3
components are x y and z. Initially as the mixture is just injected into the column there is
no separation between the 3 components. Let us see otherwise, it is very difficult you see
here.
So, in the mixture you see initially 3 gases are here. This is the time axis we are plotting.
This is the time x axis is the time y axis is the separation between the 3 gases; that
means, elutes at it elutes. You see as the time goes by x y z the 2 I mean x has a least z
has a least affinity. So, it is coming out first. You see as the time goes by, so the
separation between the I mean separation means the time required time needed to come
out of the I mean column is more and more. So, 3 components now, here the 3
components are mixtures x y z 3 are mixture here you see as the time goes by, so if the
separation; that means, elution time is becoming more and more.
So, it is coming out of the gas 1 by 1. So, z will be detected then y will be detected then x
will be detected. So, I can go back. Consider a mixture consisting of a 3 component x y
and z initially as the mixture is just injected when it is just injected in the column. There
is no separation between the 3 components that we have seen there is no separation
between the 3 components as the mobile phase drives the max the mixture down. The
column mixture down the column a small separation becomes evident when the mobile
phase is moving through the column. A small difference a small separation becomes
evident between the 3 components which is shown in figure 3 b.
Let us assume that the speeds with which the 3 components moving is sz more than sy
more than sx where sz is the speed of the component z, where sz is the speed of the
component z; sy is the speed of the component y and sx is the speed of the component x
In the other words x has the maximum affinity. So, the affinity is different for the
different components where z has the lease affinity this is partially shown in figure 3 c.
Thus, z will emerge from the column first followed by y and followed by x.
This is given in 3 this the things which we have there is a mixture there is no separation.
When the mobile phase is moving through this 1 slowly there is a separation. So, at the
end there is a large separations between separations; that means, of the of the elusion
time; that means, elusion times becoming more and more in the in the case of the x. It is
retention time is most and in the case of y the retention time is medium and in the case z
the retention time is minimum. So, there is a diagrammatic representation of a ((Refer
Time: 21:03)) of example.
Chromatic behavior of a solutes let us look at partition coefficient; there is some

terminologies we are going to introduce. Partition coefficients or equilibrium coefficient
is defined as the molar concentration of analyte in the stationary phase divided by the
molar concentration of the analyte in the mobile phase. Analyte means this is which we
are going to make the analysis, but k is given by concentration in stationary phase upon
concentration in mobile phase.
You see here is the chromatogram we are talking about the plot you see there here there
are detected signal. So, x axis we are plotting time. You see the retention time this is the
time, because after injection. So, see when it elutes the tM is coming first then tR will
come in with the peak. So, these are the 2 peaks. So, this is called this type of plot is
called the chromatogram in the case of gas chromatography.
Retention time you see here we have written tR and tM. Retention time the time between
the same sample injection and analyte peak reaching a detector at the end of the column
is termed as retention time. So, is the time between the sample injections and an analyte
peak reaching a detector at the nd of the column? This term is retention; that means, here
it is reached the detector. So, here from here it reached the detector TM is the time taken
for the mobile phase to pass through the column is called tM time taken by the, for the
mobile phase to pass the column is called tM. So, this individual gas we have shown 1
component of the gas we have shown that it is retention time tR and tM is the time.
Obviously, you can see that the tR is; obviously, that tR should be more than tM in all
the cases. So, the time taken for the mobile phase to pass through the column is called
tM. So, we will get similar graphs similar type of tR. That means, we will get similar tR
this is for 1 gas we can get you will get these for the several other gas or the mixture of
the components. Each analyte is a in a sample we will have a different retention time.
That’s I am saying, so with this we are showing for a single element. So, there will be if
you have a 2 such I mean 2 such gas I mean gas which is making mixture. So, we will
have 2 retention time tR 1 tR 2, if you have 3 so on.
Retention factor K dash it describes the migration rate of an analyte on a column. It is

also called a capacity factor. K dash equal to tR minus tM by tR. Obviously is a positive,
because tM is always less than tR. When an analytes retention factor is less than 1
elution is very fast that is very undesirable I mean elution should normally be very fast.
If the elution is fast, so it is very difficult to I mean make the analysis. Typically ideally
or I should say typically retention factor for an analyte is between 1 and 5 for to be 1 and
5.
Selectivity factor alpha it describes the separations of the components say A and B on the
column. It describes the separation of the 2 components say A and B on the column.
Alpha equal to K dash B upon K dash A. 2 components we have K dash B K dash B and
K dash A. Alpha is greater than 1 always therefore, in the above equation we have
assumed A elutes faster than B. So, A is coming out and faster than B.

Column efficiency band broadening and the resolution these are the important
parameters. As solute zones pass through the column, they broaden and the concentration
at peak maximum gradually falls. This will more important. So, whenever we more clear
when you show the graph. This broadening is important as it ultimately affects the
resolution; that means, how closely how close the characteristics I mean in the case
suppose the measurements. Suppose I have a voltmeter, so resolution means what?
What is the minimum voltage of separations it can detect? Whether it can detect 2.77 and
2.776 volt or it will be 2.7 or 2.6 volt. Obviously the voltmeter which can have
resolutions of which can measure 0.001 volt so; obviously, that will be preferred.
Similarly, in this case in the case of resolution 2 is broadening if it is broadened. So, it is;
obviously, we say that is resolution is poor. So, broadening is important as ultimately
affects the resolution. If the broadening is higher the resolutions will be poor. The solute
zones should be kept as narrow as possible by proper design. So, we should have a very
narrow zone, so that the peak will be very sharp.
Plate height this is another I mean parameters will include here. Plate height is the
quantity that measures the column efficiency and is related to the plate width. We will
show what is the plate height and plate width? It is a prime measure of peak dispersion in
chromatography. It is a prime measure or peak dispersions in chromatography, which is
given by H equal to L by 16 omega B upon tR dash whole square.
What is this ((Refer Time: 26:38)) let us explain where L is the column length; omega B
is the peak width; and t dash R is the retention time corrected for the transit time of an
unabsorbed or non retained solute.
The theoretical plate model of the chromatography we are assuming I mean in a

theoretical model, we are assuming there is a separate plate I mean remains there, but
actually it is not it is a mixture. So, the plate model supposes that the chromatographic
column contains large number of separate layers called the theoretical plates.
Now, separate equilibrium of the sample between the stationary and the mobile phase
occurs in these plates. The separate equilibrium of the sample between the stationary and
the mobile phase occurs in these plates. We are assuming there is a separate plate. The
analyte moves down the column by transfer of mobile phase from 1 plate to the next. 1
by 1 we have assuming that there are a several plates. So, 1 plate to another plates I mean
analyte is moving with a in a presence of the mobile phase.

It is important to know that the plates do not exist plates do not exist; they just serve as a
way of measuring column efficiency, because the column efficiency is important. N
equal to 5.55 tR square upon omega square omega half to the power square. What is
this? Where omega half is the peak width at half height. It is the peak width that the half
height and N equal to L by H.
Now, one more equations we will introduce is a Van Deemter equation for plate height A
more realistic descriptions of the process at work inside a column takes account of the
time taken for the solute to reach the equilibrium between the stationary and the mobile
phases. The resulting band shape of the chromatographic peak is affected by the rate of
elution and the different paths available to solute molecules as they travel between the
particles of stationary phases.
Now, considering the various mechanism which contribute to the band broadening we
arrive at the Van Deemter equation for the plate height which is given by A plus B upon
u plus cu. This is all the I mean dimensions of length you see. We will explain this a b c
in the next where u is the average velocity of the mobile phase. A b and c factors, which
contribute to the band broadening what are these a b c? Let us explain. So, these are the
factor, because band broadening will control the resolutions. So, how closely the 2
elements are there. So, how close the elements are in characteristics which we can
separate. So, this we have discussed here in this particular. So, a b c factors a b a b and c
the factors which contribute to the band broadening.
Now, A is actually what are the factors let us three factors we talk about A is the Eddy
diffusion it is in meters. It is diffusion in meters. The mobile phase moves through the
column, which is packed with stationary phase. Solute molecules will take different
paths though the stationary phase goes at random. The causes this causes broadening of
the solute band since different paths of different lengths. So, it is actually making
different paths we cannot restrict a single path, because is a there is a definite width or
definite diameter of the column.
So, this will make a different path. So, it will reach at a different. So, that will make the
different broadening of the band. So, the length which we are talking about length in
chromatography is not is actually the height if it comes straight away from the bottom to
top. So, we can say that it is a x ((Refer Time: 31:03)), but if it comes suppose this way
or something like this so; obviously, the length will be define. So, if there are in numeral
I mean number of I mean path through which the liquid can moved through the column.
Longitudinal diffusions this again the problem we are always I mean talking about the
lateral diffusions with longitudinal diffusion which is meter square per second The
concentration of the analyte is less at the edges of the band than at the center at the
center. So, it is less so; obviously, if it is not homogenous mixture, because it is a
mixture after all it not a compound that is the entire this column that through which in
which we are actually putting the all the materials which actually I mean absorbed or
adsorbed and elutes the gases or liquids.
So; obviously, it is very difficult to make the concentration we are assuming that the
concentration of the analyte is less at the edges and band and at the center hence analyte
diffuses out from the center to the edges causing band broadening. So, it is going out to
the, so there is band broadening is there. If the velocity of the mobile phase is high then
the analyte spends less time on the column which decreases the effects of the
longitudinal diffusions is high then analyte spends less time on the column which
decreases the effects of the longitudinal distribution diffusion.
Now, resistance to mass transfer which is I mean component C. The analyte takes a
certain amount of time to reach equilibrium between the stationary and the mobile
phases. It is takes certain amount of time to reach equilibrium between the stationary and
mobile phase. If the velocity of the mobile phase is high the analyte has a strong
attraction for the stationary phase, then the analyte in the mobile phase will move ahead
of that in the stationary phase. The band of analyte is broadened now again.

So, you see here, so a typical Van Deemter plot. So, minimum plate height is like this
one. So, we are seeing assuming the plate height is like this 1 optimum velocity will give
you this 1 right which will make them minimum plate height.
Resolution is another important factors another measure to know how well the species
have been separated is provided by the measurement of resolution. The resolution of the
2 species A and B is given by R equal to tR B retention time of the component B minus
retention type of the component A upon point five plus omega A plus omega B.

Scale up of chromatographic columns, because depending on the different dimensions of
the volume of the liquid you have to scale up. Now, there is some typical formula how
we can scale up the, because I have the existing column laboratory column I mean
laboratory model if I want to use it industrially with a large analytical I mean a quantity
and volume flow of the gases. So; obviously, I have to increase the dimensions, so I have
to think of dimension; however, I choose this diameter of the diameter and the length of
the column that is to be discussed now.
During many instances it becomes imperative to increase the size of the chromatography
column in order to increase the production capacity of a process doing, so will require a
scale-up of the column to suitable size for the job. The scale-up of chromatography
column is obtained by maintaining the same column length and increasing the cross-
sectional area. So, we are changing the cross sectional area increasing the cross sectional
area, so the length remains same. So, I am changing the cross section area by a formula
let us look at.
A simple way to calculate the scaled diameter is given by the following equations. We
have a load 2 upon load 1 equal to radius 2 upon radius 1 to the power square. What is
load 1 load 2, let us look at where load 1 is the present volume of the column it is the
present volume of the column. And load 2 is the desired volume of the column and if it is
the present volume of the column and desired volume of the column radius1 is a ((Refer
Time: 35:41)) present radius of the column. Obviously 3 is in the equation 4 are
unknown 4 are four parameters 3 are known so; obviously, radius 2 can be calculated is
the calculated radius of the column. So, we can calculate an accordingly I can change
very simple formula though I mean with some limitations obviously.
To recreate identical operating conditions on the scaled column as the unscaled 1 the
linear flow rate must be kept constant. The flow rate must kept constant we should have
flow meters to know that because the feed mobile phase is moving through this 1. The
volumetric flow rate required for keeping the linear flow rate constant is given by
volumetric flow rate 2 upon volumetric flow rate 1 equal to load 2 by load 1. So;
obviously, we know that the here in this case we have volumetric load 2 from these we
can calculate the parameter.
What is the where volumetric flow rate; one is the present volumetric flow rate that is for
the laboratory models and volumetric flow rate calculate a volumetric flow rate that is
we can calculate.
Now, quantitative analysis this is also very important in the case of gas chromatography.
Let us look at that peak area integration what is that peak area integration. let us look
at… In column chromatography the analog signal generated by the detector is
graphically recorded in the form of chromatographic peaks which is called the
chromatogram that we have seen it looks like this. So, it is coming like this one, you
have seen it is coming like peak like this 1, peak like this 1, so this is called the
chromatograph.
The area under these peaks can then be integrated in a variety of ways. So, we are
integrating that areas and the resulting data related to the composition of the samples can
be studied this is most important thing. Now, we come to the detector side once you have
I mean with some detectors we can collect the analog signals electrical signals now, it
will be plotted. So, the area under these peaks can then be integrated in a variety of ways
and the resulting data related to the composition of the samples can be studied. Because
the area will give you the relative contents of the different compositions the ways in
which the area under the peak can be calculated are as follows. There are various
methods by which areas can be calculated we will discuss 1 by 1.
Height times width at the half height. Height multiply by this basically height multiplied
by the width at the half height what is that? It involves the multiplying the actual peak
height with the width of the width at the half height It looks like this; that means, I will
show you actual peak height suppose we have a peak height is this. This is multiplied by
the half height width of the half height we multiply by this suppose this is x this is y. So,
x in to y is the height time’s width. Height times width at the half height will be will be
measured of the integration.

A new baseline is drawn for measuring peak height, and the width at the half height since
the normal zero signal baseline causes large deviation due to tailing.
Planimetry; this is another method of integration. In this method the peak is traced with a
planimeter a mechanical device which measures area by tracing the perimeter of the
peak. In this method the peak is traced with a planimeter a mechanical device, which
measures the area by tracing the perimeter of the peak. Planimetry is less precise than the
height width integrations due to errors caused from the placing the baseline tracing the
peak outline and obtaining a reading.
Triangulation; in this method tangents are drawn to the sides of the peak at the inflection
points and area of the triangle formed by these tangents and the baseline is determined.
So, it is looks like this I have a peak I have a peak here I will show here. So, I am
drawing a tangent here tangent here. So, we will give you the area of the triangle will
form this and the baseline is determined right. So, tangents we have drawn one at the
inflection point inflection point like this one inflection points rather. So, we are trying to
tangents where it meets. So, we are making one triangle. So, that area of the triangle
formed by this tangents baseline is determined.

The height is measured from the baseline to point where the tangents intersect right.
Ball and disk integrators; this is very mechanical integrator is useful the ball and disk
integrator shown in figure 6, in the next figure in an automatic mechanical type of
integrator. A ball positioned on a rotating flat disk will rotate at a speed proportional to
the distance from the centre of rotation.
The ball is positioned on the disk at a distance from the centre in the same relationship as
the position of the recorder pen to the baseline of the chromatogram. If the disk is rotated
at the constant speed then the ball will rotate at a speed proportional to the position of the
recorder pen from zero right.
The speed is then transmitted to a roller through a second ball, which by means of a
spiral in and spiral out cam actuates the integrator pen at a speed directly proportional the
position of the recorder pen clear? We use more clear once we show the diagram
The drive between the disk and the ball is by traction through an oil film, which is
similar to an induction motor where slip is proportional to the driven load. Reading the
integrator trace shown in figure seven is done as described in the following points. First
step is to establish the desired chart time interval from the recorder pen trace of the
chromatogram, and the project directly down to the integrator trace.
The value of an interval is obtained by counting the chart graduations crossed by the
integrator trace. So, it will be a like this one we will see it will be like this one. So, it will
be graduation crossed by integrator trace. So, interval is obtained by counting the chart
like this one.

A full stroke of the saw-tooth pattern in the either direction represents hundred counts
and every horizontal division has a value of 10. So, full stroke of the saw tooth pattern in
either direction represents the 100 counts. So, may be the horizontal division has a value
of 10. Values less than 10 are estimated; that means, only 10 is minimum resolutions we
can get. So if in between comes anything that needs to be estimated in the example the
interval for the main peak is 1084 counts.
You see this is the disk and we are this pen is going to a recorder that I am not showing.
Spring ((Refer Time: 43:46)) a valve and the disk this input linkage is here. This is a
pulsar, this is a roller and this is the time motors. So, it is going to the pen in the next
graph we will show.
You see here this pen goes here which is recording on a chart right. So, ball and disk
integrated symmetric. So, this is about the about our, the gas chromatography basic
recorder. Let us look at the plot now.
You see there are total 1084 counts 10218 1083. So, the estimation of the peak areas
with ball and disk integrators, so we have making ten divisions here. So, each has a 10
counts here. So, we will get 1084 counts here. So, with this we can estimate the peak
areas of the gas chromatography peak areas of the systems. Now, gas chromatography as
you know it is extensively used not only for analytical. Because as the beginning we say
that the initially it is not only for the making the analysis of the gas mixtures which is
coming out in this industrial process.
But also to know if suppose a some other companies making some particular mixture of
gases what is the components of the gases. It is very vague though it is very difficult to
tell. Now, once you separated the gas components; obviously, you can make the analysis
you can say will the other method to know what is the component? But the basic
separation is the most important thing where you have to wait one by one it will come.
So, basically I can say that the this chromatography is something like that if I pass the
gas mixtures through a column whether it is open channel open tubular form or a flat bed
reactor or flat bed column.
So, it will absorbs or adsorbs for sometime it will retained the gast for sometime then it
will slowly all the gas will not come out immediately or not it will come after certain
amount of time. It will come one by one from the packed bed column. So, these actually
makes our gas chromatography flexible right. Now, in these case one thing we should
know that this is the just a tremendous not only in the chemistries. It has a tremendous
use in the industrial operations in many situations you know we have to trace the
presence of gases of the gas mixtures and complete the process.
So, that type of situation we can collect a sample gas and inject through this one through
the feed, because we have seen that we have a stationary phase and the mobile phase. So,
that, so this will go through the feed as a feed through the systems or the carrier through
the systems. So, slowly it will separate and it will tell the composition. Now, different I
mean compositions I mean determination of the percentage and all this that is that is
coming next that is not I mean we are coming lateral. But the basic difference I mean
basic finding that is presence of particular gas or composition detecting the particular gas
or the components of the gas is most important.
Later on you can use the spectroscopy and all those things to find what are the elements
with the gases are there mixture of the gases the first is separation. So, this is only
method of the mankind. So, far developed this is a very excellent method as. So, far you
know to find the making a separations. Before that as I told you this is very unreliable
method this is a quite reliable method, and is not that I mean sensitive to temperatures.
And all though in some cases, we have to maintain a particular temperature, while you
are calibrating, but most of the cases the, I mean temperature I mean maintaining the
temperature is not very crucial. That is the reason we were saying it is not necessary it is
not very harsh. It is not very severe environment while we are... So, many cases suppose
there is some gasses, suppose it is I mean getting decomposed or in some other particular
temperature that type of cases also we can we can use the separation.
In other method, you can have high temperatures and all this lower low temperature
those things are not necessary in the case of gas chromatography. So, with this I come to
the end of the gas chromatography 1 and the lesson 36 we will continue with this
particular. That mean the after chromatography which is a chromatography 2. Preview of
next lecture; welcome to the lesson 36 of industrial instrumentation. This lesson is a
continuation of the lesson 36 where we have discussed the chromatography 1. In this
lesson, we will discuss chromatography 2. So, let us look at the contents. It is a
chromatography 2 as I told you.
The contents are gas chromatography, liquid chromatography, chromatogram. We will

see that in this particular lessons. So, we discuss the actual the how that chromatogram
should like and what are the, if the multiple picks comes and what are the problems what
exactly I mean the shape of the chromatogram should be. I mean that depends actual the
finding the concentrations of the different components of the gas and liquids precisely.
At the end of the lesson or end of the chapter the viewer will know different
equipment’s, and different operation techniques used in gas and liquid chromatography.
Different detectors used in the chromatography and their purpose and their classification.
We will find there are different types of detectors available. There are thermal
conductivity detector there are flame ionization detector there are electron caption detect
capture detector.
So, there are different I mean the principles are I mean somewhat different in though the
thermal conductivity detectors are the oldest one, and most widely used detectors in the
gas chromatography. But we will discuss the other detectors also where you will find that
still making the derivatives of the particular element of the mixtures. So, that to find the
concentration accurately, how do operating conditions like pH value temperature affect
the chromatogram? So, we will find these things also the pH value though I mean its I
mean not very importance in the case of gas chromatography. This is very important in
the case of liquid chromatography.
Gas chromatography; already that we have discussed that gas chromatography makes the
use of a pressurized gas cylinder and a carrier gas such as helium which is inert nature to
carry the solute through the column. Because we will find we have discussed these things
that there is a column and through column these the gas would flow right and can and
there is a packing material inside the column. So which will adsorb or absorb and elute it
after sometime one by one. Gas adsorptions that is gas solid chromatography involves a
packed bed comprised of an adsorbent used as the stationary phase. Common adsorbents
are zeolites silica gel and activated alumina these are the most common sort of I mean
adsorbents are used in a chromatography. These method is mainly used for separating
mixture of gases we will find this is basically used for mixture of gases, right.
Gas liquid chromatography is a more common type of analytical gas chromatography. In

this type of column an inert porous solid is coated with a viscous liquid which acts as a
stationary phase. And diatomaceous earth is the most common solid used and solutes in
the feed stream dissolve into the liquid phase and eventually vaporize.
Current flow across the external resistor is sensed with a voltage drop and it is amplified
and displayed on a recorder. Now, hydrocarbon groups are introduced into a flame and a
complex process takes place in which positively charged carbons spaces and electrons
are formed. Now, the current is greatly increased. Now, this frame ionization detector
response only to the substance that produce charged ions when bond in hydrogen air
flame that is most important thing. In an organic compound the response is proportional
to the number of oxidizable carbon atoms. So, this is basic principles of the flame
ionization detector.
Now, if I look at the thermionic emission detector you see this is thermionic emission
detector it is principle is something different. It employed a fuel poor hydrogen plasma.
These low temperature source suppresses the normal flame ionization response of a
compounds not containing hydrogen and phosphorous although the response to carbon is
not entirely eliminated. A non-volatile rubidium silicate bead here you see here non-
volatile rubidium silicate bead centered about 1.2 centimeter above the plasma jet is
electrically heated by a variable current supply to between 600 to 800 degree centigrade.
These arrangement permits the fine adjustments of the beads temperature and
independent of the plasma as a source of thermal energy. With a very small hydrogen
flow the detector responds to both nitrogen and phosphorous compounds enlarging the
plasma. The changing polarity between the plasma and the detector responds only to the
phosphorous compounds. So, that I can make it nitrogen I can make it sensitive to the
phosphorous compounds also. Now, compared to the compared with the flame ionization
detector the thermal emission detector, thermionic emission detector is our 50 times
more sensitive for nitrogen and about 500 times more sensitive for phosphorous.
So, any traces of phosphorous any traces of nitrogens can be better detectors in the case
of. I mean thermionic emission detector compared to the flame ionization detectors
though flame ionization is also comparatively relatively new compare to the thermal
conductivity detector. The minimum detectable limit is around point zero six pg I mean
per second for nitrogen. So, this is the thing you can see here that there is a flame tip
which is one point around one point per 2 second above the plasma jet. So, this is our, I
mean thermionic emission detector. Last, we will discuss the electron capture detector.
You see this is a electron capture detector. Now, electron capture detector has two
electrodes within the, with the column affluent passing between one of the electrodes is
treated with radio isotopes. That emits high energy electrons as it decays and this emitted
electrons produce. I mean amounts of large amount of low energy thermal secondary
electrons in the gas chromatography carrier gas all of which are collected by other
positively polarized electrode either of the positive polarized electrode. Molecules that
have an affinity for thermal electrons captured electrons as they pass between the
electrodes and deduce the steady state current thus providing a electrical reproductions.
Or the gas chromatography peak of the two general designs; the plane parallel and the
concentric and the later design is preferred since it is easier to construct a small volume
cell in this form.
Now, particular radioactive sources which are used that the titanium adsorbed in the
titanium or scandium and nickel 63 as a foil or platter on the interior of the cathode
chamber. So, your chromatograph will change; obviously if the chromatogram change
efficiency selectivity everything will change. So, this is to be, so whatever the desired
value of the chromatogram that is to be maintained right.
Four fold information is obtained from the chromatogram. What are the different I mean
components I am getting from the chromatogram, what are the different I mean
informations I am getting from this chromatogram? So, you know already we have
discussed let us downright. What one by one, the level of complexity of the sample is
indicated by the number of peaks, how many what is the complex mixtures I mean that
we can detect from this one? So, level of complexity of the sample is indicated by the
number of peaks, which appear.
Qualitative information about the sample of composition is obtained by comparing peak

positions with those of the standards. Qualitative information about the sample
composition is obtained by comparing the peak positions with those of standards.
Quantitative information regarding the relative concentrations this is most important we
are measuring the relative, because of the sample. If the we are absolutes measuring we
are making the relative comparison relative concentrations actually we are measuring the
gas chromatography. Because otherwise we cannot measure it, because we need the
injections at the, it is only ten micro liter that is the best other samples betrays the
detection. Quantitative information regarding the relative concentration can be obtained
from the peak and area comparisons. Column performance can be checked by comparing
with the standard outputs.
This column performance also, because if I have if I knew if I knew the particular
suppose if I want to calibrate the column performance I know the particular gas with
which actually we are injecting. So, the, that time when an unknown gas is coming, if I
want to calibrate the how the column actually working? So, the column efficiency can
also be known I mean I mean can be known by looking at the output. So, these are the 4r
I mean features I mean which will get from the gas chromatography. The most important
thing is the relative I mean first of all the, what are the different components of the gas
present in the mixture? Second thing is the relative this is the most prime most
importance. Second thing is the relative concentrations of the gas or liquid or whatever it
may be in the mixture. So, with this I come to the end of the lesson thirty-six on
Prof. A. Barua
Lecture - 36
Chromatography – II
Welcome to the lesson 36 of industrial instrumentation. This lesson is a continuation

of the lesson 35 where we have discussed the chromatography 1. In this lesson, we will
discuss chromatography 2. So, let us look at the contents, this is the chromatography 2 as
I told you.
The contents are gas chromatography liquid chromatography chromatogram. So, we will
see that in these particular lessons, we will discuss the actual the how the chromatogram
should look like. And what are the, if the multiple peaks comes and what are the
problems, what exactly I mean the shape of the chromatogram should be? I mean that
depends actual the finding the concentrations of the different components of the gas and
liquid precisely.
At the end of the lesson or end of the chapter the viewer will know different equipment’s
and different operation techniques used in gas and liquid chromatography. Different
detectors used in the chromatography and their purpose and their classification. We will
find there are different types of detectors available. There are thermal conductivity
detector; there are film ionization detector; there are electron capture detector. So, there
are different I mean the principles are I mean somewhat different.
And though the thermal conductivity detectors are the oldest one and most widely used
detectors in the gas chromatography. But we will discuss the other detectors also where
you will find that is we will be making a derivatives of the particular elements of the
mixtures. So, that to find the concentration accurately, how do operating conditions like
pH value temperature affect the chromatogram. So, we will find these things also the pH
value though I mean it is I mean not very importance in the case of gas chromatography.
This is very important in the case of liquid chromatography.
Gas chromatography; already that we have discussed that gas chromatography makes use
of a pressurized gas cylinder and a carrier gas such as helium which is inert in nature to
carry the solute through the column. Because we will find we have discussed these things
that there is a column and through column these the gas would flow right and can there is
a packing material inside the column, so which will adsorb or absorb and elute it after
sometime one by one. Gas adsorptions that is gas solid chromatography involves a
packed bed comprised of an adsorbent used as the stationary phase. Common adsorbents
are Zeolite silica gel and activated alumina these are the most common sort of I mean
adsorbents are used in a chromatography. This method is mainly used for separating
mixture of gases. We will find this is basically used for the mixture of gases right.
Gas liquid chromatography is a more common type of analytical gas chromatography. In

this type of column an inert porous solid is coated with a viscous liquid which acts as a
stationary phase. And diatomaceous earth is the most common solid used and solutes in
the feed stream dissolved into the liquid phase and eventually vaporize and the
separation is thus based on the relative volatilities. So, these you can see that these are all
different these already we have discussed in the lesson 35. Now, we are discussing in
more details you see the last line which is most important the separation is thus based on
the relative volatilities.

Now, capillary gas chromatography uses a glass or fused silica capillary walls which are
coated with an absorbent or other solvent. And the column has only a limited capacity
because of the small amount of stationary phase. However, this method also yields rapid
separation of the mixtures right.
Now, chromatography equipment’s, what we need and the entire chromatography

equipment? Let us look at that. Basically we will see that an figure in the next diagram.
In the next slide basically a gas chromatograph consists of six parts a supply of carrier
gas in a high pressure cylinder, a sample injection system. Number 3 is a separation
column where we have the packing materials and all those things. The detector which is
most important in this particular lessons we will discuss a detectors in very much details.
There are various kinds of detectors we will discuss the detectors and we are already we
have discussed about the the packing materials in the case of. I mean in the case of gas
chromatography we will discuss also some to some extent that thing in this particular
lesson. An electrometer; that means, and a separate thermostated compartments for
housing the column and detector. So, as to regulate their temperatures right, so detector
usually followed by a strip chart recorders or any other electronic recorders where we
will I mean record the peak. In the strip chart recorders we will get a hard copies. It can
be computerized also where you will get the actual plot on the screen itself.
Now, see this is the symmetric of a gas chromatography you see can we can see here this
is the sample injections systems. We have a column here you see the columns are made
like this. Columns are these are the columns we can see here these are the columns. Then
it is going out and this entire thing placed in a thermal chamber. You can see this is the
thermostatic chambers we has used and there is a bridge what is the bridge. And there is
a reference detector we will find that to make the I mean other things nullified, because
you see these types of measurements already we have did in the case of Wheatstone
bridge. We will find the same principles is utilized especially in the case of I mean
thermal conductivity detector this is used and we have a measuring detectors and we
have a recorder.
This is the reference detectors it has no column nothing is there only detector is there and
there is a measuring detector. Because measuring detector I mean suppose the detector if
the temperature I mean if I want to make the temperature independent or ambient
temperature radiation independent. So, in that case I should use a reference detectors.
Reference detector has no column nothing but it has a same I mean same material I mean
what are the insight the components. So, that, so that temperature variation everything
will be nullified and I will get the only the only the, this will go to the two offset offs of
the Wheatstone bridge. So, that it can be nullified we will see these thing in the next I
mean in the subsequent slides.
Now, sample injection system; let us look at the sample must be introduced as a vapour
in the smallest possible volume sample should be very very small. So, the smallest the, I
mean smallest amount that is better for our detection, the peak will be better that is most
important thing right. So, the sample should be, because ultimately we are measuring the
relative concentrations. So, it does not matter I mean if it is small or large, so if it is
small, so it is good for us. The detection will be easy the peak will be more separable we
will see the, that type of things. The sample must be introduced as a vapour in the
smallest possible volume and in a minimum of time without decomposition. It should not
be decomposed that is most important thing. Liquid samples; it is order of 1 to 10 micro
liters in a volume are usually injected by a micro syringe through the self sealing silicone
rubber septum. So, it is a very small amount that we can see 1 to 10 microliters and it
should be it should have micro syringe. So, that a very small amount can be injected.
The most accurate and precise method for the gas sample used a calibrated sample loop
of 0.5 10 milliliters and a multi port rotary valve. So, different types of rotary valves will
be used in the case of in the case of this type of sample injections because sometime we
need a valve. So, the precise amount can be injected. The smaller the sample the better
the peak shape that I told you the peak shape should be better and better and how the
peak should look like let us look at that.

You see that peak should the ideal peak should be suppose I have a suppose I have 2
components. So, time it is in 1 2 3 4. So, it is coming like this the 2 peak is coming like
this right. It is a good separation, separation of the 2 components the 2 peaks are coming
like this. It might be sharp also it may not necessary.
It will be like this when it can be I mean sharp also it can be like this one. It can be very
sharp like this one also. Now, if the, this is always possible if the sample quantity is very,
very small.

If the sample quantity is large you can see that what type of peak you will get? You will
may get a peak which looks like this. Actually do it actually look like this one. So, it is
very difficult to find the area under the curve from this graph. So, this is another way the
why we I mean give so much emphasis on the small quantity of the sample right. The
smaller the sample the better the peak shape when a very dilute solutions are to be
analyzed the use of a concentration precolumn such as two six diphenyl-ah p- phenylene
oxide porous polymer allows the quantitative transfer of up to twenty micro liter of
sample. So, this is a precise transfer of the particular amount of liquids we are
considering here.
Now, derivative formations derivative play an important role in gas chromatography for
analysis of polar compounds such as fatty acids steroids drugs biological amine and
phenols etcetera. Derivatives make a polar compound less polar improve quantitation
and increase the volatility of high molecular weight compounds.
Chromatographic columns, so let us look at the columns first there are basically 2 types
of column the packed bit column open tubular or capillary column that is all we have
already discuss in the lesson 35. The packed columns are made of stainless steel copper
or glass tubing of diameter 1.6 to 9 millimeter, and typically is 3 meter in length. We
have seen that it is a zigzag fashion, because there is a large column length 3 meter of
length you can see you cannot accommodate. So, it should be zigzag phase.

Capillary columns have an open unrestricted path for the carrier gas within the column.
And it is a long narrow tubing of length 50 to 150 meters extremely long and of
diameters 0.25 millimeter extremely small. And it is inner wall is coated with a liquid
stationary phase to about one micron thickness right. Columns constructed of silane-
treated Pyrex glass have the most desirable features. A good packed bed column will
have a nominally one thousand to three thousand plates per meter already we have seen
that this concept actually came from the distillation columns. So, more number of plates
better is a separation. Capillary column is range from 1000 to 4000 plates per meter this
is the theoretical value right.
The sample capacity of capillary column is determined principally by the thickness of the
stationary phase on the column walls. The coating usually done is of silicon gum. Packed
columns are usually formed into several coils and placed within the oven compartments.
Its several coils we have seen its coil looks like this that we already we have seen. It
looks like this right, its looks like this it can be coiled like this one also, but it is better
and if you this. So, it will get it will come and it will go out like this one. So, the packing
materials are inside this column. So, the packed column are usually formed into several
coils and placed within the oven compartment and oven compartment. It is nothing but a
temperature I mean is a thermostat based or more precisely controlled temperature
environments. Capillary columns are tubing coiled into an open spiral and basket coil or
a flat pancake shape.

Supports; the purpose of the support is to provide an inert surface onto which the
stationary liquid phase can be placed in a pan in a packed column. The diatomaceous
earth supports may be either firebrick derived materials like Chromosorb P and Anakrom
ABS etcetera. This is special materials the acid-washed grade will perform quite well for
the analysis of relatively non polar samples.
Liquid phase the stationary liquid phase provides the separation of the columns. In
addition to selectivity the liquid phase should be chemically or thermally stable, that is
most important thing.
Now, ovens as I told you the oven is used for maintaining the precise temperature control
around the column. Hence the column oven should be free from the influence of
changing the ambient temperature and should have a well designed and adequate air flow
systems. There should be good air flow systems otherwise temperature control will be
difficult.
Detectors now detectors will discuss in very much detail as I told you. The detector
senses the presence of the individual components as they leave the column as they leave
the column. So, we will have, so all the detectors we have injection systems we have
carrier gas we have a solvent pump everything is there. So, it will go to the column from
the column it will come out it will elute depending on the absorbent material absorbent
material and ultimately it is come to the. So, all the columns will be followed by detector
and detector will be followed by another recorder. The detector output after amplification
is traced on a recorder as I told you right. The different types of detectors generally used
are thermal conductivity detectors flame ionization detectors, thermionic emission
detector, flame photometric detector, electron capture detector and helium detector. So,
we will discuss some of the things. So, we will discuss some not all we will discuss in
this case. So, that anybody can refer to some standard book on the gas chromatography
for this.
Detectors in chromatography are generally operated in 2 different ways they respond

either to the concentration of the solute or the mass flow rate. Those responding to the
concentration yield a signal which is proportional to the solute concentration which
traverses the detector. An elution peak will resolve will resolve when the signal is plotted
against time right. So, there can be different components. So, how does it look? Let us
look at.
You see here that the selectivity efficiency this is very important in the case of the peak
right. So, those are the things we should discuss what is the selectivity? What is the
efficiency? Let us look at, I have you see here this is a injections I have made here this is
adequate selectivity, but poor efficiency. You see here good efficiency and selectivity.
This is the desired characteristics and selectivity and you see this another one where you
can see like this one. This is a good efficiency, but poor selectivity. So, selectivity
efficiency I mean everything we have discussed for the 2 columns in this particular slide
right.
So, these are the desired characteristics for any I mean the peak should be look like this
one. So, who will decide this one actually this will be decided in the detector itself
efficiency of the detector selectivity of the detector. So, the two most important thing for
the detector is the selectivity and efficiency and also the partition ratio right. So, the
elution peak that I just drawn elution peak the peak is that what I we have drawn is for
the elution peak right. So, this is the peak which we have drawn talked about. So, this is
the suppose…
So, this peak is called the elution peak, because it will come out of the gas and suddenly
it will detect at the output.
For such detectors the area under the peak is proportional to the mass of a component
and the inversely proportional to the flow rate of the mobile phase. Hence an important
care of keeping the flow of mobile phase constant for such detector is necessary. In
differential detectors that responds to the mass flow rate. The peak area is directly
proportional to the total mass and there is no dependency on flow rate of the mobile
phase, you see let us go back first of all. Now, you see when I talked about the thermal
conductivity detector. Thermal conductivity detector is the one of the oldest detector and
its steel is used because of it is simplicity of the system. It is very simple system and it is
very widely used. Data’s are available over the years peoples are using these things and
it is nondestructive. This is the most important thing for the many of the gas
chromatography right.
So, let us look the, how it looks? I have a block looks like this. We will take a new page
block is there and here actually I will put the detector. So, here it will look like this then
it will go like this. So, this is our block, so this is the gas flow in and this is the gas flow
out and this is the block. It will work as a heat sink please note work as a heat sink,
because there is a some sort of heating here and these actually, if I want to make the it is
independents of the detector. Independents of the other parameters of the detector
because detector output should be should be solely depends on the, I mean different
components of the or the elution peak. So, in that type of cases the, it should not
influence by the any other I mean any other component any other components of the
other detector itself. So, that is the reason we are using 2 detectors. So, let us look at how
the 2 detectors methods are using.
It looks like this we have a one detector here. Then we have a here 100 ohm. This is 100
ohm which balanced though there is an ammeter 0 to 500 milliampere. This is coming
through a resistance 100 ohm it is coming here. So, this is excitation voltage Ex; this is
excitation. So, this is coming to the recorder, because this is the unbalanced voltage. So,
this is the detector 1 and this is the detector 2. So, same carrier gas is passing through
both the detector. So, in both the cavity; that means, cavity means actually we have
showing that you can have a four same sensors also. So, it looks like this you see this is
the thing let me let me first explain then I will go to that, see 2 detectors are there. So,
same carrier gas is passing through the both the detectors, but there is no column or
anything on the detector 2.
Initially what we will do? I will balance with the by varying this potentiometer I will
balance the bridge. So, whenever the detector I mean I mean elution peaks comes; that
means, when the gas is coming out of the elution peaks. So, detector output will change
because the there is a thermal conductivity there. So, it will change, so automatically if it
changes I will get a unbalanced voltage that will be recorded in the recorder right. Now,
again you see the in the hydrogen flame ionization detectors for high sensitivity analysis
of organic compounds. Hydrogen flame ionization detector is used that is different thing
we will discuss that thing also. So, 2 detectors are using you see we can have 4 detectors
also 4 detectors in one block itself, right.
It looks like this, you see... So, what they have? We have 1 4detectors is a block. I can
have 2 detectors and I can have 4 detectors also. It is coming like this one right. Let us go
back now.
Now, you see this is the flame ionization detector as I told you this flame ionization
detector is one of the detectors since the different components in the flame ionization
detectors. This is a exit end of the column let us look at you see here a jet capillary is
there it is the I mean flowing through this one. This is the hydrogen air flame and air is
coming through this one and this is with the air and hydrogen. So, it is I mean. So, we
have a collector assembly nut. We have insulators and we have a collector holder. We
are putting across that high voltage inside the detectors plates. So, what is this? Let us
explain the flame ionization detector is currently is one of the most popular detectors,
because of its high sensitivity wide range and great reliability. As shown in the figures I
mean this particular figure; figure 2 I think figure two the column affluent enters the nut
base through the multiple filters.
So, through this one this is enters and it is mixed with the hydrogen gas and mixture
bonds at the tip. This is the tip of the hydrogen you see this is the tip let me. So, this is
the tip you can see here this is our tip hydrogen tip. The jet air with the air or oxygen,
because through air it is coming. So, it is going down and is bonding. Ions and the free
electrons are formed in the flame and these enter the gap between the 2 electrodes. We
have 2 electrodes and there is a high voltage and the flame jet and the collector which
may be a parallel plate or might be a cylindrical. Now, this is mounted around 0.5 to 1
centimeter 1 centimeter above the flame tip. So, above the, around 0.5 either 2 0.5 to 1
centimeter half centimeter to 1 centimeter above the flame tip and across the 2 electrodes
a high voltage of 400 volt is applied right. And this lowers the resistance across the gap
and a causes a current to flow, right.
Normally an externally bucking voltage is produced to balance the potential generated by

the ions and free electrons generated in the pure hydrogen air flame. This ensures that a
net current flows only when ionized materials enters the gap and the net current flows
when the ionized material enters the gap. Thus enhance enhancing the differential
sensitivity of the detector the current flow across the external register is sensed with a
voltage drop and is amplified and displayed on a recorder. Now, hydrocarbon groups are
introduced into a flame and a complex process takes place in which positively charged
carbons spaces and electrons are formed. Now, the current is greatly increased. Now, this
flame ionization detector responds only to the substance that produce charged ions when
bond in hydrogen air flames that is most important thing. In an organic compound the
response is proportional to the number of oxidizable carbon atoms. So, this is basic
principles of the flame ionization detector.
Now, if I look at the thermionic emission detector you see this is thermionic emission
detector it is principle is something different. It employs a fuel pore hydrogen plasma.
This low temperature source suppresses the normal flame ionization response of a
compounds not containing hydrogen and phosphorous. Although the response to carbon
is not entirely eliminated a nonvolatile rubidium silicate bead here you see here
nonvolatile rubidium silicate bead centered about one point two centimeter above the
plasma jet is electrically heated by variable current supply to between six hundred to
eight hundred degree centigrade.
This arrangement permits the fine adjustments of the beads temperature an independent
of the plasma as a source of thermal energy with a very small hydrogen flow, the
detector response to both nitrogen, phosphorous compounds. And enlarging the plasma
the changing the polarity between the plasma tube and the collector, and detector
response only to the phosphorous compounds. So that I can make it nitrogen I can make
it sensitive to the phosphorous compounds also. Now, compared to the compared with
the flame ionization detector; the thermal emission detector. Thermionic emission
detector is about 50 times more sensitive for nitrogen and about 500 times more sensitive
for phosphorous.
So, any traces of phosphorous any traces of nitrogens can be better detectors in the case
of I mean a thermionic emission detector compared to the flame ionization detector.
Though flame ionization detector also is comparatively new compared to the thermal
conductivity detector. The minimum detectable limit is around 0.06 Pg I mean per
second for nitrogen. So, this is the thing you can see here that there is a flame tip which
is one point around 1.2 centimeter above the plasma jet. So, this is our, I mean
thermionic emission detector. Last we will discuss the electron capture detector.
You see this is a electron capture detector. Now, electron capture detector has two
electrodes within the, with the column affluent passing between one of the electrodes is
treated with radioisotopes. That emits high energy electrons as it decays and this emitted
electrons produced. I mean amounts of large amount of low energy thermal secondary
electrons in the gas chromatography carrier gas. All of which are collected by other
positively polarized electrode either of the positive polarized electrode. Molecules that
have an affinity for thermal electrons captured electrons as they pass between the
electrodes. And deduce the steady current thus providing a electrical reproductions of the
gas chromatography peak of the 2 general designs. The plane parallel and the concentric
cell and the lateral design is preferred since it is easier to construct a small low voltage
volume cell in this form.
Now, particular radioactive sources which are used that the tritium adsorb in the titanium
or scandium and nickel 63 as a foil or plated on the interior of the cathode chamber.
Tritium sources have a high specific activity giving a large standing current and high
sensitivity, but the beta energy is. So, low that the sources extremely susceptible to
contamination. The maximum working temperature is 225 degree centigrade. Now,
nickel 63 is a higher energy sources that can be used up to 400 degree centigrade. So,
this you see this is the, I mean we are talking the electron capture detectors. Let us now
go back. Now, adsorbent in cases of gas solid chromatography
Inorganic molecular sieves are naturally occurring for synthetic or synthetic zeolites
which are comprised of interconnected cavities or pores of uniform size. Four types of
molecular sieves are varying in pore diameters. This is the absorbent because is different
types used are the dimensions of the pores are. So, that the I mean it is a Armstrong
level, so that the some of the larger cannot enter the pore some will enter which are
smaller in diameter. Varying pore diameter and alkali metals contents are available.
Type 3 A; a potassium aluminosilicate with the pore diameters of 3 Armstrong. It will

absorb the molecules such as water and ammonia. It will absorb the molecules of water
and ammonia. Type 4 A; it is a sodium analog of type of 3 A; it will absorb the
molecules of carbon dioxide, right. Sulfur dioxide, hydrogen disulfide, ethane ethylene,
propylene and ethanol, right.
Type five A; with calcium replacing part of the sodium content of type four A and it will
separate most of the hydrocarbons. In many I mean this particular industry this is very
important. Type 13 x; a sodium aluminosilicate of different crystalline structure from the
preceding types has a pore diameter of 10 Armstrong.
The molecular sieves separate molecules not only according to size and configuration,
but according to the polarity and degree of unsaturation as well. Columns prepared from
carbon molecules sieve type B with pore radius of 5 to 15 Armstrong are also available.
Porous polymer packings are analogous to the porous gels used in the exclusion
chromatography. Those made from the copolymers of aromatic hydrocarbons provide
column packings of low to moderate polarity. Polymers made from the acrylic esters and
provide the packing of moderate to high polarity.
Solid adsorbents like silica gel alumina etcetera are used for specific applications right.
So, these are the some specific applications we will used silica gels alumina etcetera
right. It is also these are also basically adsorbent please note. The large retention of silica
gel for carbon dioxide which elutes after an ethane is useful in multicolumn systems
because there is a large retention which is it will retained it for a longer time. So, it will;
obviously, help to get a better peak. So, the large retention of the silica gel for carbon
dioxide which elutes after ethane is useful in the multicolumn systems. Similarly,
alumina is useful for retention of unsaturated hydrocarbons, because unsaturated
hydrocarbons. So, it will; obviously, helpful for retention, because we have to retain
right. So, that is the thing of the packing materials.
Multicolumn systems gas solid chromatography is more complicated than the gas liquid
chromatography. Since all gas mixtures contain components which are not separated on
or will not pass through a particular column reasonable time. Backflushing substantially
it reduces the analysis time.
In series-cross-detector across detector one column is used before the detector and a
different column is used after the detector. The same packing may reside in both the
columns. In two column series bypass technique a column switching valve, valve enables
a column 2 to bypass by the carrier gas and at the selected times. So that the certain
components can be stored temporarily there while the separations are made on column 1.
Liquid chromatography let us know only 15 percent of the known compounds lend
themselves to the analysis by gas chromatography owing to the insufficient volatility of
thermal stability. This is the most important the gas should not vaporize every time we
have said right. The liquid column chromatography does not leave the limit this
limitation. The interchange or combination of the solvents can provide special selectivity
effects that are absent when the mobile phase is a gas.
Traditional liquid chromatography was achieved by gravity hence analysis took place at
a slow rate. In modern liquid chromatography known as the high performance liquid
chromatography, a pressure is applied to the column forcing the mobile phase through
the through at a much higher rate and the pressure is applied using a pumping systems.
Instead of gravity we are using pumping system.
You see this is a typical liquid column chromatography the entire instrumentations we
are showing here. We have a reflux systems we can see here. I am sorry you see here.
So, we have a pressure gauge here a filter heated solvent reservoir pump and this is a
precolumn and this is analytical column. So, it is a injection head here syringe and there
is a detector and this is come to the detector. So, this is entire instrumentation system.
So, this thing should be put in a thermostatic path. So, that the constant temperature can
be achieved, right.
The general instrumentation for liquid liquid chromatography incorporates a solvent

reservoir for the mobile phase a solvent pump, a precolumn except for the bonded
phases; a pressure gauge; a sampling or injection device to introduce a sample in to the
column; a separation column and a detector.
Solvent delivery system; the solvent delivery system should have the following features;
precise delivery of solvent over a relatively broad flow range; maximum pressure
attainable. Compatibility with other components in the high; performance liquid,
chromatography system, compatibility with the wide choice of the solvents and low
noise levels in the detector resulting from any pulsations.

Removal of the dissolved air and other gasses is also necessary. So, there are three main
types of delivery pumps used for solvent delivery system namely reciprocating pumps,
syringe-type pumps and constant-pressure pumps.
Final choice of pump is made after the considering; whether the isolation or gradient
elusion is to be performed and the minimum detectability limit desired type of separation
column. The detector employed precision in quantitation and the cost of the packaged
chromatograph.

Sample introduction; the ideal sample introduction method should be able to insert
reproducibly and conveniently a wide range of sample volumes into the pressurized
column as a sharp plug with little loss of efficiency. Two kinds of sample introduction
methods are mainly used one is the syringe injection.
In the syringe septum injection a small 10 microliters already we discussed the smaller
the amount better is the peak sample is introduced into the pressurized column with a
high pressure syringe to a self ceiling elastomer septum and directly on top of the column
packing. It is to be given on the top of the column packings we have seen that thing. If
you look at the figures you will find you see syringe it is at the top of the column
packings we are giving here. You see here top of the column packing we are giving
sample introduction right. Now, sampling valves and loops sampling valves allows the
sample to be introduced reproducibly into a pressurized column without significant
interruption of flow.
Separation column; heavy wall glass or stainless tubing which can withstand high
pressures are generally used to construct liquid chromatographic columns. Columns with
an internal diameter of 5 millimeter provide a good balance between the sample capacity.
The amount of packing used the solvent required and column efficiency.
2.1 millimeter bore packed column requires about five times the inlet pressure or as the
same length of the 4.6 millimeter bore column for the same flow rate. Replacing the steel
columns excuse me steel columns with comparable cartridges permits the cartridges
permits the dedication of the cartridge to each application thus maximizing the life.
Column packing that lie in the range of three to seven micrometre in diameter are used
for exclusion chromatography.
Column temperature is maintained by circulating air baths or by using water jackets or

thermostats, that is already we have discussed it. Oven is nothing but a thermostat to add
it is constant temperature is maintained or you can we can make I mean instead of oven
we can circulate with the jackets where the constant temperature. And you can by
controlling the circulating waters we can control the temperatures of the oven also of the
column. Preceding the separation column should be short five centimeter protection
column to adsorb or filter unwanted materials. We can see there in the, if you go back to
the initial slide you will find on the instrumentation systems we have a filter. So, this will
be I mean filter all the unabsorbed unwanted materials.
Detectors choice of detectors mainly depends on the problem at hand. The different types
of detectors used we have photometric detectors, fluorometric detectors, infrared
detectors, differential refractometers.
Types of high performance liquid chromatography methods; adsorption chromatography;

adsorption chromatography is probably one of the oldest type of chromatography around.
It utilizes the mobile liquid phase that is adsorbed onto the surface of the stationary solid
phase.
The equilibrium between the mobile and the stationary phase accounts for the separation
of the different solutes. In this method the components are separated on the basis of their
polarities. Physical selectivity is dominant in adsorption chromatography. Adsorption
chromatography is used when the sample is completely soluble in organic solvents and
the molecular weight is less than 2000 for all the sample components. We have for more
than 2000 we have some different methods.

Liquid-liquid partition chromatography; this form is chromatography is passed on the
thin film formed on the surface of a solid support by a liquid stationary phase clear?
Solute equilibrates between the mobile phase and the stationary liquid.
LLC is best used to separate homologs where the separation is based on the molecular
weights selectivity right. A successful separation is achieved by establishing the proper
balance between the attraction of the mobile phase solvent and the stationary liquid
phase for the sample. Good separation can be achieved by matching the polarity of the
sample and the stationary phase and using a solvent which has different polarity.
Ion exchange chromatography; here in this type of chromatography the resin the
stationary solid phase is used to covalently at attach anions or cations on it, right. Solute
ions having opposite charge in the mobile liquid phase are attracted to the resin by
electrostatic forces right.
This method is used for compounds with the ionic and ionizable functional groups. If the
sample is the insoluble in the organic solvents, but soluble in water giving a solution that
is not neutral or is only soluble in dilute acid or alkali then ion-exchange
chromatography is used.
Molecular exclusion chromatography this is also known as known as a exclusion

chromatography sometimes it is a simple exclusion chromatography. But actually it is a
molecular exclusion chromatography let us look at it is also known as the gel permeation
or gel filtration chromatography. This type of chromatography lacks an attractive
interaction between the stationary phase and the solute sometimes there is a, this
desirable properties.
It is generally used when the molecular weight approximately exceeds 2000s in case
where it is less than 2000 we have adsorbent method here we have a molecular
exclusion. I mean gas chromatography which is used for the molecular when the
molecular weights more than, approximately exceeds 2000. This method is based on the
pores property of the substrate to sort and separate sample mixtures according to size.
The pores are normally small and exclude the larger solute molecules, but allows smaller
molecules to enter the gel causing them to flow through the larger volume. This causes
the larger molecules to pass through the column at a faster rate than the smaller ones
right. So, the pores are normally small and exclude the larger solute molecules, but
allows smaller molecules to enter gel causing them to flow through the larger volumes.
This causes the larger molecules to pass through the column at a faster rate than smaller
ones.
The chromatograph output or which is called the chromatogram. So, let us look at
consider the output from the detector that is chromatogram as shown in figure 6.

You see this is a typical chromatograph which we got at the detectors or the strip chart
recorders although we have different components, we have a gas we have a components
of A B C D, this will be I am sorry. So, this will be A B C; this will be D different
components and this will be E right.
So, this is the chromatogram for the, another one see it is a absorptions as absorptions
this is also you see here.

Since, the sample is separated in the column different peaks on the chromatogram
corresponds to different components in the mixture clear? Since the sample is separated
in the column itself. So, when it is either adsorptions it elutes and come to the detectors.
So, it is one by one is detector. Different peaks on the chromatogram corresponds to
different components in the mixture this I told you several times. The chromatograms
were obtained by separating a protein mixture using ion exchange chromatography
depends on the what type of... So, in ion exchange chromatography you can use a protein
you can separate a protein mixture. The separation corresponds to the chromatogram and
was performed at a lower pH value than 1 in b. If you look at you see the pH value is
was performed at lower value in a then b, if a has a lower pH value than b. You see the;
obviously, the separation is better in a.
So, this will tell you that is actually the how the pH will affect the actual my
chromatogram. So, pH should be maintained to a particular value. So, it should be
neither low or neither high like that this about this chromatography we are talking about.
Since the sample is separate in the column different peaks in the chromatogram
corresponds to different components in the mixture. The chromatograms were obtained
by separating a protein mixture using a ion exchange chromatography. The separation
corresponding to the chromatogram that is the a was performed at a lower pH value than
one in b this actually should like this PH we should not write like this. It should like this
PH should like this.

This shows that the operating conditions such as pH and temperature affect the output of
the chromatography if the pH changes. So, your chromatograph will change so;
obviously, they if the chromatogram change efficiency selectivity everything will. So,
this is to be. So, preside whatever the desired value of the chromatogram that is to be
maintained right. Four fold information is obtained from the chromatogram, what are the
different I mean components I am getting from the chromatogram. What are the different
I mean information’s I am getting from this chromatogram. Already we have discussed,
but let us downright, what one by one. The level of complexity of the sample is indicated
by the number of peaks. How many what is the complex mixtures I mean that we can
detect from this one. So, level of complexity of the sample is indicated by the number of
peaks which appear.
Qualitative information about the sample of composition is obtained by comparing peak

positions with those of the standards. Qualitative information about the sample
composition is obtained by comparing the peak positions with those of standards.
Quantitative information regarding the relative concentrations this is most important. So,
we are measuring the relative, because the sample if the we are not absolute measuring
we are making relative comparison relative concentrations actually we are measuring the
gas chromatography. Because otherwise, we cannot measure, because we need the
injections and the so many tin microliter, that is the best smaller the sample better is the
detection sorry.
Quantitative information regarding the relative concentration can be obtained from the
peak and the area comparison. Column performance can be checked by comparing with
the standard outputs. This column performance also because if I have if I know if I knew
the particular. Suppose if I want to calibrate the column performance I know the
particular gas with which actually we are injecting. So, that time when an unknown gas
is coming if I want to calibrate the how the column actually working. So, the column
efficiency can also be known I mean I mean can be known by looking at the.., So, these
are the 4 I mean features I mean which we will get from the gas chromatography.
And the most important thing is the relative I mean first of all the, what are the different
components of the gas present in the mixture? Second thing is the relative this is the
most prime importance. Second thing is the relative concentrations of the gas or liquid or
whatever it may be in the mixture. So, with this I come to the end of the lesson 36 of
industrial instrumentation. Preview of next lecture; welcome the lesson 37 of industrial
instrumentation. In this lesson we will study the pollution measurement when we were
talking on the pollution. Basically we are talking about the environmental pollution or
the air pollutions or the type of gases that brings the carbon monoxide and the nitrogen
oxide. All these of type of gases how it is making the air polluted and we must measure
it.
Because you know the certain I mean if it crosses some limit that is and it is not safe it is
hazardous for the human being. Pollution will be there we cannot in industrial
industrializations you know there will be pollutions you cannot avoid pollutions. You
have to be with pollutions, but what is the level we must know, because if you I mean
stay in some in forest or there is; obviously, the pollution will be less. But if you leave in
a city; obviously, there will be some pollution, because of the exhaust of the cars and
truck and all those things burning of the fuels burning of the waste. Because if you know
the waste is the big problems in a I mean cities or metropolis. They cannot dump it and
they usually they burn it, because that will reduce the volume of the waste.
So, that will cause the pollutions? The factories; the power stations all these thing will
make the air polluted. Now, we must know what is the concentrations of the different
gases? And if it is within the safe limit fine otherwise we have to check it and we have to
warn. Our goal is to warn the public or the. So, that they will also know that what type of
pollution they are going through and what is the, what should be the safe level of
contents? Now, in some countries we will find that the forest fire that also will cause the
pollutions some countries where there is where if the I mean like Australia when it is
very dry weathers.
When nitrogen monoxide or nitrogen dioxide in the sample reacts with the original,
which is produced by passing externally supplied O2 over UV lamp part of the NO is
oxidized to become a nitrogen dioxide right. Part of the NO2 generated in is in the exited
state exited states actually we are defining it by this please note.

This phenomena is called the chemiluminescence actually you see here. So, we are
defining this NO2 and this asteric means it is in exited state. I am sorry it is in exited
state NO2 asteric you see it is exited state plus O2. So, it is NO2 exited state it is
liberating the actually this will be h mu. I am sorry this will be h mu NO2. Second
equations if you see NO2 asteric which goes to NO2 plus h mu right. You can this will
be this will be h mu right. So, the light will be emitted, so above mentioned reaction is
very fast and only NO is involved without almost any effect of the other co-existent
gases. So, this also you should remember because it should not react with the other
gases. So, the other concentrations or the other gases cannot be detected by that type of
method. So, the phenomena is called the chemiluminescence. So, by this it is a nitrogen
oxide reacts with ozone. It is making nitrogen dioxide which is in exited state it will go
back to the normal nitrogen dioxide.
Now, you see we have them now this is a color code of representations of AQI. We have
given the color code representations. That is anybody can understand that is in green
means always friendly as you know the environment the green environments we are
talking about always is a good environment. So, we have given a color of green, so 0 to 5
and my AQI values. So, it is green color it is good 51 to 100 marginal moderates we
have yellow. Well, I cannot justify why it is yellow and it is some standard is to be
followed. Obviously, the green means that is our environment its green always is better
for us. Then we have a 100 and 1 to 200 unhealthy poor I mean orange. Then we have a
201 to 300 AQI value very unhealthy very poor which is on a red alert.
Red means always alert and 301 last is a critical I mean there is no more for
consumptions of the in a I mean for that air is consumption for the human being or any
living animals. So, these are things which we have considered in this I what we mean to
AQI. So, because if you give the color because it is very difficult to remember all this
numbers. So, instead of giving the numbers, so we will measure this one, but once we
define when to the public. So, give with this some colors, so that with green it is better
yellow is fine orange is not that good. Red is quite bad and if the purple it is extremely
bad right. So, with this AQI color coding we can explain to the general public. So, what
that, what the color they should have in their particular area of residence or the particular
area where they are working? So, with this I come to the end of the lesson 37 of
Prof. A. Barua
Lecture - 37
Pollution Measurement
Welcome to the lesson 37 of industrial instrumentation in this lesson we will study the
pollution measurements when we will talk of the pollution basically, we are talking
about the environmental pollutions or the air pollutions or the type of gases. That brings
the carbon monoxide, nitrogen oxide all these of type of gases how it is making the air
polluted, and we must measure it. Because you know the certain I mean if it crosses
some limit that is and it is not safe it is hazardous for the human being. Pollution will be
there we cannot in industrial industrializations you know there will be pollutions you
cannot avoid pollutions, you have to live with pollutions, but what is the level? We must
know. Because if you I mean stay in some in forest or that is; obviously, the pollution
will be less.
But if you live in a city; obviously, there will be some pollution because of the exhaust
of the cars and truck all these things burning of the fuels burning of the waste. Because if
you know the waste is the big problems in a, I mean cities or metropolis they can on
dump it and the usually they burn it because that will reduce the volume of the waste. So,
that will cause the pollutions the factories the power stations all this thing will make the
air polluted. Now, we must know what is the concentrations of the different gases and if
it is within safe living fine otherwise you have to check it or you have to warn our goal is
to warn the public or the. So, that they will also know that type of pollution they are
going through and what is the, what should be the safe level of content?
Now, in some countries we will find that the forest fire that also will cause the
pollutions. Some countries where there is a I mean where if the I mean like Australia
when it is very dry weathers. So, there is a pollutions like means, when the wind blows;
obviously, it will pollute now you see some of the pollutions. I mean some of the exhaust
of the gases like I mean you can 1 can see either by suits it is nature or by it basic. I mean
colour when it is dark in colour, but some are, so small that you cannot, it is very
difficult to see in the I mean this I mean without any microscope. So, that, so this are the
all different type of pollutions we are not concerned the pollutions of ((Refer Time:
03:06)) on this particular measurements and I mean lessons we will consider the only the
pollutions in the air, right. So, let us look at the contents pollution measurement.
Introduction, the contents or introduction, then we have a sample collection because

sample is most important, because you will know we will collect the samples from air.
Excuse me and we will how what is the type of container that should be and what type
what type of I mean how long will you store it? So, those are very important for the, for
this type of measurements. Then the detection of carbon monoxide, because carbon
monoxide is the one of the I mean extremely it is extremely poisonous gas. So, that it is
contains it is I mean it is presence in the air always is harmful for the public or for the
human or for any animal being. So, it is concentration should be below some limit. So,
that we must determine then we have detection of the nitrogen oxide and nitrogen di
oxide which is called as NOx analyzer.
NOx analyzer is not only common in the streets. I mean we are also in the environment
also in the in the in the many industries they measure this one, because to have a safe I
mean good environment for the workers or the people those who are working in the
industry or the plant. Detection of sulphur dioxide; this is another pollutions we must
measure it then detection of ozone, detection of hydrocarbons. This hydrocarbons also is
a like methane this are very I mean it is not good for the human being. So, this contains
or you must know the concentrations of the methanes also and air quality index air
quality index means visibility. Then what is the PM 10 concentration particular matter
10; that means, the particular matters with the diameter of 10 micron and less. So, that it
can detect, and this type of thing we will study in this particular lesson, right.
At the end of the lesson the viewer will know the principal pollutants in the atmosphere,
then the detection and measurement techniques. What are the different detection
measurement? Please note that in all the cases the most of the cases the detection
techniques based on the based on the optical system. So, there is a some filter. So, we
will take out the output and measure it. So, basically it is optical method of measurement
in all the cases of the gas detections. Then we have gas detection based on optical filter
as I told you earlier. Then you have colour code of air quality index we have a colour
code. So, the looking at the colour I can say the, what is the quality index of the air or the
air or the environment? So, this all the things we will study in this particular lesson.
Introduction let us see, the sources of pollutions are many and varied as I told you and
include factories automobiles burning dumps, power generating stations, household
furnaces and so on and so forth right. These are the different type of pollutions you will
get. There are many other pollutions you will find that this is the basic principle
pollutions which will have the factories automobiles is the most. I mean main source of
pollutions in any big cities or metropolis power generating stations even though
sometimes it is remote from the area from main cities. But you will find you pollute the
neighboring areas neighboring villages. Then it destroys the nearby paddy fields all those
things are there.
And household furnaces though it is not very common in our country and you know that
many places still the people uses the coal based stove, I mean the coal based or calcium
that is also pollute the environment. So, we will study all these things. Now, common
harmful chemicals that are emitted into the atmosphere include hydrocarbons oxides of
nitrogen which is called NO x or NO x nitrogen dioxide nitrogen monoxide and carbon
monoxide and sulphur dioxide. These are the main I mean basic pollutions which you
will get in the which are harmful.
Now, 10 to 17 milligram permeter cube of carbon monoxide is common on a busy streets

and more than 120 milligram per meter cube is considered dangerous. So, if it is that is
why I told you cannot we cannot I mean I mean we cannot say that the they are the any
environments will be totally pollution free. So, when you are living in a cities, we have
to live in a city large number of people live in a city. So, there will be some pollution,
but if it is within the safe limit that is fine no problem, But if it crosses that particular
limit which is hazardous. The various ill effects of these pollutants on the human health
and environments are well known, right. So, there are various types of there are causes of
from ((Refer Time: 07:49)) asthma you know this all this pollution make the people sick
So, that this content should be less and less.
It is now recognized that many chemicals undergo petro photochemical decompositions

and reactions in the atmosphere forming different pollutants which may be even more
toxic than their precursors. So, this is another problem, because you see to it will react.
So, it is a photochemical decomposition. So, it will make some other pollution pollutants
which is more dangerous. The nitrogen oxides and hydrocarbons are the most important
source of secondary pollutant smog. Smog is now it is called as smoke and fog. So many
cities you will finds in the morning this is full of smog; that means, visibility is
extremely poor. So, the I mean it is very difficult for the persons to drive in the street or
the very difficult for the air aircraft to land very difficult for the train to move, so due to
this smog. So, this is also very important sort of pollutions.
Analytical chemistry and photo spectrometry plays a very important role in the
measurements of environmental pollutions. So, these are the most important things, we
are usually measuring this type of as I told it is. Basically photo spectrometry plays a
important role in measurements of pollutions or quantify the pollutions we are talking of
just in qualitatively. But you must quantify otherwise how will you know whether it is 10
to 17 milligram per meter cube. Whether it is not it has it does not cross or did not cross
the 1 20 milli milligram per meter cube. The complete process can be broadly split up
into 2 steps sample collections and sample analysis. So, obviously, you will first the
collect the sample of the air which I want to analyze, which I want to know how much
the pollution there then the sample analysis sample collection.
One of the most important steps in air analysis is the collection of the sample. Once the
sample has been collected standard measurement techniques may be employed. The
topic of sample collection can be broken down into 2 broad categories. General
consolidations and sampling train. So, what is the different precautions we will take for
sampling collections? What is the velocity of such making the suction from the
environment, and all these things.

General considerations; the following are the points which must be kept in the mind
before going for sample collection.
Size of the sample; the volume of the size I am talking about the I mean the size of the
dust particles of the size of the sample how much will how much air you will collect
from the environment to analyze it? That is I am talking about the size of the sample.
The volume of air sample is governed by the minimum pollutant concentration that must
be measured. The sensitivity of the measurements and the information desired. The
sample size is generally chosen by a trial and error method. And trial samples of more
than 10 meter cube may be required to determine ambient concentrations sometimes we
need large meter. I mean volume to know the ambient concentration or an average
concentrations instead of taking very small, but we may need very small, but we take a
large then; obviously, that is also will be on average we can average over a particular
area.
Rate of sampling: The useful sampling rate will vary with that sampling devices and
should be determined experimentally, that we have to determine experimentally. The
most sampling devices for gaseous constituents have permissible flow rate as I told you
we have to, I mean collect it with some suction pump. So, that type of thing. So, velocity
will be 0.003 to 0.03 metre cube per minute.
Now, duration of sampling that is also important. Time of the day and the duration of
sampling will be determined by the information that is desired. Because the time because
if you go to a busy streets if you take a sample at the 5 o clock in the morning there is no
use hardly that time in pollutions, but at the 10 o clock in the morning if you take
something you will get the estimate. You will get the exact the how much the pollution is
there or what is the worst case pollutions in the streets in the evening or in the morning?.
Sampling period will give the indications of only the average concentration during that
periods means suppose, I am taking for 1 hour, so that means, I am collecting over
averaging over 1 hour. So, that is that is the good measurement.
Then, storage time of sample storage; this is one important storage time of the air sample
should be kept to a minimum. This should be because in some cases it can be
decomposed. So, that should be after making the collections you should make the quick
analysis. They should be protected from heat and light this is another important thing.
Care should be taken that the desired test component does not react with the other
constituents or with the container.
So, the container should be I mean sort sort of inert some minute probably made of glass
or stainless steel. So, that it will not whatever the pollutions or the like the gases that
should not react with the container those are those things should be considered. Gaseous
samples are sometimes collected by the absorption adsorptions on to a solid and the
gaseous must not be lost by desorption prior to analysis. Many samples you know
collected by adsorption; that means, it will be on the solid material and it should not be
released or desorption prior to the analysis. So, because if I do that if it desorbs before
the analysis; obviously, that particular gas will be lost we cannot analyze. Even though it
will be showing there is no that type of gas is there, but actually it was there by that time
it is lost.
Sampling train, the requirement for intermittent air sampling are a vacuum source a
means of measuring the amount of air sampled and a collector. So, you have vacuum
source. So, that and the means of measuring the amount of air sample and the collector.
Because if there is no vacuum source that we cannot collect the sample, right.
Vacuum source; what are the vacuum source motor or hand driven vacuum pumps
aspirators and automobile vacuums are generally used to draw the sample through a
collection devices, clear. So, this is a basically pump, by which you will collect the
sample from the environment and store in a collector. When vacuum devices are being
used to draw the sample through a filter in which the pressure loss may build up during
the sampling it is recommended that some constant flow devices to be attached it, right.
Some constant flow devices should be there. So, that the, because if I use a filter, so
obviously, there will be some pressure loss across that filter.

Now, metering devices, flow measurement devices are of 2 general types we have seen
this we have studied this flow measurement devices in this particular course of industrial
instrumentation. Thus some I will make as you know some will make the rate
measurement, some will make the volumetric flow measurement, some will make the
mass flow measurements. We are more interested here instead of the rate measurement
on the volumetric measurement, right. Flow measurement devices are of 2 general types
those measure the rate and those measure the volume or you can make the total volume
also. This devices the devices that measure the flow rates are small and inexpensive, but
have the disadvantage of measuring only instantaneous rate of flow 1 is rotameter.
Rotameter is just you make the instantaneous rate of flow it will not make the total
volumetric flow of a device over a certain amount of time.
Devices that measures the volume record that flow passing through the, them through
them and are therefore, more useful they are; however, bulkier and are more expensive.
Rotameter is a very simple devices as you know it is a very I mean indicating sort of
instruments, we have studied this in details in the in the flow meters. But other devices
where we can suppose a flow meters like a turbine flow meters. And all these things or
any other gas flow meters you can measure this one like orifice meter also, because that
is bulkier obviously. All gas measuring devices should be calibrated before use because
the you are measuring the volumes. So, how much collections we are taking? So, it
should be calibrated before. A direct calibrations with the actual gas can be carried out
by means of gasometer system. Here the gas flowing through the flowmeter during a
time interval is trapped in the gasometer bell, and it is volume is measured a gasometer
bell is there and the volume is measured.
Collector the last component of the sampling train is the collector which may be of
variety of types depending, on the particular applications right. So, these are the very
important thing. So, for consideration…

Basically, if you look at the gasometer basic principle is something like that this
gasometer is basic principle is that it has a known volume of I mean device direct
calibrations. So, the actual gas can be carried by means of gasometer system and the gas
flowing through the flow meter is trapped in the gasometer bell and that it is volume is
measured, right. It is a like this one it is a I mean if you look at it will be like this, one is
a, if I take a new page and new color.
So, it is a bell sort of thing over this 1 and water is there. So, that it is trapped and going
like this. So, the bell will come now this water also gas will be trapped here in this. So,
that the volume will be known us, it is as simple as that that is it is called basically
gasometer.
Collector: The last component of the sampling train is the, a collector which may be
variety of types depending on the particular applications.
Aerosol contaminants; one of the most commonly used means of collection aerosol
contaminations is a filtration. So, fiber filter wood fiber paper glass fibre asbestos these
are granular filter glass or metal or porous ceramic sand and membrane filters are used,
cellulose ester. So, these are the different types of filters which are used commonly mean
of aerosol contaminants in the filtrations.
A second type of collection device for both the solid and liquid aerosols contaminants is
a impinge, right. The aerosol impinge on a surface exposed to the air stream and those
contain both dry and wet impingers. More sophisticated collectors include electrostatic
re-liberate ((Refer Time: 18:50)) that energy in the form of a heat when they recompress.
Carbon monoxide detectors can be split into 2 types the method depends on the
measurements of energy of heat liberated after IR rays have passed through the sample.
After the heat I mean after the IR rays pass through the sample. So, that is one method
next method is; that means, the as you as I told you; that means, I have a atoms I mean it
will delete will show you. The method depends on the electrochemical analysis, these are
the 2 methods; one is the total I mean optical method another is electrochemical method
right.
Now, detectors based on the heat absorptions; that means, it looks like you see the what
will happen? That means, it will have like this one. So, I have here. So, we have like this
one. So, after light energy it is bonds like this one. So, it is going on like this one excited
after one the when the, if it is infrared passed. So, this will be detect liberate the light
energy that will be detected, right. That is a basic principle.
Carbon monoxide analyzers pass an infra red light beam, alternatively through a
reference cell containing a non absorbing gas like N 2 nitrogen and the sample cell
containing air with carbon monoxide to be measured. The infra red light of 4.67 micron
wavelength is absorbed from the sample to the extent that corresponds to the carbon
monoxide concentration, right.
The next step to receive the infra red beam through a nitrogen filled reference cell and a
sample cell, which now have a different intensities due to light absorptions by carbon
monoxide in the sample cell then nitrogen filled reference cells, right. So, this is the
thing. The difference in the measure intensity is converted to a carbon monoxide
concentration, right. So, these 2 difference; one will be without carbon monoxide and
other is with carbon monoxide. So, be the difference.
Now, detectors based on the electrochemical analysis looks like this, when the gas is
electrolyzed, while maintaining the potential between the electrodes immersed in
electrolyte the current assumes values proportional to the gas concentration, right. The
gas is electrolyzed maintaining the potential between the electrodes immersed in the
electrolytes the current assumes value proportional to the gas concentration why you
see?
Because the electric potential at which the electrolysis takes place depends on the gas
only carbon monoxide concentration and the carbon monoxide concentration is
measured. The carbon monoxide concentration is therefore, obtained from the potential.
So, if I measure the potential that potential can be calibrated in terms of the carbon
monoxide concentrations in the gas because electric potential at which the electrolysis
take place depends on the gas only. So, depends on the concentrations of the carbon. So,
if that voltage is calibrated in terms of you can directly calibrate that voltage in terms of
carbon monoxide concentration.

The NO x measurements; it is a very common not only in our environment, but also in
industry NO x; that means, nitrogen oxide and nitrogen dioxide this 2 types of pollutions.
I mean gas we are calling it the NO x NO x analyzer NO x measurement NO x gas.
Whenever we are talking about we have to I mean of this NO x means NO x it means
NO subscript x it means that the nitrogen oxide and nitrogen dioxide. When nitrogen
monoxide sorry I was not instead of nitrogen oxide nitrogen both are nitrogen oxide. So,
NO is the nitrogen monoxide and nitrogen dioxide is NO 2. When nitrogen monoxide or
nitrogen dioxides in the sample reacts with the ozone which is produced by passing
externally supplied O 2, over UV lamp part of the NO is oxidized to become a nitrogen
dioxide, right. Part of NO 2 generated in is is in exited state exited states actually we are
defining it by this one please note.
This phenomena is called the chemiluminescence actually you see here we are defining
this NO 2 this asteric means it is in exited state I am sorry is in exited state NO 2 asteric
if it is exited it is plus O 2. So, it is NO 2 exited state it is I mean liberating the actually
this will h mu I am sorry this will be h mu NO 2 second equations if you see NO 2
asteric which goes to NO 2 plus h mu right. We can this will be if I this will be h mu
right. So, the light will be emitted. So, above mentioned reaction is very fast and only
NO is involved without almost any effect of the other co-existent gases. So, this also you
should remember because it should not react with the other a gases other concentration
of the gas cannot be detected by that type of method. So, the phenomena is called the
chemiluminescence. So, by this it is a nitrogen oxide reacts with ozone it is making
nitrogen dioxide which is in exited state it will go back to the normal nitrogen dioxide
state or nitrogen dioxides liberating light, right. So, the, this method is very fast and only
NO is involved without almost any effect on the other co-existent gases.
If the NO is at a low concentration the quantity of luminescence is in proportional to the

concentration. The nitrogen monoxide is at low concentration the quantity of the
luminescence is in the proportions of that to that concentration. Now, NO x analyzers
separate the sample gas into the 2 portions in 1 portion nitrogen dioxide is reduced to NO
by the NO x convertor and used as the sample gas for the measurement of NO x which is
NO plus NO 2 nitrogen oxide and nitrogen monoxide and nitrogen dioxide.
In the other the NO sample gas is used as it is, these sample gases are switched to the NO
x NO and reference gas lines by the solenoid valves every .5 seconds and are in turn
introduced in the reaction chamber. On the other hand the open air that was separately
sucked through the air filter is dried by the self reproducing type silica gel dryer then
introduced as ozone gas into the reaction chamber by the ozonizer. On the other hand the
open air that was separately sucked through the air filter is dried up dried by the dried by
the self reproducing type silica gel dryer ad then introduced as ozone gas into the
reaction chamber by the ozonizer.

In the reaction chamber the sample reacts with ozone. The light emission involved in the
reaction is detected by the photodiode. So, whatever the light detection is coming out is
detected by the photodiode. It is in turn gives a current that is calibrated as a
concentration. So, photodiode output can be I mean photodiode output will be current.
So, that can be you know the photodiode always can be used for detection of some lights.
So, that can be calibrated in terms of concentration.
Now, what is the precautions and preprocessing of these that we must note one of the
problem with No x measurement is it is affinity to water. So, any water, so immediately
it will be absorbs and you cannot find the NO x. So, that a container should be totally air
or water free that is most important thing for this reason it is important to remove the
practically all the water from the sample because when it is in the sample. So, it may
have other ah, so that, so the when before analyzing. So, this from the sample before
analyzing the sample the all the water should be removed. Dryer is used for this purpose
because silica gel dryers and all these dryers. Obviously, it will remove the moisture
context in the NO x keeping the NO x as it is for analysis purposes because ultimately
we have to make the analysis of this NO x.
Now, sulphur dioxide is another form of pollutions in the environment. So, we must look
at what is the this type of I mean pollution this is a very hazardous . So, we must know
the contents of the sulphur dioxide in the air. Analysis of sulphur dioxide is based on the
principles of fluorescence spectroscopy as I told u mostly these are basically depends on
spectroscopy. So, this is also fluorescence spectroscopy. Sulphur dioxide has a strong
ultraviolet absorptions to wavelength between 200 nanometer to 240 nanometers within
that it has strong ultraviolet absorptions. The absorptions of photons at these wavelengths
results in the emission of fluorescence because where this energy will go it cannot just
get lost. So, that this energy will the consumed in the sulphur dioxide atom and it will
release the fluorescence, right. We will measure that fluorescence and calibrate the by
spectroscopy and and calibrate that in terms of the sulphur dioxide concentration.
A zinc discharge lamp and a band pass filter and a optical band pass filter; obviously,
radiates ultraviolet light at 215 nanometer into the reaction chamber where it interacts
with the sulphur dioxide molecules, right. The fluorescence is measured perpendicularly
to the beam using a photomultiplier tube. Photomultiplier is nothing but is a series of I
mean instead of now a days instead of photomultiplier tube you can use a photodiodes
also or photo transistors. So, photo multiplier tube is just it multiplies the it increases the
output current as simple as that nothing more you must learn it in increases the current at
the output. The amount of fluorescence is directly proportional to the concentration of
sulphur dioxide which we are measuring by the photo multiplier tube.
One of the major problems with the sulphur dioxide is extremely corrosive nature and
affinity to water same as the sulphur dioxide as I told you. But sulphur dioxide I mean
sorry nitrogen NO x not that corrosive, but sulphur dioxide has is corrosive in nature as
well as it is affinity to water. So, both you have to take it is more I mean we should take
higher precautions for analyzing the sulphur dioxide then nitrogen nitrogen dioxide if I
remove all the water that is fine. So, there is by some dryer or some other means no
problem, but here that we have to think that whatever containers we take that is that
should not react to the sulphur dioxide. The presence of any condensate will quickly
remove all traces of sulphur dioxide from a measuring system. So, all this if there is a
little of presence of any water; obviously, it will remove all the traces. So, ultimately
while we are analyzing you would not get any SO 2 giving you a wrong measurements
or wrong indication that is un-desirable. In addition there is a absorptions line for a water
very near line for the SO 2, this is another problem.
This means that the presence of any water vapour will affect the low readings of sulphur
dioxide to be expected, there is another problem. So, not only the reaction, but this is line
also lies very close. So, that there will be low reading of sulphur dioxide in presence of
the water, right. Water vapor or water I mean the steam whatever you call. Infrared
sensor for sulphur dioxide has the advantage of lower cost, but at the expense of greater
size infrared sensors are there, but it has a has a advantage of the lower cost, but;
obviously, the sizes will be quite big in this case. Nevertheless infrared source sensors
are 1 of the most preferred methods are measuring SO 2 concentration in the field
nowadays, right. Because of those advantages of his I mean that that line corresponds to
a very close to the line of the water vapor.
Now, ozone detections you see the how it will detect the ozone zone detector works in
the principle of absorption spectroscopy or spectrometry or photometry right. The
wavelength range for ozone is 220 to 330 nano meter and the absorption spectrum is
extended to 253.7 nanometer, which corresponds to the main emission line of mercury.
So, the ozone absorption coefficient was experimentally defined at 253.7 nanometer.
To determine ozone concentrations a sample of ambient air is drawn through a flow cell,
which is irradiated with ultraviolet light of wavelength 253.7 nanometer from a mercury
vapour lamp mercury vapour lamp will make a particular wavelength this 253.7 as I told
you it is a line is same as the mercury. So, it will produce that particular wavelength
stable wavelength. The intensity of ultraviolet light detected by the instrument is reduced
depending on how much ozone is in the sample to absorb it if there is more ozone;
obviously, it will absorb more. So, the intensity will be reduced. So, that is a intensity
ultraviolet light detected by the instrument is reduced depending on how much ozone in
the sample to absorb it, if there are more less ozone. So, we will get more better higher
output if the more ozone I will get less output. So, that is the indications of the
concentration.
A second source that has been scrubbed of ozone using a manganese dioxide scrubber is
then drawn through the absorption chamber resulting in a higher intensity of light
reaching the detector, clear? The difference in the intensity of the 2 yields the
concentration yields the concentrations of the ozone in the sample difference of these 2
intensity will give us the sample.
Now, measurement of hydrocarbons as I told you hydrocarbons means combinations of

me at this methane ethane. All these things are basically we are calling a hydrocarbons
no petrochemical industries sometimes we call it hydrocarbon industry also. Now, the
flame ionization detector is the most commonly used detector for the analysis of organic
compounds or hydrocarbons in the both in the environment as well as in the factory. Its
sensitivity linearity and user friendliness make it the ideal detector for the measurements
of hydrocarbons.

The ionization mechanism of organic substances in the flame is carried out in 2 phases,
what are the 2 phases let us look at. The cracking of organic compounds in the centre
zone of the flame and forming the CH CH 2 CH 3 radicals, CH radical, CH 2 radicals
and CH 3 radicals by cracking these all ready compounds in the centre zone of the flame
and forming of this type of hydro radicals. Chemical ionization in contact with the
oxygen according to the reaction CH radical I mean plus O which will give you CHO
plus plus electron 1 electron.
The molar response is defined as being the proportional to the number of atoms in the
carbon of the molecule. The electrons are extracted from the flame using a voltage
polarization applied between the nozzle and the collecting electrode between the nozzle
and collecting electrode it will be.
Now, this is the figure of general functional diagram of our you see the polarizations. We
have a collecting electrode we have a flame here then we have a zero air the sample is
injected through E 1 and this electrometer, I am sorry the electrometer signal is
electrometer is nothing but a high input impedance meter voltage meter. So, it will
collect the signals and like this right.
Current obtained from the collecting electrode is converted to a voltage subsequently it is

digitized by ADC for further processing right, because if it is analog signals I have to
convert to the digital domain by ADC. Figure 2 shows the detail mechanism which I
have just shown the detail mechanism of this type of detector. The flame ionization
detector sends a signal proportional to the number of atoms composing the hydrocarbon
molecules present in the sample or total hydrocarbons right. Now, another important
thing is the air quality index the, what is the quality, how we will index, how we will I
mean make the air safe? And all these things you must have some index because the if
you look at the general I mean public they will not understand what is this concentration
or what is the safe thing? So, let us better it is expressed what is the concentration of the
carbon monoxide in the air what is the concentration of sulphur dioxide? Instead of that
we make the air quality index it is inclusive of visibility also the also the quality of the
air. So, this air quality index, now let us look at meaning and use.
The air quality index is an index or rating scale for reporting daily the ambient air
pollution recorded in the monitoring sites. So, it will usually it I mean records or it will
gives you it will gives you index how the air quality in a particular site is good or bad
right. So, that is I am saying air quality index is an index or rating scale for reporting
daily the ambient air pollution recorded in the monitoring sites.
The 2 objectives of AQI are to inform and warn the public about the risk of exposure to
daily pollution levels. As I told you earlier usually we will define these things;
obviously, we must have a goal that we have to reduce this to some level. So, people also
will know. So, there should be some protective measure to come down to bring it down
to particular level. To enforce regulatory measures for moving AQI improving AQI. So,
that what is the precautions we should take, what type of you know that you see that the
reason is we have the air quality is very extremely poor in India and the metropolis like
Delhi, because there is a lot of automobile for pollution there. So, there converted to all
everything to CNG. So, which will give you less pollutions. So, this actually this type of.
So, there every time they will measure the, what is the quality what is the quality of the
air we are inhaling what is the dangerous level. And once they have switched to CNG all
the cars all the buses were people found the air quality index improved a lot. So, this will
give you some measure or the viewer with a single index people will know the, how my
water how the environments of the air which I am inhaling good or or how good it is.
The AQI or air quality informs the public about the quality of the air and associated
health effect resulting from inhaling polluted air what is the ill effects and how you can
you reduce? And all this thing is to be discussed the higher the AQI value greater is the
level or air pollution. So; obviously, the higher the value always the desired value should
be lower and lower. So, that is most important thing for air pollution four major pollute
pollutants including the suspended particular matter is PM 10 are considered for
reporting AQI PM 10 is a particular matter of 10 micron of size 10 micron or less there
are PM 2.5 also, but we are considering a PM 10. That means, not more than that right.
So, we are considering the PM 10 particular matter of size 10 micron.
Measurement and calculations the concentration of the pollutant is measured by air

quality monitoring equipment at a monitoring station. This pollutant concentration is
converted to an index value using the following equations index value, pollutant
concentration, pollutant goal concentration into 100. What is our goal? What is actually
desired value of the concentration and what is the actual concentration now? So, that will
give you the index value multiplied by 100? That is we will some quantitative measure
for the quality of air people are breathing.

The pollutant goal concentration is used to calculate the index value is either the air
NEPM standard for that standard for that particular pollutants or the EPP goal in the case
of visibility. Let us look at, what is NEPM and EPP?
NEPM is the NEPM is an national environment I should take a different colour. I think
this are not very good colors NEPM is a national environment protection measures every
country has this thing. So, this is the national protection environment protection
measures is a basically are broad framework setting statuary instrument. We find in the
national environment protection council and the outline. And agreed national objectives
for protecting or managing particular aspects of the environment this is basically NEPM.
And also we have EPP you see there that EPP is basically is environmental protection
policy environmental protection policy. So, this also we will look at all these things now.
So, these are the I mean now see this PM 10; that is also we are talking about the PM 10.
Let us take a new page. PM 10 is the particulate matter particulate matter of includes
particles with a diameter includes particles with diameter 10 micron 10 micrometer
which is coming as 0.0004 inches or I should say or 1 seventh width of human hair right.
We have then PM 2 point five that is we are not discussing previously it has a tsp
monitors that is also we are not discussing here. The pollutant goal concentration used to
calculate the index value is either the air NEPM standard for that particular pollutant or
the EPP air goal in the case of visibility. Visibility is most important, because you see
that if the air I mean if there is a if there is a suspended particles or the if there is the PM
10 is there. So; obviously, it will reduce the visibility of a person. So, what is the goal?
What should be the visibility whether it is a 10 kilometers or 20 kilometers? That we
must know. So, this also we will give you some standard for particular country. So, let us
look at that the air NEPM standards and EPP air goal used to in calculating the index
value are presented in table one.
You see here that is an air NEPM standard that ozone it is 0.10 parts per million. So, we
have taken the average time of 1 hour nitrogen dioxide 0.12 parts per million that is 1
hour sulphur dioxide 0.20 parts per million for 1 hour carbon monoxide. We have taken
for 8 hours and the PM 10 we have taken it is 50 microgram per meter cube for 24 hour.
This is the standard an NEPM standards and pollutants and the visibility is 20 kilometers.
So, EPP is that air that is our goal is 20 kilometer right for 1 hours the average in time.

The EPP air goal for visibility reduction particles is 20 kilometer visibility which means
you would be able to see the clearly for at least 20 kilometers. Obviously, these the sort
of pollution, because some pollution we should why we have taken the PM 10 you see
the this 2 are should be different because some gases are this carbon monoxide. These
are not visibile ozone is not visibile that will pollute the air the 2 types we are
considering one is the visibility concentration another is the health hazard just health
hazard. Because visibility is also making the it is not hazardous to the I mean people or
the persons or the public. But it makes the our other things difficult; that means, the
landing of the aircraft or taking off or moving of the train or moving of the cars
everything become difficult. That is also measurement of the pollution that is the reason
we have PM 10 we have introduced this goal is related to the light scattering coefficient
value measured using a nephelometer. A scattering of 235 or less is equivalent where
Mm minus 1 is the per million meters is equivalent. So, a scattering of 235 millions per
millions meters or less is equivalent to visibility of more than 20 kilometers. So, this is
also will give you that what is the visibility? Usually in many countries nowadays they
straight away will give you the visibility. So, that from that measurement they can see
the visibility.
Meaning or interpretation of AQI reading the higher the AQI value the greater is the
level of air pollution and greater is the danger to health not PM PM 10 is also, but that is
also considers with the visibility. AQI for suspended particles is 305 and is highest out of
the 5 pollutants and it is reported as an AQI of 305 right air quality index of 305.
AQI indicates the maximum safe levels for a pollutant and above this level the pollutant
has undesirable impact on the people and the environment. So, this will give you some
indicates the maximum safe levels for the pollutants how much pollution it can cost. So,
it goes above that level. So, it is not good for the health or the persons or the any animal
being. If AQI falls below 50 or the air quality is considered extremely good if it is falls.
Obviously, as I told you if the AQI is higher pollution is more an AQI reading between
51 to 100 indicates that the moderate air quality and AQI reading from 101 to I think this
should be 2 200 I am sorry this will be it should 200 . So, if AQI falls below 50 and the
air quality is considered good an AQI reading between from 51 to 100 indicates
moderate air quality. And AQI reading from 100 very poor air quality and an AQI of
above 301 is regarded as a critical to human health. This is some category it depends it
varies from country to country, but this is the; obviously, always we will desire that the it
is desirable that the AQI should be below 50, right.
Now, you see we have then now this is a colour code of representations of AQI we have
given the colour code representations that is anybody can understand. That means, green
means always friendly as you know that the environment green environments we are
talking about always is a good environment. So, we have given a colour of green. So, 0
to 5 I mean AQI value. So, it is green colour is good, 51 to 100 marginal moderate
yellow. Well, I cannot justified why it is yellow? It is some standard is to be follow;
obviously, the green means that our environment is always is green always is better for
us. Then we have 101 to 200 unhealthy poor I mean orange then we have a 201 to 300
AQI value very unhealthy very poor which is once a red alert. Red means always alert
and 301 plus is a critical. I mean it is a no more for consumptions of the, I mean for that
air is consumption for the human being or any living animals. So, these are the things
which we have considered in this, what we mean to be AQI. So, because if you give the
colour, because it is very difficult to remember all this numbers.
So, instead of giving the numbers we will measure this one, but once we define to the
public. So, will give you this some colors so that green is better yellow is fine orange is
not that good red is quite bad and if the purple it is extremely bad right. So, with this
AQI colour coating we can explain to the general public. So, that what the colour they
should have in that particular area residence or a particular area, where they are working?
So, with this I come to the end to the lesson 37 of industrial instrumentation. Welcome to
the lesson 38 of industrial instrumentation. In this lesson, we will cover the control valve.
In fact, this lesson 38 39 both will be dedicated to the I mean control valve 1 and control
valve 2. Control valve as you know it is a final control element in a process and it is most
important part of any industrial instrumentation. Now, so far we have studied as you
know the many sensors which actually measure the possible variable flow temperature
pressure and viscosity pH of so on and so forth.
Now, all this actual measurement is basically to control some that particular process
suppose in the case of temperatures or in the case of flow we have to control the
temperature we have a set point. So, our goal is to I mean to make that whatever the set
point I mean determine or fixed by the linears. So, our system should attained that
particular value right. So, some or the other we need control valve now control valve is
basically is a is a device which we which will resist the flow of the fluid either liquid or
gases through a pipe right. This is the basic purpose and directly it is controlling the flow
indirectly it is controlling the temperature and any other process parameters right.
Suppose I have a indirect heating of any of a boilers. So, we are heating it by a super
heated steam in that case by controlling the amount of steam. Also we can raise the
temperature of the boiler and directly in the case of flow measurement. There is no
problem I mean these are the only factors which is to be included which I have to think
of otherwise not.
The general form of the sizing equations or the compulsively fluid is…
You see Cv equal to Q upon NFpP 1 Y to the power X upon G T 1 into Z equation
number 19 where the legends are X is a ratio of the pressure drop to upstream absolute
static pressure. Ratio of the pressure drop to upstream absolute static pressure that is the
reason we are interested in P 1 the previous equation number 18. Also we have seen that
the P 1 is important not the P 2. Y is the expansion factor ration of the flow coefficient
for a gas to that for a liquid at the same Reynold number. Z is the compressibility factor
and T 1 is the upstream fluid temperature in Kelvin. So, all these I mean I think all the
legends are defined now every time you see that Cv is the most important thing in the
bulk parameter. There is no other, because if you define Cv or if you find Cv all other
parameters will be automatically picked there are some parameters like small d capital D.
So, first in the while you are choosing the valve first I have to calculate the, what are the
desired Cv you want for a particular applications once you find Cv. So, from the selected
list of the valve, so you try to choose Cv which will match very close you may not get
the exact match. You may get 7 to 90 percent of the value of the Cv which we have seen
that will be the desired range of the control valve. So, that particular control valve is to
be chosen and is to be installed in the process. Now, with this I we will cover this in
industry I mean this control valve we will continue this control valve in the lesson
number 39. Also we will go further details and we will solve some problems in the
lesson 39. So, at that with this I come to the end of the lesson 38 of industrial
instrumentation
Prof. A. Barua
Lecture - 38
Control Valve - I
Welcome to the lesson 38 of industrial instrumentation. In this lesson, we will cover the
control valve. In fact, this lesson 38 and 39 both will be dedicated to the I mean control
valve 1 and control valve 2. Control valve as you know it is a final control element in a
process, and it is most important part of any industrial instrumentation. Now, so far we
have studied as you know the many sensors which actually measure the process, variable
flow temperature pressure and viscosity pH and so on and so forth. Now, all this actual
measurement is basically to control some that particular process suppose in the case of
temperatures or in the case of flow we have to control the temperature we have set point.
So, we our goal is to I mean to make that whatever the set point. I mean determined by
the or fixed by the engineers our system should attain that particular value, right.
So, some of the other we need control valve. Now, control valve is basically is a device
which will resist the flow of the fluid either liquid or gases through a pipe, right. This is
the basic purpose and directly it is controlling the flow indirectly it is controlling the
temperatures and any other process parameters, right. Suppose I have a indirect heating
of a boiler. So, we are heating it by a super heated steam in that case by controlling the
amount of steam also we can raise the temperature and of the boiler. And directly in the
case of flow measurement there is no problems we have seen that by changing the flow.
That means, there changing the flow through the pipe, I can control it. I can achieve
some set point which is pre-determined.
Now, all this I mean basically is to as you can see that this control valve is basically is a
variable resistance. It will offer a variable resistance to flow. And why we to we are
telling it control valve? Because and there is a distinct difference between the valve and
the control valve, because in the control valve signal will come from some controller.
And it will offer some resistance either I mean another open the valve stream or it will
down the valve stream. So, that I will achieve the, if the desired flow is more I mean then
the, what is actually existing? So, that I have to open the valve if the desired existent
existence is I mean is less than the I mean more then I have to close the valve. So, if we
can do it manually, but in the case of control valves it is done automatically right. So, let
us study this control valve in details. Lesson 38 control valve 1 the contents are.
Introduction to the control valve and basic control valve terminology then you have
valve sizing, because the valve sizing is also important.
Now, at the end of the lesson the viewer will know the different components of a
controller. What are the different components? What are the different terminologies?
You must know first then we will go for the other details construction. What are the
different flow characteristics? We can achieve in a control valve we will find there are 3
different characteristics we will achieve in the control valve. What is the valve sizing
coefficient and valve discharge coefficient? These are 2 important parameters for
designing a control valve. This also we will know basic sizing procedure for the laminar
liquids and turbulent liquids and chocked flow and gases. Chocked flow you will find as
the most of the cases the flow is chocked. So, it is called chocked flow. And also for the
gases because all these thing. We are not considering the solids here we are at best we
can consider the suspended particles in the liquid. So, this we will cover in details let us
look at the introduction.
The control valve is the most popular final control element. That is the last device in a
control loop right; that means, I am measuring with some sensor, suppose I have a flow
meters. So, flow meters in the case of orifice meter, venturi meter or any other meter, but
ultimately to control the flow in the pipe I need a valve. So, it is basically a final control
element. I will measure the flow actual existing flow it is if it is more than the set point.
So, I will I will close the I mean I will reduce the flow through the pipe by the control
valve if it is less than the desired value, so I will open it. So, that the more flow can go,
so ultimately, so that is the reason we are calling it a final control element in the any
process.
The control valve manipulates a flowing fluid such as gas water stream or chemical
compounds to compensate for the load disturbance, and keep the regulated process
variables as close as possible to the desired set point. Now, this question my come
several times I am saying that the if it is more if it is less, why it will be like this one?
Because I have a set point it is not a. So, I know that the flow should be suppose 10 to
the power 3 liter per hour. So, why should it change? Because I have already set points I
have a some pump which is flowing to the I mean which is pumping the liquid to the
pipe. So, why should it change? It will change due to the load disturbances, due to the
load disturbances if it changes. So, I have to open the control valve I have to close the
control valve. So, that the whatever the desired flow of liquid. What are the set points did
there? So, liquid should flow accordingly through the point, right.
A control valve selected on the basis of the process control system requirements and the
characteristics of the process fluid, right. There are different types of control system
requirements. First of all you will you have to look at and we have to look at the process.
What type of fluid it is whether it is a clean fluid whether it is a dirty fluid whether it is a
very highly viscous fluid. So, all this depend I mean considerations we have to keep in
mind to choose a control valve, or control valve characteristics I should say.
Now, you see this is a basic control valve you can see here, you see this is a control
valve. Now, you see here that we have a pneumatic control valve, I mean we have a
bonnet here. You see here there is a diaphragm and let me take a different color. I have a
diaphragm here, and this is called you see this is called the plug of the control valve. The
liquid fluid is coming either gas, I mean or anything is coming through this and it is
going out through this one, right. And you there here that there is a there is a plug and
this is called the seat of the control valve. This is called the seat this is plug and this
vertical I mean this is called the stem of the control valve you see this portions. I am
talking about this is this portion this is called the stem of the control valve, you can see
here this is called the stem of the control valve.
Here you see there is a packing this packing is necessary. So, that the valve this stem can
move only in the vertical directions, as well as it will resist the flow of the fluid through
this portion. Liquid will flow from this one according to the opens it will come to this 1
and it will fluid out it will go out like this one, clear? And we have a bonnet here. A
spring is necessary, because when you when it is removed when the pressure is removed;
that means, when you if you if the pressure from this pneumatic signals is removed. So,
valve should this stem should go to it is previous position. This type of valve you can see
it is called air to close. Because air is to be come through pneumatic signal will come
through this 1 and it will close it. So, it is called air to close valve right and this is a cage
and this is a packing as I told you this is the spring which is needed to. I mean move the
stem of the control valve to the position when it was normally when there is no
pneumatic signal there.
So, this type of say I mean control valve when there is no pneumatic signal. So, the valve
will be fully opened. So, to give the full pressures once you give the full pressures air
and through pneumatic signal. Because this is totally sealed this is circular in shape. So,
the seat will be and the valve stem will sit on will come on the seat of the control valve,
right. And this is the fluid and liquid is flowing through this one and it is coming out to
this one. This is the total details of a typical control valve. Now, there is different shape
of this plug you will find that, I mean if we choose a different set shape of the plug. I can
get a I mean different types of characteristics that basically linear characteristics, equal
percentage characteristics, quick opening characteristics depends on the plug. What is the
shape of the plug? We will go through the details of that one.
This is the the control valve previously what I have discussed it is called a single seat
valve. But here we have a double seat valve, the double seat valve has a some I mean I
have not drawn the full. I have drawn only the stem and all these things double seated
valve is also necessary for many applications, fine.
Bonnet it is the portion of the valve that contains the packing box and the stem seal and
can guide the system stem, right. As I told you if I look at bonnet, this is the called the
bonnet, right. It provides the principle opening to the body cavity for assembly of the
internal parts or it can be an integral part of the valve body. It is can be of the integral
part of the valve body.

It can also provide for the attachment of the actuator to the valve body, clear. Typical
bonnets are bolted threaded and welded pressured seals, because this are totally sealed
otherwise there will be a problem.
Cage is another part of the control valve. What is cage? A part of a valve that surrounds
the plug seat and it can provide the flow characterization, right. So, the plug and the seat,
which I have shown several times. So, that is actually contains the that is makes the cage
of the control through which the liquid is flowing. It is coming to the coming through the
in in path fluid in path and it is going out of the fluid out path, right. So, it can provide
the flow characterization what type of I mean I mean characterizations you want that is
can be determined by this one. It also provides stability guiding balance and alignment
and facilitates assembly of the other parts of the valve, right. The valves of the cage
contain openings for inlet and outlet of the valve flow we have shown this one. So,
through one it is the inlet the valve is coming and through which the valve is going out.
Diaphragm, it is a flexible pressure responsive element that transmits force to the stem,
right. It is a flexible pressure responsive element. That transmits force to the stem it is at
the top of the control valve, it is steel diaphragm. We have already studied diaphragm
gauges all this things. This is basically giving the force to the against the spring which is
already in existence to the stem. So, that the stem will move down as well plug will
move down it will go and put on the cover the seat of the control valve. Plug; it controls
the volumetric or mass flow of the fluid through the cage it controls the volumetric or
mass flow of the fluid through the cage. Port; these are the openings through which the
fluid will flow in flow out of the valve.
Seat; it is the area of the contact between the closed member of the cage and its mating
surface. Already we have shown this seat, actually there is a contact between there is
always when the valve when the plug totally rest on the seat that is it is totally closed
when it is fully opened. So, that will totally liquid. So, the through the seat actually
always the liquid is flowing in between the space in between the space in between the
seat and the plug the liquid is flowing through the valve. Trim; the internal components
like plug seat and stem that modulate the flow of the control fluid are collectively called
as trim. So, the plug seat and the stem of the control valve, that are collectively called as
a trim. Stem; the part that connects the diaphragm to the final control element, final
control element here in this case of the control valve is the plug. So, the when the plug is
connected plug is connected to the diaphragm through the stem. And the stem is
surrounded by a spring.
Now, basic control valve terminologies control valve assembly. It includes all the
components normally mounted on the control valve like the valve body assembly,
actuator positioned, transducers limit switches etc.
Flow characteristics; the flow characteristics of a control valve is a relationship between

the flow rate through the valve as the stem travels from 0 to 100 percent. This travelling
of the stem is called in the process control industry we call it the lift of the wall
percentage lift of the wall. If the valve is fully opened; that means, it is 100 percent
travelled, like this one. That there are 2 types of flow characteristics 1 is a inherent flow
characteristics another is the install flow characteristics we will studied this in details
inherent flow characteristics we will study more details you will find.
Inherent flow characteristics; it is the relationship between the valve flow capacity and
the stem travel when that the stem travel as I told you is repeatedly it is a basically the
percentage lift of the stem, right. That is called the stem travel. Stem travel when the
differential pressure drop across the valve is held constant, right. I will repeat it is the
relationship between the valve flow capacity And the stem travel when the differential
pressure drop across the valve is held constant.
These characteristics are plotted on a curve where the horizontal axis is labelled in
percentage movement of the stem percent horizontal axis stem and the valve axis. And
the vertical axis is labelled as the percentage flow as shown in figure 3. Percentage label
or lift as I told you as shown in figure 3.There are basically 3 types of inherent flow
characteristics we will find there are 3 types of basic inherent flow characteristics these
are equal percentage linear and the quick opening. We will show we will see this in
details.
Linear flow characteristics, here the flow rate is directly proportional to the amount of
valve stem plug movement. There is no there is a direct I mean relationship it is a linear.
In other words change of flow rate is constant with respect to valve plug movement. It is
the linear curve there is nothing much to say. Valve with linear characteristics are often
specified for liquid level and for flow control application requiring constant gain. If you
need a constant gain that type of things in that type of situations I need this type of linear
characteristics.

We see here, that is you see it is equal percentage valve equal percentage characteristics.
This is a linear characteristics gain is same in all places and where is this is a quick
opening characteristics. Here is a percentage percent of movement lift of the plug and
there is a percent of maximum flow it is a 100 percent flow this is 100 percent lift
hundred percent movement of the plug. So, obviously, when the valve will be fully
opened, the stem is fully opened. So, the there should be hundred percent flow of the lift
through the valve. So, this is the inherent flow characteristics valve characteristics of the
control valve.
Equal percentage flow characteristics, we have already discussed the linear

characteristics let us look at the equal percentage characteristics. For equal increment of
the valve plug travel for the change in flow rate may be expressed as a constant
percentage of the flow rate at the time of change. Gain is not same, but it is equal
increment valve plug travel valve plug travel. The change in the flow rate may be
expressed as a constant percent of the flow rate at the time of change. The change in flow
rate observed regarding the travel will be relatively small when the valve plug is near its
seat, and relatively high when the valve plug is wide open. So, near the seat it is low and
when it is valve plug relatively high when the valve plug is wide open.
Therefore a valve with an inherent equal percentage flow characteristic provides precise
throttling control through the lower portion of the travel range and rapidly increasing
capacity as the valve plug nears the wide open position, clear? These valves are used in
pressure control applications. Since, a large amount of the pressure drop is normally
absorbed by the system itself with only a relatively small amount is available at the
control valve. This is applications because there is a various kinds of applications as I
told you for the pressure control. So, this type of equal percentage valve will be more
acceptable. You can see the characteristics how the characteristics looks like as I told
you.
If I take a blank page it will look like, this is our stem movement or stem lift. This is our
flow percentage flow percentage lift. So, characteristics will be like this 100 percent, this
is 100 percent of the lift 100 percent say characteristics of the lift of the stem. So, that is
the reason we call it equal percentage. Percentage will be equal at that point of the stem
lift, clear. These valves are used in pressure control applications. Since, a large amount
of pressure drop is normally absorbed by the system itself with only a relatively small
amount is available at the control valve.

These valves are also used in applications where highly varying pressure drop condition
can be expected. So, different applications are there where we are using this type of
different types of controller. Equal percentage is the most common valve characteristics.
Since, in most physical system the inlet pressure decreases as the rate of flow increases
as the rate of flow increases inlet pressure decreases. So, that will be it will be suitable
for that type of applications.
Now, third one is a quick third one of inherent valve characteristics is a quick opening
flow characteristics. What is this? Let us look at. Here there is a maximum change in
flow rate when the plug is near the seat. And total and slowly it will get saturated, clear.
However it is very simple you see here.
In the case of quick opening I mean percentage lift of the stem and there is a percentage
flow is 100 percent this is 100 percent, right. So, at the initially there is a large change
slowly it will saturate like this one. For a small change here, if small change of the lift
there is a large change of flow slowly it will saturate in many applications we need this
type of valve characteristics. So, please note that we have 3 types of inherent valve
characteristics one is the linear characteristics in some applications we need it if the gain
is equal. Whereas, equal percentage of particular point of lift the percentage the
percentage of the valve opening will be percentage of the flow will be I mean equal.
And in the case of quick opening that the initially it is when the valve is very when the
seat seat is very near to the plug. Plug is very near to the seat you will find there is a
large flow and whenever the plug is coming out going far from the seat as it move as we
move the stem upward. So, there will be a little change of the flow, even though there is
a change in the stem position. That is basically quick opening characteristics. Here there
is a maximum change in flow rate when the plug is near the seat. The curve is basically
linear, through the first 45 percent of the valve plug travel. Then it almost saturates to
indicate that the little increase of flow rate occurs as the travel approaches the wide open
position. All is very nicely drawn in figure three you can look at.
Control valves with quick opening flow characteristics are mostly used for on-off
applications were significant flow rate has to be established quickly as the valve begins
to flow right in on-off characteristics we need this type of thing. The linearity which is
absorbed in the early part of the curve early part of the curve decreases sharply after the
flow area generated by the valve plug travels equals to the flow area of the port, right.
This is the actually the plus as I told you earlier that the, this is the actually shape of the
plug. Seat will remain same shape of the plug will give us this different flow
characteristics right flow characteristics I mean through the versus the lift of the stem.
So, all the flow characteristics are in corresponding to the lift of the stem, actually. And
the x axis we have drawn the lift of the stem is basically the movement of the plug is not
it. So, the movement of the plug we are plotting in the x axis percentage movement of
the plug. We are plotting in the x axis and percentage of the flow we are plotting in the y
axis. Now, you see this all this actually basically all this linear equal percentage and as
quick opening depends on the type of plug, how will you shape the plug though seat will
remain the same. This is the linear you see this is the linear characteristics. This is a
equal percentage and this is a quick opening, right. So, these 3 actually characteristics
this is different shape of the plug will give us a different flow characteristics inherent
valve characteristics, clear?
Mathematical derivation; let us look at the mathematical derivations. In general the flow
through a control valve for a particular fluid at a given temperature is a function of S P 1
and P 2. Let us look at the legend. Q which I am saying that the the flow at any particular
flow rate or volumetric flow is equal to the function of S P 1 P 2. What is S? Let us look
at where q is the volumetric flow rate S is the valve stem position or lift stem position
stem movement lift. These are all basically same in many process book you will find we
have to expressed as a percentage lift or basically here for the use of I mean
understanding I am writing as a stem movement, clear? So, stem movement and lift
please note these are 2 same thing. Then P 1 is a upstream pressure of the valve, because
liquid is flowing through the restriction there will be a pressure drop across the valve
please note there is a sufficient I mean pressure drop across the valve. So, in the P 1 and
P 2 P 1 is the upstream pressure and P 2 is the downstream pressure. That means that in
pressure static pressure of the liquid and the upstream at the inlet. And is, at the static
pressure of the liquid or fluid at the outlet of the control valve, right these are we are
designating at P 1 and P 2 respectively.
For inherent flow characteristics P 1 and P 2 are constant. Hence q is a function of S

only. If the pressure drop is constant, say I can say this is constant. So, q the volumetric
flow rate will depend on S; that means, q is a function of S. What is S? S is the valve
stem movement or movement of the plug, clear? Now, I suppose that if I introduce a
factor capital F equal to q upon q max q q max and x equal to S by S max. So, what is
about this? Where q max is the maximum flow and q is a flow at any instant of time.
So, this n is variable depends if it is variable depending on the I mean position of the
stem or particular liquid I mean amount of liquid is flowing through that. S max is the
maximum lift of the valve stem and x is a fraction of that. So, ratio if you take the ratio.
So, it will give you the percentage of the flow right percentage of the lift x is the
percentage of lift. And F is actually the percentage of flow right when q equal to q max F
will be 100 percent and x when S equal to S max x will be 100 percent. That means,
valve has totally open valve is totally open, x is the fraction of the maximum lift and F is
the fraction of the maximum flow, clear?
Now, sensitivity we define as the fractional change in flow upon fractional change in
stem position. Fractional change in flow upon the fractional change in stem position will
give you, the dF by dx derivative of F with respect to x. For linear valve dF by dx equal
to alpha, dF by dx equal to alpha for linear valve equation number 3. Integrating equation
3 within the limits F equal to 0 at x equal to 0 we are assuming that that at at 0. That
means, when the valve is totally shut off there is no flow and F equal to 1 at x equal to 1.
F will be maximum flow when x x maximum value of x can be only the 1, is not it?
Because when the totally valve is moved up the stem is moved up. So, that is 1 fully
open 100 percent open at that time the flow is 100 percent. So, of F is; obviously, one
right. With this limit I can write integral dF 1 integrations of dF derivative of F
integration from 1 to 0. Because F can vary dF can vary only from 1 to 0 into alpha in to
dx where the limit is 1 to 0 again.
Integrating these equations and putting the limits and knowing that x equal to 1 excuse
me and F equal to 1 we get alpha equal to 1, right. Putting alpha 1 in equation 3 we get F
equal to x. So, this is the linear valve right. So, this is the relation when the I mean will
make like this 1. So, the F equal to x; that means, for any position, whenever I am getting
some position. So, F by x is always a constant please note for any position the amount of
flow if you divide it. So, the gain will be constant. So, it is the linear valve.

Now, for the equal percentage valve it is slightly different we are defining as d derivative
of F with respect to x. That means, derivative of flow with respect to fraction of the flow
with respect to fraction of the stem position will be mu in to t where mu is a constant
integrating equation 4 we get dF by F where varies from F to F naught and mu into dx
when x to 0, right.
The natural lock if I integrate it will be F by F naught equal to mu into x, right. Where F
naught is the flow at x equal to 0. This is very interesting you see the actually the initially
the linear valve what we have assumed that is not exactly true. A control valve will never
I mean used for totally short of the liquid flow. There will always some liquid flow
through the control valve right. So, we are assuming that is special in the single seated
valve this always will be true or double seat valve rather I should say this is always true.
That means, x equal to 0 I mean F naught we have to assume otherwise we cannot solve
the equation right; that means, at x equal to 0. That means, when the valve is at the
minimum position there is still there is not a totally shut off there is a non 0 value of F
which is F naught. It is true for most of the practically practically in all the cases of the
valve especially double seated valve otherwise we cannot calculate. Equation 5 is the
reason why equal percentage valve is also called the logarithmic valve. The basis for
calling the above type of valve equal percentage is explained below. Re-arranging
equation 4 in the form you see d by dF by F into mu into dx or delta F by F into mu into
delta x equation number 6.
From equation 6, it can be seen that an equal percentage change in flow delta F by F
occurs for a specified increment of change in stem position delta x regardless of where
the change in stem position occurs along the characteristics curve. It does not matter. So,
if I plot it in a similar graph paper what will happen you will find it is a linear
characteristics, right. I have drawn in the linear curve, if I drawn in a similar graph paper
it will be linear characteristic. That is the reason it is telling that it can be seen that the
equal percentage change in flow delta F by F occurs for a specified increment of change
in stem position delta x regardless of where the change in the stem position. Let us go
back and see.
You see here 6 for sorting change in the stem position, percentage change of the flow
will be constant mu. Because delta F by F by delta x will be equal to mu, which is a
constant, clear? That is the reason we are calling it equal percentage valve, clear?
Now, we are talked about the inherent valve characteristics let us come to the installed
valve characteristics or install characteristics. The relationship between the flow rate and
the travel of the plug as it is moved from the close position to fully open position is
influenced by the varying process conditions. This again the process, because so far we
have not considered the process condition, but this will also depends on the process
conditions, right.
Inherent valve gain, it is the magnitude ratio of the change in the flow through the valve
to the change in the stem position under condition of the constant pressure. It is equal to
the slope of the inherent characteristics curve at any travel point, right. Mathematically
inherent valve gain can be written as change in flow upon change in stem position or
change in lift which is the slope of the inherent characteristics curve, right.

Installed valve gain it is the magnitude ratio of the change in the flow through the valve
to the change in the valve gain relationship. That occurs when the valve is installed in a
specific system and the pressure drop is allowed to change naturally according to the
dictates of the system. Initially we have assumed the pressure drop will remain constant.
It is the function of the valve gain and is equal to the slope of the installed characteristics
curve.
Now, fail closed. So, far we have considered we have seen that the fail open I mean
valve, right. So, it is a fail closed a valve condition in which the valve closure member
moves to a closed position when the actuating energy source fails. In case of pneumatic
valve where the actuating signal is a air such condition is air to open right is called air to
open, right. If the air supply below the diaphragm is lost, then the valve will be fully
closed fail closed since the spring would push the stem in the valve.
You see here here is spring, so that what will happen you see. If I move the otherwise
what will happen you see that valve will always fully open. Sorry if I give a air here,
then what will happen this because there is a spring it will try to open, right. So, it is a
fail closed or air to open. So, I have to spring will be always closed like this 1. Now,
what I have to do I have to give a pressure. So, that it will move up clear? So, this is the
air 2 close, because due to spring actions valve will be always closed like this 1 because
spring will be always like this one.

Now, if I put the pressure what will happen? Because the spring will never be in this
position it will be always in the squeezed position right, got it like this one spring will be
like this one. So, if I give the air spring will be like this one. So, what will happen if this
is the position? Let us go back let us see if it is that position what will happen? You see
valve will be fully closed now if I put the air. So, valve will be fully open. Now, fail
open a condition in which the valve closer member moves to an open position when the
actuating energy source fails when the actuating energy source fails the condition. In
which the valve close member moves to an open position when the actuating energy
source fails. So, it is actually air to close. These are all valve which we have shown and
if the air supply above the diaphragm is lost the valve will be fail open. Since, the spring
would push the stem and the plug upwards, what how do does it look you see here. What
will happen that if the valve if the air supply fails? If the air supply air supply fails the
spring will move the valve in the upward position, right. So, it will move in this
downward conditions.
Rangeability; rangeability is the ratio of the maximum to minimum controllable flow,

valve coefficient Cv. It is the number of gallons per minute of 60 degree Fahrenheit
water. That will pass through the valve opening with a pressure drop of 1 psi. And it is
the measure of capacity of the valve and hence is also called the capacity factor of the
valve, right. Cv is equal to Q upon square root of delta p where Q will be flow rate in
gallon per minute and delta p is the pressure drop in psi.
Now, the valve discharge coefficient; it indicates the relative capacity between the
different valves. And is defined in terms of Cv of the valve and it is diameter d as Cd Cv
by d square. Let us go back I think if you have any confusions. Here what will happen?
You see the valve will be always closed like this 1 because, I am sorry, because this
spring will always try to open. if it is open, then what will happen if the movement the
stem will this seat will this valve plug will always sit on the seat there is a seat. So, valve
will always close the flow, right. Normally, if there is no air pressures spring will try to
open, clear? This spring will try to open right if there is let me take this 1 the spring will
try to open like this one.
Normally if there is no air pressure, suppose there is no air then what will happen spring
will try to open. So, what will it will pushed the stem downwards, so that the stem. So,
these plug will totally sit totally on the seat. So, there is no flow of fluid air or liquid
through the this region. Now, if I put the air then what will happen? So, this stem will
move up, if the stem this stem will move up, because diaphragm at that time will be go
up. So, the stem will move up, so the liquid will start to flow through this 1. So, this type
of thing is called the air to open. And next whatever I have shown is called air to close
that we have seen that is if it is normally it is spring will always try to move expand it is
like this one. Spring will always try to move in this direction normally when there is no
air.
So, it will open it, because if there is fixed here if it is we have a plug here. So, it will
move upwards right if the plug it cannot move this direction there is a diaphragm here.
So, it will move upwards now, if I want to show the valve is fully open now once we
have to give a pneumatic pressure. What will happen? This spring will now start to
squeeze the diaphragm is there the spring will start to squeeze and it will totally choke
the flow, right. So, this type is called air to close clear let us go back again. Valve
discharge coefficient; it indicates that the relative capacity between the different valves
and is defined in terms of Cv of the valve and its diameter d as Cd equal to Cv upon d
square.
Valve sizing; instrumentation society of America ISA put forward standard standardized
guideline for control valve sizing. The basic equation used for control valve sizing is q
equal to Cv route over p 1 under the square p 1 minus p 2 upon G under the square route
this is equation number 7.
Where Q is the volumetric flow rate, Cv is the valve sizing coefficient determined by
testing, P 1 is the upstream pressure, P 2 is the downstream pressure and G is the liquid
specific gravity, clear?
Valve sizing procedure for normal liquid flow you can take the simplest case of liquid
flow involves the basic equation. Re-arranging equation 7, we can get Cv equal to Q
upon under the square P 1 minus P 2 by G under the square route, right.
Based on a given flow rate and the pressure drop a required Cv value has been calculated
this is very important this is called the characteristics of the valve right. The Cv is then
compared to the Cv values for a particular valve size and value valve design and a
particular valve is selected . So, that from the series of we have a valve. So, that we have
to select a particular valve from that right first we will calculate Cv and compare to the
standard Cv of the valves available valves in the process we will choose a particular
valve whose Cv will match with the Cv which we have calculated. The required Cv
should fall in a range between 70 and 90 percent of the selected valve Cv capability not
exactly the same it is within that range. Because ultimately we will find this Cv again,
right.
For selecting the valve along with a Cv minimum and maximum flow pressure condition
should be considered that is also should be considered static pressure. The above
equation covers the condition governed by the test assumptions. Unfortunately many
applications fall outside these bounds and hence certain correction factors have to be
incorporated. If I include that these factors when incorporated change the form of the
equation as follows.

Q Cv valve discharge coefficient equal to Q upon N FPFR under the square root P 1
minus P 2 by G what is that where N is the numerical coefficient for unit conversion FP
is the correction factor FR is the function of Reynold number, and describes the degree
of turbulent flow for laminar flow FR equal to 1.
And finding N, N can be selected based on the following tables. So, you have to use
there is the table from it you can find the value of N. This is the table you see for
volumetric flow what is the value of N we have seen different values of N are there. So,
as you see this is a correction factor. So, if it is in kilo Pascal Q is in meter cube per hour
pressure is kilopascal. So, N will be 0865 and if it is meter cube per hour and pressure is
in bar. So, it is 0.86 that is unit conversion nothing else please note it is just a unit
conversion and it is 1. If it is Q is volumetric flow is gallon per minute and psi absolute
right pressure pounds per square inch. Now, this is for the volumetric flow rate and for
the mass flow rate we will choose if it is P is in kilo Pascal gamma which is the I mean
density this kg per meter cube then the weight is kg because it is a mass flow rate.
So, it is a kg per hour if the pressure is in bar and is kg per meter cube then the mass flow
rate is kg per hour at that time is 27.3 we choose and if it is in FPS unit. That means, if it
is in psi pressure is psi pressure drop is psi absolute and gamma is pound for fit cube that
time the volumetric flow mass flow rate is pound per hour. So, we are I mean taking 63.3
N 3 value is a small d and capital D this might be either in millimeter or in inch. So, in
that case if it is in inch a millimeter it is 0.021 and if it is a inch 0.890 right. So, this will
give you the conversion factor. So, which will use for this value N this will simplify our
problem solving, because some country is using some if you import some valve form US
they have some I mean APS units and from Europe we are using SI units. So, this chart
will help you to find the value of N which particular value of N will be use which
particular I mean units of Q and P will use.
For volumetric flow rate I can write Cv Q upon N 1 F p Q upon N 1 F p under the square
route P 1 minus P 2 by G is G is a specific gravity as I told you and for mass flow rate.
So, we are writing Q into F p under the square route P 1 minus P 2 into gamma gamma
in that case will be density.
Calculating Fp; Fp is the correction factor that accounts for the pressure losses due to the
pipe fittings such as reducers elbows. Because all the valves you know I mean it must be
preceded by some elbows some reducers suppose the valve sizing is not matching with
the pipe diameter. So, you have to use a reducers which are attached directly to the
control valve these are all will be control attached to the control valve, because not
necessarily you will get the exactly the pipe diameter and the inlet diameter of the valve
same. So, I have to use some reducers if no such connections are made then Fp is equal
to 1 if this no connections are made if there is no such reducers elbows. But; obviously,
Fp will never will be 1 we use always some reducers some elbows something like that.
Fp can be determined experimentally by using the specified valve in actual test. So, if the
line test or the length calibrations we have to do alternatively Fp can be determined
mathematically using the following equations, what is this equations?
Fp equal to 1 plus summation small k N 3 into Cv by d square whole square under the
square route to the power minus half right. So, it is 1 by square route of this 1 right. N 3
is the numerical constant found in equation constant table. We have already seen that N
what is N 3 either in millimeter I mean it depends what type of d small d and D. You are
using either in the millimeter or inch accordingly you choose the value of N 3. D is the
assumed nominal valve size it is always less than capital D which is the Cv is the valve
sizing coefficient at 100 percent travel for the assumed valve size right. That means,
valve is fully opened Cv whenever you are talking about. Please note it is the valve is
fully opened that is stem is totally moved there is a full flow of liquid through the valve.
Summation k is the algebraic sum of the velocity heat loss coefficients of all the fittings
that are attached to the control valve.
Small k equal to K 1 plus K 2 plus KB1 minus KB 2 this is equation number 13 what is
that? Where K 1 is the resistance coefficient of upstream fittings; K 2 is the resistance
coefficient of downstream fittings. KB 1 is inlet Bernoulli’s coefficient and KB 2 outlet
Bernoulli’s efficient. So, this will may give summation k, right.
Now, the Bernoulli’s coefficient KB 1 and KB 2 are used only when the diameter of the
piping approaching the valve is different from the diameter of the pipe leaving the valve.
If the 2 valve diameter is same pipe diameter diameter of the pipe I mean same I mean at
the inlet and the outlet of the same I do not have to use then KB 1 will be equal to KB 2.
Otherwise it will be different, but most of the process cases you will find this will be 0
because KB 1 minus will be equal to KB 2. So, KB 1 minus KB 2 will be 0. So, this 2
terms and the last terms can be calculated whereby KB 1 or KB 2 equal to 1 minus small
d by D to the power 4, right. Where d is the nominal valve sizing and D is the internal
diameters of the piping, right diameter of the piping; obviously, D is always greater than
the small d.
If inlet and outlet pipings are of equal size then we can say inlet and outlet piping to the
valve or to and from the valve then KB 1 will be KB 2 and they are dropped in such a
situations. So, for inlet reducers I can write K 1 equal to 0.51 minus d square upon
capital D square to the power whole square. And for outlet reducers I can write K 2 equal
to 1 1 minus small d square by capital D square to the power whole square and it is
fifteen point b equations.
For a valve installed between identical reducers I can write K 1 plus K 2 equal to 1.5
multiplied by 1 minus d square upon small D square by capital square to the power
whole square this is equation number 16.
For chocked flow you see that actually in the process we will find there is a chocked
flow. That means, even though I am expecting that the flow will be always I mean if I
use a delta p if increase a delta p the flow through the fluid and gallon per meter.
Whatever it may be I mean the litre per hour whatever the link is it will increase. So, it
will saturate. So, that even though increase in the delta p differential pressure will not
increase the flow through the pipe. So, this reasons we are calling it chocked flow right is
delta p allowable and actual delta p you see to the predicted using the actual delta p this
is a chocked flow we are talking about.
A plot of basic equations of figure 7 gives that the relationship holds this is figure seven
a plot of basic equations figure 7 gives that the relationship hold only for a limited range.
For the limited range this will happen otherwise it will not at the pressure differential as
the pressure differential is increased a point is reached where the realized flow increase
is less than the expected. That means, there is saturation or there is value sort
characteristics and finally, it saturates to a particular value right. So, as we increase
unlike the when stem movement when if you stem movement if you make the change of
stem do not confuse with the stem movement x axis. Now, we are plotting please go
back x axis we are plotting delta p differential pressure.
We are increasing differential pressure in this direction we are increasing the differential
pressure increasing the differential pressure in this directions. We are now in the case the
valve characteristics we have seen I mean the valve this seat I mean your plug
characteristics. We have seen as we increase the stem movement as we increase the lift
of the valve. Obviously, the flow is increased it saturate little bits in the case of quick
opening, but in the other case linear and other case equal percent is always increasing
right. But that is not the situations here when you are using having a chocked flow. That
means, if you increase delta p not necessary it will wait it will saturate after some time.
As the pressure differential is increased a point is reached when the where the realized
flow increase less than the expected.
And finally, it saturates to a particular value. Any increase in delta p after that has no
effect on the volumetric flow rate even if you increase it I mean that will not affect the
flow rate of the valve unlike. As I told you several times unlike the stem movement
whenever you have a stem movement there is always a change of flow right this is
necessary also. This phenomenon is known as the chocked flow this is called the
chocked flow of the valve. That means, if I go which ever has a chocked flow this
portion is called we are calling a chocked flow. You see this is the choked flow region
right even though I am increasing this flow is not increasing it is getting saturated clear?
So, this phenomena is known as the chocked flow of the fluid in a control valve clear?
In such cases maximum flow rate and the pressure drop conditions are very important
what is the pressure drop conditions and the maximum flow? These 2 we have think on
these are very important in this particular cases clear? We have to think of this thing do
not confuse the 2 from this let leave them. So, maximum flow rate q max equal to N 1
FL CV and under the square route P 1 minus delta F Ff Pv upon this is equation number
17 where FL is the rated liquid pressure recovery factor which is actually. Because you
know the liquid will always the pressure will be recovered recovery factor and Ff is the
liquid critical pressure ratio factor and P 1 is the upstream absolute static pressure. This
is very important up upstream pressures absolute static pressures I mean it here you see
we are not considering the deemed differential pressure. PV is the absolute vapor
pressure of the liquid at the inlet temperatures this will make the maximum flow rate q
max.
So, maximum allowable pressure drop which is to be calculated delta P max equal to Fl
square which is P 1 minus Ff into Pv this is equation number 18.
Gas and steam sizing the procedure for sizing when the process fluid is liquid air gas are
closely parallel. These are almost same and is not much of difference the only additional
steps involve the correction of the physical properties of the particular gas and pressure
ratio factors which determine the degree of the compressions and predicted chocked
flow. That is I mean these are the only factors which is to be included which I have to
think of otherwise not. The general form of the sizing equations for the compressible
fluid is.

You see Cv equal to Q upon NFp P 1 Y route over X upon GT 1 into Z equation number
19 where the legends are x is the ratio of the pressure drop to upstream absolute static
pressure ratio of the pressure drop to upstream absolute static pressure. That is the reason
we are interested in P 1 the previous equation number 18 also we have seen that there is
a P 1 is important not the P 2. Y is the expansion factor ratio of the flow coefficient for a
gas to that for a liquid at the same Reynold number. Z is the compressibility factor and T
1 is the upstream fluid temperature in kelvin. So, all these I mean things all the legends
are defined now every time you see that a Cv is the most important thing in the valve
parameter there is no other, because if you define Cv or if you find Cv all other
parameters will be automatically fixed. There are some parameters like small d capital D.
So, first in the while you are choosing the valve first I have to calculate the, what are the
desired Cv you want for a particular applications once you find a Cv? So, from the
selected list of the valve, so you try to choose Cv which will match very close you may
not get the exact match. It will be 70 to 90 percent of the value of the Cv which you have
seen that will be the desired range of the control valve. So, that particular control valve is
to be chosen and is to be installed in the process. Now, with this I we will cover this in
industrial I mean this control valve we will continue this control valve in the lesson
number 39. Also we will go further details and we will solve some problems in the
lesson thirty-nine. So, at this with this I come to the end of the lesson 38 of industrial
instrumentation
Control Valve-II; Welcome to the lesson 39 of industrial instrumentation. In this lesson

we will continue with the control valve control valve we have started in lesson number
38. So, in the lesson 39 also we will continue we will go in the deep details of control
valves, because as I told you earlier that the control valves actually means the final
control element in any industrial process they are actually the whatever the measurement
techniques we have seen or we have studied. So, far like temperature pressure flow pH
humidity so many so on and so forth. We will see that ultimately is to all these things are
basically and these measurements are necessary to control the temperature control the
flow control the process parameter. So, basically I mean the either you can make the
heater power control, but most of the cases will find that we have to control the flow
whether it is a gas fluid it does not matter. So, we will control the flow. So, by
controlling the flow I can control the temperature I can control the flow I can control the
level and so on. So, we will continue with the control valves. So, this is a control valve 2
of lesson 39. Let us look at the contents control valve 2.
Contents are types of control valves we have seen that the we have last time in 38 we
have seen that we have considered there the equal percentage valve and quick opening
linear. But those are not the types of control valve those are basically the characteristics
of the control valve I mean flow characteristics versus the stem movement. I mean if you
vary the stem movement how the flow characteristics varies accordingly with the equal
percentage linear or quick opening. But the overall the different types of control valves
are available in industry. So, we will study those in details in this particular lesson types
of control valve.
So, this is all about I am talking about this I mean valve positions and all these things
you see here this arms attached to stem to sense the valve position. And this is arm it is
sense the stem of the valve position this will come to parallel and these are the valve
position and signal from the pc will come, right.
As shown in figure 11 a typical valve positioner has the following features what are the
features of the valve positions? Let us look at a set point that calls for a desired stem
position from the primary controller Pc. It is coming from the primary controller a
measurement in the form of stem position x. So, this 2 will be I mean this 2 will be
compared this Pc will be compared with this; obviously, Pc will be compared with x in a
valve position. Accordingly it will generate a control signal Pv which will go to the
diaphragm of the control valve.
A pneumatic output in the form of a pressure to the valve top Pv will go this will actually
give the proper position of the control valve. So, this is all about the positions of the
control valves and the control valves I mean and most of the you see the most of the
figures. Actually we have taken from the the Fisher’s control they are the largest
manufacture of the control valve this is available in the web sites. There is no problem
those who want to look at the details of the figures. They can go to the, this particular
site and you get those figures right. And with this I come to the end of the lesson 39 of
the industrial instrumentation.
Prof. A. Barua
Lecture - 39
Control Valve - II
Welcome to the lesson 39 of industrial instrumentation. In this lesson, we will continue

with the control valve. So, control valve we have started in lesson number 38. So, in the
lesson 39 also we will continue we will go in the deep details of a control valves,
because as I told you earlier that the control valves actually means the final control
element in any industrial process. They are actually, whatever the measurements
techniques we have seen or we have studied so far like temperature pressure flow pH
humidity so many so on and so forth.
We will see that ultimately is to all these things are basically and this measurement is
necessary to control the temperature control the flow control the process parameter. So,
basically I mean the either you can make the heater power control, but most of the cases
we will find that we have to control a flow whether it is a gas fluid it does not matter. So,
we will control the flow. So, by controlling the flow I can control the temperature by
controlling the flow I can control the level and so on. So, we will continue with the
control valves. So, this is the control valve 2 of lesson 39. Let us look at the contents
control valve 2.
Contents are types of control valves. We have seen that we have we have last time in 38.
We have seen that we have considered there the equal percentage valve and quick
opening linear, but those are not the types of control valve. Those are basically the
characteristics of the control valve I mean flow characteristics. But there was versus the
stem movement. I mean if you vary the stem movement how your flow characteristic
varies? Accordingly with the equal percentage linear or quick opening, but the overall
the different types of control valves are available in industry. So, we will study those in
details in this particular lesson.
Types of control valve that is the contents then control valve selection how you will
select the control valve? This is the very important criterions how we choose the control
valve, because you will find the varieties of the control valves in the process according to
need we have to select the particular control valves. Then the cavitation is a problem you
will find in industrial process that it is you must prevent. We have to prevent the
cavitation otherwise that the erosion of the control valve will start and is very common in
a petrochemical industries or hydrocarbon industries.
Where you will find that whenever there is a pressure drop, because always you will find
and in this suppose in the, for an example that LPG plant when the liquid petroleum gas
is flowing through a pipe. So, there is a mixture of the liquids and the vapour gas,
because the now, when it is passing through a control valve there is a differential
pressure. So, upstream pressure is slightly higher than the downstream pressure. Now,
whenever this the liquid will flow through this one. In the downstream what will happen
due to the low pressure there is a large vaporization of the liquids into vapour.
And those will I mean start to flow in a high speed, because the pressure is there. So, all
these things will make the problem. We will study the details of the cavitations in the
control valves. There is a formation of the bubbles then the collapse of the bubbles all
these things will create lot of erosions in the controller. And usually there is no way that
you can you have to make either routine maintenance or replacements of the control
valve or by any means of the others by some means of other you have to stop this
cavitations. One of the good example, you will find in the process industry that an LPG
plant you will find that suddenly a control valve stars to what happens?
So, you see that valves starts to sweating in the outside. It will you can see that the
control valve outside is moist you can immediately say that the cavitation is started. That
means, the liquid is flowing with the high velocity on a downstream I mean the gas is
flowing. So, there is a fall of temperature that is there is an outside air is condensed.
Outside I mean your humidity is getting condensed form and it is depositing on the
outside body of the valve. So, this is one very rough estimate or rough I mean indications
of that the cavitation will start within the control valve.
Then we will start we will study the control valve accessories. There are various
accessories, but we will find that the valve position is one most important thing you must
learnt. Because valve positions will give you we will eliminate the hysteresis of the valve
of the stem of the valve. As well as it will make the make the precise control of the
position, which is not there? So; that means, hysteresis can be completely removed if you
use a valve position it is additional control excuse me, this is additional control it is
basically a proportional control. We will find it is working as a cascade control in a with
the main control look. So, it will make the better positions of the stem of the control
valve.
Now, at the end of the lesson the viewer will now, the difference between different types
of control valves like sliding stem valve, globe valves and butterfly valves etcetera and
their relative advantages and their applications. So, with this will know because, these
are the different types of control valves available. Globe valves, butterfly valves. I mean
this will study in details and relative advantages and disadvantages and typical
applications in the process industry. How do parameters like temperature pressure,
rangeability etcetera effect the control valve selection?
This; obviously, come in the selection of the control valve how the temperature pressure
of this rangeability? Rangeability means total range or the capacity. Sometimes we call
capacity of the control valves, sometimes we call rangeability these are same basically
effect the control valve selection. Pressure profile of a control valve, that means, how
what is the pressure profile? How does it look? The pressure profile as you go we have
seen that in the case of orifice there is a pressure profile. Similarly what type of pressure
profile exist inside the control valve if you if I proceed from the upstream.
That means, from inlet point to the outlet, how the pressure inside the control valve looks
like? That is actually we will also look at. Advantages of the valve positioners as I told
you this is the necessary. So, what is the valve position is the necessaries for the control
valve? Actually it is installed fixed to the control valve itself; that means, that pneumatic
signal control signal will not come directly to the control valve it will come first come to
the wall positioners. Wall positioners will make some additional control signal which
will we to correct the according to the positions, it will make some signal that will go the
diaphragm of the control valve. The diaphragm all this things we have studied in details
in the case of the lesson 38.
Types of control valve; let us look, at control valve here means any power operated valve
whether used for continues or on-off control. We have control valve usually, but mostly
we are talking of the pneumatic control valve. Even though we are saying the control
valve means any power operated valve because I need pneumatic power from a control. I
need a electric power in the case of solenoid valves, solenoid valves usually it is on-off
type of valves that we are not discussing. We are discussing basically control valves
control I mean we are basically discussing the pneumatic control valves. Excluding
varieties such as motorized gate valves plug valves pinch valves etcetera control valves
can be broadly classified into 3 main categories. What are this 3 main categories?
Let us look at a sliding stems valves, we have seen that the sliding stem; that means,
what is this? We have seen that.
I have a diaphragm. So, this valve sits on a seat, this inlet outlet of pressure is coming.
And there is a spring action there is a plug here or packing this is a packing. So, if the
pressure comes, so the stem will move this stem will move up and down. And it will
control it will control the position of the seat in the provision of the position of the plug
in the seat. So, that thereby is controlling the flow of the liquids, so it will look like not
this one. It will look like this right this one let me take a new page. So, it will look like a.
It has a have a stem here, so stem is like this one. So, liquid is flowing through this one
and liquid is flowing out because there is an inlet here. We should erase this part, so it
will look like this. So, liquid is flowing through this one, so there is a path and liquid is
flowing out through this one. So, it will look like this one; that means, it is like this liquid
is flowing this it is stopped here. And it will look like this one actually now, what will
happen? You see when the liquid is flowing through this one. So, liquid will come
through this one and it will flow through this directions. If the valve moves upwards, so
there is a restriction here. So, the flow will go like this one, so it will come like this it
will go like this one; that means, it will come like this it will flow and through this one.
So, position of the stem there is a spring here, so position of this is already we have
discussed in details in the lesson 38. So, you can see the better pictures there, because
these are pictures a little complicated we can see very nicely in lesson 38 anyway. Let us
come to the sliding stem valves or globe valves or angle configuration, etcetera. This is
the sliding stem valves globe valves angle configuration etcetera all comes under the
under the sliding stem valves. Then rotary valves these are ball and butterfly valves;
butterfly valves are used in many processes where I need a quick shut off like nuclear
industries even though butterflies has leakage. But with the Teflon coating and all this
Teflon liner inside the pipe you can make it Teflon liners in on the stem itself.
So, we can make a tight shut off otherwise I mean butterfly valves is used, because it
needs action is very quick compare to the other valve like the step sliding stem valve. If
you compare sliding stem valves I mean very roughly it will save the time constant. It is
much faster response butterfly valve is much less. It is sorry I mean not less I mean
saying that very fast. That means, the time of response will be very less in this case in the
case of butterfly valve in the case of butterfly valve or rotary valve. Then you have a
eccentric valve also we will study all this. These are the three different categories of the
control valve sliding stem valves, rotary valves and eccentric valves. These are the 3, I
mean different categories of the control valves then sliding stem valves. Let us look at
the details.
They are the most widely used control valves. As I told you sliding stem valves, most of
the time we have seen that when we were giving examples of the control valves we are
going to the sliding stem valve. Typical sliding stem valves are glove valves angle and Y
pattern these are the 3 different patterns we will see. We will see the sliding globe angle
and Y pattern these are the available in various sizes and various designs, very widely
used control valve.
Then ruggedness dependability, precise throttling, capabilities throttling means I mean
precise passing of the liquid. That means, if I say that the for this particular position this
much of liquid will flow or this much of liquid will this will be the volumetric flow of
the liquid that can be precisely achieved in the case of this type of valves sliding stem
valve. High overall performance and general sturdiness are some of the qualities of the
sliding stem valves. It is very rugged very sturdy, so it is widely used in industry.
They are more suitable for extreme applications, which involve high pressures, high
temperature, high noise and cavitational controls. Cavitations are we are telling it
repeatedly cavitations we will study details after few slides. You will find that it is very
important for the control valves you must know it very details. So, they are more suitable
for extreme applications which involve high pressure high temperature, high noise and
cavitations. They can also be differential differentiated as a single port and double port.
All that you have seen there are two types of valves we can have double port valves we
have a single port valves. In the single port valves a shut off is tight. Whereas, in the case
of double port valves shut off is not tight. But in the most of the industrial process as you
know this never we have I mean tight shutoff always it is within a range of liquid inflow.
Types of sliding stem valve, single port valve bodies: It has a simple construction and is
the most commonly valve body style. We have seen that in the case of 38, whatever the
examples we have given which is which a single port valve. They are available in the
various types like globe angle, etcetera.
Generally single port valves are specified for applications with stringent shut off
requirements require the shut off is very tight. I need this type of valve you may say that
why should use then butterfly valve? As I told you the response time of the butterfly
valve is an I mean very less; that means, it is a very quick action. It will take whereas, in
the case of sliding valve single port valve even though as it is single I mean sliding stem
it will takes more time.
But the off I mean if the shut off is very tight though if a need a very I mean tight shut
off I have to use a single port valve. They use the metal to metal seating surface and they
can handle most service requirements. So, metal to metal surface; that means seat is
metal as well as plug is a metal. So, handle more service requirement; that means,
corrosive requirement high temperature all those things can be withstand. Though now
nowadays Teflon can also withstand high temperature, but not high as; obviously, the
metals like steel.
Since high-pressure feed is normally loading the entire of the port the unbalance force
created must be considered in selecting actuators for single port control valve bodies.
The high pressure if the liquid is coming as a high pressure, so quite. Obviously, it will
occupy the entire port of the inlet port of the valve. Entire area of the inlet port and
unbalanced ports created must be considered in selecting the actuators for single port
control valve bodies. The different styles of single port valves and their descriptions are
as follows. Single port globe valves, it is the most common form of sliding stem single
port valve.
It has a linear motion; it has a globe shaped body. The inside shape of the body I mean
valve is a globe shape that is the reasons we are calling it globe. They are widely used in
process control application particularly in the sizes of 2.5 centimeter to 10 centimeter.
That is 1 inch to 4 inch size you will get, because you see that we are converting in SI
units or in a in a CGS unit. But most of the valves I mean valve the details you will find
like the Fischer’s and all those who are the making of control valves in over the years.
You will find they are giving the specific inches since these are American valves usually
I mean makers the American manufacturers they still consider in those LPS system, but
still in a we are telling it in CGS or SI units. A typical single port globe valve is shown in
figure 1 you can see here.
A single port globe valve you can see here. It is the inlet body inlet it is coming here
there is a seat, so here is a seat. So, if the seat closes the liquid cannot flow. This is the
stem of the valve there is a spring you can see. There is here you see where you can see
here that is there is a packing. So, spring is there, so diaphragm is not shown here. This is
a typical single port globe valve.
Single port angle valve angle valves are nearly always single port. So, you will see that it
is a angle the liquids are it is like an elbow. Let us see they are commonly used in boiler
feed water and heater drained service. And in piping schemes where proper utilization of
space is very important, because not always we can use an I mean linear I mean there is a
linear pipe when I need a bending. So, that time that time you can use a single port angle
valve. They can also serve as an elbow it is a working as a elbow. So, instead of using an
elbow I can use an I mean angel valve. Which will make the flow control as well as it
will make the, it will save an additional elbow in the process pipeline. The valve shown
has caged style caged style construction.
Changes like expanding outlet connections, restricting trim etcetera can be done to
reduce the erosion. Damage erosion is a common problem you will find in the control
valves. We will see the cavitations of the cavitation are there; obviously the erosion will
be there.
This is the single port angle valve, you can see here. It is inlet port and this is a outlet
port. This is the stem of the control valve you can see here this is the stem and this is a
we have a spring here. This is a inlet and this is the seat this is the our plug and this is the
seat. So, if it is seat tight, so the liquid cannot flow, so it is like an elbow. So, it will save
1 elbow in the process. That means, I if I need proper utilizations of the phase is very
important. So, I can use this type of angle valve. It is single port angle valve.

Single port bar stock valve they can they are small range from few centimeters to 7.5
centimeters. And they can be made a machined from any metallic bar stock material and
from some plastics. From any bar I can metal bar I can we can machined it and it is from
some plastics also you can make it. Their flow capabilities are generally lower than those
of the general purpose valves. It is not for the high capacities in as it case of the globe
valves not convention globe valves, but for the low capacities.
These valves are used more often when there are special corrosion considerations in
chemical industries. In chemical industries you see though you are saying stainless steel
we can use, but stainless steel machining stainless steel is very difficult as you know. So
in that type of cases where we need a valve; where it will not react with the fluid, which
is flowing through the pipe. So, because any metals you know sometime some metals
reacts with some chemicals. So, those types of cases chemicals are more inert I mean I
mean these plastics are more inert in that sense. So, we can use that these type of valves.
A typical bar stock valves you can see here is shown in figure. This is a bar stock valve
the same thing stem spring is there and you can see this is a bar. So, we are calling it a
bar stock you see here. It is coming through here and this going through this one. This
inlet this is outlet.
Double port valve bodies: Double port valve already we have seen in that lesson 38. So,
we will we will look at details. The dynamic force on the plug is more balanced as flow
opens one port and closes at the other. So, dynamic forces 1 on the plug is more balanced
as flow opens 1 port and close the other. So, there are 2 valves, so obviously, if in1 port
liquid will flow in other port liquid will stop. Reduces dynamic forces acting on the plug
might allow choosing a smaller actuator then would be necessary for single port valve
body with similar capacity. So, I need a lesser powerful actuator that is the reason single
double port valve is referred. Body’s are generally furnished with large sizes 10
centimeter larger and more capacities also more for this type of valves.
Many double port double port bodies reverses, so the valve plug can be installed as a fail
close of fail open modes. We have seen that the valve can be used as a fail open air to
open or air to close. Since double port bodies reverse. So, I can find the reverse. So, that
I can use it for that I mean fail close or fail open modes. This is the great advantage of
the double port body, which we do not have in the case of single port valve. So, same
valve can be used whereas, in the case of single port valve we have seen that we have to
make that total. We have to change the total valve itself, which is not necessary in the
case of double port valve we can use it either in an air to close or air to open. Figure 4
shows a typical double port valve with a globe style valve body this is a globe style valve
body and this is a double port.
You see here this is a double port valve. So, it is you see here there are 2 ports. And 2
seats liquid can flow through this one or liquid can flow through this one. So, always
when different position liquid will flow either through this one or through this one. So, it
is a single port double port globe style valve body.
Rotary valves types of rotary valves are ball valves. Ball valves show high range of
volumetric flow and the low susceptibility to wear by erosion streams. This is important
this is also true for the butterfly valve we will see that. Hence generally used for high
pressure drop throttling and on off applications. If we need a high pressure drop, because
whenever there is a high pressure there is a problem of the cavitations will be very severe
in the case of high pressure drop. Because whenever it is comes to the outlet point there
will low vapour pressure. I mean pressure there is a pressure drop and there is a
formations of the vapour and it will start to corrode the, or I mean erosion will start at the
outlet ring. So, the seat and all those things will be damaged. Hence generally used for
high pressure drop throttling and on-off applications
However, they have a sluggish response in the first 20 percent of the ball travel making
them unsuitable for throttling applications. Continuous throttling is not necessary I mean
and either you can stop it on-off sort of that type of situations I can use it. But the
continuous control of the liquid is not possible, because of this inherent problem that is
they have a sluggish response in the first 20 percent of the ball travel; that means the
stem travel and making them unsuitable for throttling applications. Figure 5 shows a
rotary shaft control valve with a V notch ball. The V notch produces an equal percentage
flow characteristics. This V notch will make a V notch an equal percentage flow
characteristics.
These valves have a good rangeability, control and shutoff capabilities. The range is
higher; that means, it has a wide range this is very important. And it is a control and
shutoff capabilities are also high. These are the advantages of this type of valves and it is
also suitable for the control of the viscous fluids and slurries containing solids and fibres.
This is the very common form of I mean situations we will see in the process industries.
That means, it will always not necessarily you will find always water there is any viscous
liquids may flow. It may contain slurries and other fibres, so that type of situation globe
valve is very important. I mean valve type of valve should be important.

This is you see, this is the wall type of valves rotary shaft; control valve with a V notch
valves. You can see here this is a V notch ball valve.
Now, butterfly valves you can see this is a very common. This is a widely this is also not
very widely used, but it is used for some particular I mean industry because if it is a
suspended particles to avoid cavitations all those things butterfly is very suitable for that
type of situation. They require minimum space of for installation and a very economical
particularly because you see since it is a butterfly. So, you will have an only a valve
motors. So, we do not have stem sort of thing only the butterfly I mean that there is a
disk sort of thing which will rotates if it rotates it will allow the liquid to flow otherwise
it will stop.
That is the butter and a very economical particularly in large sizes and flow capacity or
range. So, if you need a large range that is great advantage; that means it has usually
typically butterfly valves comes with a large range which is not possible in the case of
typical globe valves. We need if you want to larger in the size of the valve will be
enormously weak, because in the valve there is a restriction the stem size you stem size
sorry the plug size and the seat size actually will control the range of the valves also. So,
those type of distinct there is no as such seat or plug in the case of butterfly valve.
That is the great advantage that is; obviously, will lead to the higher range of the control
valve. They provide high capacity with low pressure loss across the valves. They provide
high capacity with low pressure loss across the valves. In case where operating torques
can be very high butterfly valves may require large actuators. That is another problem if
we need a large torques we need a large actuators, because they have to rotate. We will
see that how it works we will show the butterfly valve. So, that I need a large actuators
otherwise it is a very good from valve.
Butterfly valves are used in nuclear power plant applications with very stringent leakage
requirements. It is the applied, because in we have a very I mean time constants of the
valve should be I mean very small. That means, it should be immediately shut-off of
immediately start to flow. That type of situation, but the problem is if you want to make
a complete shut-off butterfly valve is very difficult because there is no such plug and
seat. So, there is always some leakage, but nowadays with some special applications we
can make it we can make the leakage almost zero. The standard liner can provide good
shut-off and the erosion protections. Erosion is not very important here it is the good
shut-off is very important. So, the commonly used liner is nitrile or PTFE which is
polytetrafluoroethylene or Teflon. This can be used as a liner.
And this you can see here this is actually a butterfly valve. So, that this pneumatic signal
you can see you can see here this is actually it will rotate and will this will rotate. And it
looks like this one; that means, I have a valve.
So, pipe is there it will look like this one. Either it will come to this position or it will
completely shut-off like this one.
If it is goes like this one there is no shutoff at all. So, this; that means, you have to rotate
it. It will rotate this one to this one; that means, if the pipe is there. So, if I rotate like this
one there is a total shutoff, so it will rotate like this one it will rotate like this one.
Eccentric control valves eccentric plug have features of both the sliding stem and rotary
valves. There are the features of the rotary valves as well as sliding stem valves. But
unlike rotary valves, it utilizes a massive rigid seat design. Seat design should be very
rigid here. So, let us look at many good aspects of both rotary and stem are combined in
eccentric valve. So, both the combinations are there so; obviously, some applications this
is very stringent. There are basically 2 types of eccentric control valves.
Eccentric disk control valves; Bodies offer effective throttling control. They provide
linear flow characteristics. An eccentric mounting of disk pulls it away from seal after it
begins to open minimizing seal wear. So, that is the great advantage, because seal is
always wear. They use standard pneumatic diaphragm or piston rotary actuators. I think
this will be seat wear.

They are intended for general service applications not requiring precision throttling
control. Precision throttling control is not possible it does not matter.
So, this our extrinsic disk control valve. You see here.
Eccentric plug control valve: Valve assembly combats erosion. In a rugged body and the
trim design to handle the temperature as high as 400 degree centigrade and shut-off
pressure drops to 10,000 kilo Pascal extremely high pressures we can see here. And path
of the eccentric plug minimizes the contact with the seat ring when the opening and
reducing the seat wear and friction and prolong prolonging it is life. Forward or reverse
flow and the tight shut-off in either direction is possible. Forward or reverse flow and
tight shut-off in either direction is possible. So, this is a great advantage though it is very
uncommon in any process.
Eccentric plug control valve: you can see here

Control valve selection: General criteria for selection of control valves are the pressure
rating this is most important. Body pressure ratings are considered according to the
ANSI pressure classes. The ANSI pressure classes for steel are 150 300 and 600.
The maximum pressure rating is inversely proportional with the temperature. The
required pressure ratings are dictated by the application.
Flow capacity or the range of the I mean range of the valve that is also important. On the
basis of the flow capacity we can roughly categorize the selection as follows. Sliding
stem valves are used for lower ranges 1.25 centimeter to 7.5. I am sorry this will be in
centimeter this will be centimeter. Ball valves in intermediate capacities it is coming on
10 centimeter and to for 15 centimeter. Butterfly valves for a very large sizes 20
centimeter and above.
Special consideration like high - noise application, cavitation, etcetera require valve
designed with a special trims. Trim as I told you it is a entire assembly we are calling
about the trims.

Flow characteristics a complete dynamic analysis is required for determining the proper
flow characteristics desired. What are the typical flow characteristics I need? Equal
percentage linear or quick opening these is basically flow characteristics. However, there
are certain general guidelines, which hold true, but there are few exceptions.
You see we are seeing at the constant delta P, this is the table a selection based on the
flow characteristics, selection of the control valve, control valve pressure drop that is the
pressure drop across the control valve. Inherent characteristics constant delta P linear,
decreasing delta P with the increase load linear we will we will use linear control valve
delta P at a maximum load decreasing delta P with increasing load equal to percentage.
Increasing delta P with increasing load delta P at maximum load 200 percent of the
minimum load the linear, increasing delta P with increasing load it is quick opening.
Rangeability: it is the ratio of it is maximum and minimum control flow rates. it is the
ratio of the maximum and minimum control rates. We have seen these things while we
are doing the analysis I mean at the equal percentage valves in the lesson 38.
Exceptionally wide ranges may be required to handle wide load swings or combination
of start up to normal and maximum working condition. But usually you know
rangeability is not very important we have actually the variations of the flow capacities it
is not that wide in any process. The rangeability is obviously; there I mean it depends on
what type of fluid is flowing? You may have a very high range you may have a small
range, but the variations of the flow capacity is not very common in any process. In
general rotary valves have greater rangeability than the sliding stem valves right quite
obvious.
Temperature limits the thermal considerations include the strength or ductility of the
body material as well as the relative thermal expansion of the valve internal parts. Now,
this it does not matter much if I use a I mean I mean metal like a steel and all those
things. This ductility and all those things will not matter much, but sometimes it will
have if that liquid temperature is extremely high. I mean this is you should consider this
thing also what type of material is to be used to make the seat to make the plug. And all
those things to the inlet I mean through which the liquid is flowing that is all those things
to be considered. Temperature limits may also be imposed due to disintegration of soft
parts like seat rings, seal, piston etcetera, which are made of various elastomers plastics
etcetera. At high temperatures or loss of resiliency at low temperature, this is also
important.
Shut off leakage; this is very I mean we should consider tight shut off is very necessary
in case of high pressure valves, because leakage can cause seat damage. So, the tight shut
off is necessary in case of high pressure when the pressure is high. Obviously, if the shut
off is not tight there will be always leakage. So, since leakage can cause seat damage. At
the same time; however, we must consider we should not over specify the shut off
leakage in some application may unnecessarily increase the cost.

Pressure drop this is another important thing pressure drop across the valve. Sliding stem
valves can tolerate more pressure drop then the rotary valve. Sliding stem valves can be
tolerated more pressure drop then the rotary valves. Noise and cavitation both of which
are accompanied by high pressure drops must also be considered while selection.
End connection requirement this is another important thing. End connections desired are
important considerations, because sometimes the desired connection style is not available
and the valve style being considered. This is another important thing what type of pipe
you are installing that type of what is the end connection. That means, inlet connection
and the outlet connection to the pipe is to be considered also not always all the valves.
We have the same all types of end connections you can make it according you have to
choose the one there is a choice of valves obviously. If piping specifications calls for
welded connection only, the choice is limited to the sliding stem valves welded
connections only since rotary valves available are expensive.
Material selection generally materials like carbon steel, alloy steel and stainless steel
which have a better strength pressure retaining and corrosive properties are used.
Stainless steel I earlier also told it is not very easy to machine the stainless steel. I mean
alloy steel carbon steel; obviously, the stainless steel is also a low carbon steel basically.
The life cycle cost this not only includes the initial cost. But also the maintenance cost,
because control valve is to be maintained very regularly. Usually all control valves will
be bypassed by a globe valves always, because during the routine maintenance a plant a
24 hours when the plants is running you cannot I mean shutoff the plants. So, that the
global the actual control valve is removed, and you will put a globe valve there.
Cavitation; we have talked about the cavitation so much. What is this cavitation? Let us
look at every detail, because this is very important you must stop the cavitation.
Otherwise the erosion will start and valve will be entire calibrations whatever the typical
flow characteristics, you have desired; you may not get these things ultimately leading
the poor product qualities. So, those things are very important. So, cavitation must be
you must study the cavitation in details and you must we have to stop the cavitations
inside the wall. At the fundamental level cavitation is nothing but the growth and
collapse of cavities in a liquid growth and collapse of a cavities in a liquid. These growth
and collapses result from the fluid pressure dynamics as I told you high pressure to low
pressure.
In case of control valves a bubble is formed when the local pressure of the liquid drops to
the vapour pressure when the bubble will be formed. That means, when a local pressure
of the liquid drops to the vapour pressure, because at the outlet of the valve or the just
when it crosses the seat or the seat of the valve then the pressure will drop that there. So
if the pressure drops; obviously, the bubble will be formed, and it collapses when the
local pressure, because as it proceeds to the pipe I mean it will collapse. So, it will
collapse in the local pressure, because the pressure will be recovered once we proceed to
the and outle. The pressure drop the pressure loss will be recovered as it happens in the
case of orifice venturi same thing here also. Then that bubbles will collapse again,
because pressure is increase the bubble will collapse clear rises above the vapour
pressure.
Now, these rise of bubble and it is consequent collapse leads to the formation of cavity.
That means the formation of the bubble rise of bubble and it is collapse will lead to the
formation of cavity. Consider the mean pressure profile shown in figure nine in the next
figure we will see. The shape of the, this general curve is a consequence of the pressure
changes resulting from the velocity changes and the available energy dissipation. There
is a energy dissipation inside the pipe and the shape of this general curve is a
consequence of the pressure changes velocity changes and the dissipation.
This you can see mean pressure profile of a control valve you see here pressure is here
then of pressure is there. There is a large pressure drop and it will increase little bit at the
outlet it will come to the P 2, but inside there is a large pressure drop. So, the cavitations
will start here when the pressure drop is high when the pressure difference is high. So,
the cavitation will start here this is P 1 minus Pc this is the minimum pressures here. So,
you can see there is a large pressure drop. So, the pre-cavitation will start here right. So,
this is a flow direction; that means, I am proceeding from the inlet to the outlet of the
valve this inlet this is to the outlet of the valve clear?

Now, the mean fluid pressure decreases from the inlet valve to some minimum value and
recovers partially at the outlet. That is I told you like Orifice it is the same thing there is
the restriction is not it valve is very similar to restrictions we have seen the orifice
venture. All these things we have a restriction there we have seen in the case of venturi
the restriction. I mean the valve pressure recovery is better pressure drop is not that here
you see the pressure will be fall is a restriction again it will be recovered to some extent.
This mean the fluid pressure decreases from the inlet valve to some minimum value and
recovers partially. This is the minimum value PV what we have seen this is our
minimum value PVC. So, this is P 1. So, P 1 minus PVC is this much this is the
minimum value of the control where is the pressure.
Cavitation theoretically occurs when the minimum pressure is equal to the vapour
pressure and the outlet pressure is above the vapour pressure. Now, where it actually
happens you see there is it is what I am saying cavitation theoretically occurs when the
minimum pressure is equal to the vapour when it happens it actually will happen when
the liquid coming in through the inlet point. It is going through the going through the
plug going through the seat through the plug plug will I mean make the control and to the
outlet. So, the cavitations will start where cavitation will start just when it will leave the
leave the seat of the valve clear just when it will leave the seat of the valve the
cavitations will start, because at the as it proceeds to the as it proceeds to the outlet.
So, that is the portions actually when the cavitation will occur. Cavitation theoretically
occurs when the minimum pressure is equal to the vapour pressure minimum pressure is
equal to the vapour pressure very common in the case it is very common in the case of
LPG industries where the vapour pressure is little less . So, whenever this pressure falls
then; obviously, cavitation will start and the outlet pressure is above the pressure when
the above above the vapour pressure. What will happen to the outlet pressure? Because
pressure will be recovered that time that bubbles will be collapsed. So, the cavitation will
start.
How to overcome the cavitation, so cavitation is there you must I mean check how to
overcome the cavitation. Material selection it is sometime possible to create a cavitation
tolerant environment by selecting materials more resistant to the cavitation attack we
have to see the inside the valve. So, that it will more resistance to the cavitation that is
number 1 number 2 introducing resistance in flow path due to this the overall fluid
pressure recover is reduced.
Obviously, the pressure drop will be there is a pressure loss; obviously. So, this pressure
loss actually will lead to the more pumping cost; obviously, but the, we can avoid the
cavitation. Introducing resistance in the flow path and due to this overall fluid pressure
recovery is reduced thus avoiding to above the vapour pressure. Even if it does not stop
cavitation all together it is certainly due to it is intensity intensity of the cavitation will be
reduced.
Now, accessories of the control valve there are accessories I talked about at the initial
stage. So, let us discuss little bit of accessories. There are many accessories associated
with the control valves like valve positioners, limit switches, solenoid valve, manifold,
supply pressure, regulator, fail safe systems, pneumatic lock up system etcetera. But
most of important of these all these things are the the valve positioners. These
accessories help the control valve function better with more accuracy and also act as a
safety precautions. This is also safety is very important as a safety precautions also will
be used. Out of all these valve positioners are the most important are the out of all this
valve positioners are the most important accessories we will study the valve positioners
in details clear?
Now, valve positioners what is these valve positioners you see that is as the name
implies actually it will position the position the stem position the seat position I am sorry
position the plug on the seat of the valve more precisely much better way right. So, that
there is no stress the valve positioners is necessary due to the hysteresis we will first look
at what is the control valve hysteresis we will look at that. Need of the valve positioners;
valve hysteresis; this is mostly the friction in the packing, because we need a packing,
why because you see the inlet pressure inlet liquid is coming inside the valve which is
might be globing shape. Or whatever it may be through the through the seat it is in the
plug the space between the seat and the plug liquid is flowing through the outlet line, but
this plug actually connected to the stem.
So, the plug must this entire globe should be separated from the other part of the control
valve from the bonnet all those things. So, what will happen? We need some packing.
So, pack packing will prevent the liquid flowing out there is only one path of the inlet or
one path of the outlet if there is improper packing. Then what will happen? Liquid will
flow through the packing region that is not allowed, because the high pressure liquids.
So, the, is a very good packing. So, the packing the problem is there packing will
introduce a frictions. So, that will make the hysteresis of the control valve. The friction
in the packing and the guiding surfaces of a control valve causes a control valve to
exhibit hysteresis as shown in figure 10. The stem position increases along the lower
curve as pressure is increased, let us look at…
You see I have signal to the control valve Pv and this is the stem position or lift what we
call it something. Sometimes we call it life y axis I am placing lift and x axis the control
signals pneumatic control signals I am giving the control signals. But I am getting any
stem position suddenly there is a stem position it is increasing I am changing the signal.
Now, it is not now when it is friction overcome the friction it is going like this while I
reducing you see while I reducing the I mean this signal to valve. Because x axis is
signal to valve we are reducing the signal pneumatic signal the stem position is not
coming down it will remain for sometime the same position after that it will go. So, this
is H hysteresis of the control valve to kill this hysteresis I need a valve positioner let us
look at.
It decreases along the upper curve as pressure is decreased we have seen as the moment
air pressures signal reverses the stem position stays in the last position until the dead
band H is exceeded. This H is exceeded until unless I cross the H H is the dead band of
the pneumatic signal. After that it will start to react right both in the when it is going in
the stem is moving up and moving down might be this H is small whatever we have
shown that I mean to understand. So, that you will understand nicely, but it usually quite
small even if you do not use valve positioned
After this the pressure increases or decreases along the path shown by the arrows we
have seen where pressure is increases stem position; stem position changes this is the
pressure. If the valve is subjected to slow periodic variations in pressure a typical path
taken by the stem will be PQRS. If I now give a slightly cyclic variations of the
pneumatic signal Pv we will find that there is a typical path taken by the stem will be
PQRS clear? What is PQRS? You can see the hysteresis curve clear?
The affects of the valve hysteresis; what is the affects of this hysteresis? Let us look at
the hysteresis is a non-linear phenomenon. I cannot be expressed by a transfer function
the presence of hysteresis in the valve can cause the controlled signal to oscillate or
exhibit a limit cycle it is show a oscillate or limit cycle. This limit cycle contributes to
wear of the valve; that means, valve will be damage very quickly.
Now, to eliminate the hysteresis valve hysteresis let us look at. Hysteresis being a non-
linear phenomenon cannot be removed or reduced using controller tuning controller
tuning will not do which is non-linear phenomenon, what we can do the only way to
reduce the affect of the hysteresis is through the use of the valve positioner. We can
valve positioners you will see is a another control loop which is with the proportional
control which will control which will position the valve precisely at the position where
we need. This is the main reason of using valve positioners; this is to kill the hysteresis.
Let us look at about the valve positioners.
Valve positioners is nothing but a high gain proportional controller receive a set point
signal from the primary controller and a measurement from the valve stem position. I am
getting a set point signal from the main controller and I have a stem position if there is an
error. It will make some I mean it will make proportional control signal we will go to the
diaphragm of the control valve. So, by these I can get a precise. So, that now the stem
position also is to be measured this is most important previously we have seen that we
are not measuring the stem position. But if I install the valve position valve position is
the integrated part of a valve body it is actually rigid to the valve body.
But I need a if I need a valve positioners then what will happen? Obviously, I need a I
mean I have to measure the position of the stem if the actual position which is coming
from the control signal or controller does not match with the position which is showing
by the valve stem positioner position sensor. So; obviously, I have to take some action I
have to move up and down until unless this position I mean exactly matches with a with
the correspondence to the control signal which is coming from the controller let us look
at this. The addition of the valve positioner to control valve introduces a form of cascade
control. This is basically a cascade control we have seen the cascade control two loops in
nature.
So, we have seen the cascade control extensively use in process thus increasing the
accuracy and also speeds up the response of the valve. So, the addition of the valve
position to control the valve introduce a form of cascade control and thus increases
accuracy and also the speeds of the response of the control valve. So, the control valve
response also will be increased by this position. This is very important that is the reason
people are using the valve positioner most of the controllers. Specially if the valve is a
large in size we need a controller the only pneumatic signals may not be sufficient to. So,
I need a valve positioner there right. A typical control valve with a positioner is shown in
figure eleven you can see here.
You see this is our a control valve a control valve already you have seen in many
situations. But you have not seen the valve positioner this is the valve positioners of the
control valve you see what will happen you see here. So, this is our valve typical valve
body I mean the liquid is coming in through this flowing through this one which is going
to the outlet fine. This is the stem; this is the, our plug and this is our seat you know this
is the seat and this is spring. And you see here is a packing this packing should be very
tight in the case of control valve. Otherwise, that is a problem, because if there is a
packing why I am saying repeatedly about the packing you see what will happen if the
pressure liquid is high.
So, it will go through this one. So, that is not allowed. So, this packing should be very
tight right. Now, if I use a tight packing there is a large of frictions of the stem. Because
stems will move only the stem of this control valve will move nothing else the stem will
move the spring is squeezed. If it goes up now when the pneumatic signal removes this it
will go down right clear? Now, what will happen? You see if I if I now sorry if the
pneumatic signal is I mean removed actually the valve will go up. So, it is air to close if
the pneumatic signal this is a diaphragm. So, diaphragm give the pressure it will go like
it will squeeze and it will come otherwise what will happen? It will go like this one.
Now, what will happen? You see here there is a valve positioner. So, there is a friction
now the valve positioners is a device is a cascade controller. So, I am getting a signal
pneumatic signals of Pc from the controllers that is 3 to 15 Psi of the signals. I am getting
it is coming to the now I am measuring the valve positions also through this one stem
position the stem positions I am measuring. So, what will happen? It is getting signal and
this is also this if this is not matches then what will happen if it does not matches it will
produce some control signal it will produce some control signal. And this will this
process will continue until unless this valve position whatever it is showing when the
indicator or measurement system will exactly matches with the signal control signal Pc.
Pc is basically corresponds to some valve position stem position, right.
So, this way, so the control loops I mean this way I can use a output of the control will
come here. So, this way I can make the valve positions and make the systems very steady
I mean I can make the system. So, that it will work I mean without any error in the stem
position this is very common, because if it is a large friction. Obviously, what will
happen that there will be a some mismatch of the position now that can be. I mean
recovered that can be exactly valve can be positioned valve stem can be precisely
controlled by using the valve positioners right. So, this is all about I am talking about
these I mean valve positions and all these things you see here these arm attached to the
stem to sense the valve position. And this is arms it is sense the stem of the valve
position this will come to travel and this is the valve position signal from the Pc will
come right.
And as shown in figure eleven a typical valve positioners has the following features what
are the features of the valve positioners let us look at. A set point, that calls for a desired
stem position from the primary controller Pc, which is coming from the primary
controller, a measurement in the form of stem position x. So, these two will be I mean
these two will be compared this Pc will be compared with x. Obviously, Pc will be
compared with x in a valve position accordingly it will generate a control signal Pv
which will go to the diaphragm of the control valve.
A pneumatic output in the form of a pressure to the valve top Pv will go this will actually
give the proper position of the control valve. So, this is all about the positions of the
control valve and the control valves I mean and most of the, you see the most the figures.
Actually we have taken from the Fischer control they are the largest manufacture of the
control valve this is available in the web sites. So, there is no problem those who want to
look at the details of the figures they can go to the particular site and look at those
figures right. And with this I come to the end of the lesson 39 of the industrial
instrumentation.
Prof. A. Barua
Lecture - 40
Signal Conditioning Integrated Circuits
Welcome to the lesson 40 of industrial instrumentation. This lesson is basically, we will

consider some of the signal conditioning circuits, which are available in IC chip form, as
you know, that there are some sensors like LVDT thermocouple. We need some
conditioning circuits in the case of LVDT. We need some circuits for null voltage
reductions; we need some circuit for the phase compensation. Because in LVDT always,
you know there is an input and output phase difference. So, that can be killed, only in the
by using some signal conditioning SRP’s, because in null voltage also I mean, we need
some lead lag networks type of things for the phase. I mean compensation like means, if
I want the input and output should be at the same phase right, because we need some
excitation in the in the case of LVDT.
Now, similarly, we have also we have also in the case of thermocouple. We need cold
junctions compensations for that reasons, we have already, we discussed some chip like,
I AD 5 8 0 that has some advantages in AD f. So, but there are some more chip. I mean
chips, came out which has as a better performance. Similarly, we will also discuss, this
the current voltage to current convertor, which is also available in the chip form. One
thing I should mention that you can do it with discrete component, but; obviously, if you
use some CMOS chip, it will you will get some additional advantages. It is of the shelf
devices, you can take an interface with your either ability or transducers, so for better
results or better performance.
So, in this context we will, so this I mean, lessons is actually, signal conditioning
integrated circuits right. So, basically signal conditioning circuits, we have already,
defined already, discussed some filters, some amplifiers those things. We have already
discussed some logarithm amplifiers antilogarithm amplifiers samples. And whole
circuits all these things, we have discussed in this particular lesson. We will concern
some of the self chips right, which are very handy. We can use it for our for signal
conditioning purposes let us look at the contents.
Contents are we will discuss fast LVDT signal conditioner, then we will look at cold
junction compensator, because as you know that cold junctions has a problem in the case
of thermocouple. Thermocouple, because it is I mean industrially a cold junction is not
possible; that means, one junction cannot be put in a ice. So, there is always a variations
of the one junction and measure junction accordingly. Whatever the temperature, we
want to measure accordingly, it will vary, but cold junction cannot be put in a ice bath, it
is not possible. So, for temperature change, I need some compensation, which will
correct that temperature change right. So, that is a cold junction compensator.
And thirdly, we will consider the voltage to correct convertor in many situations. We
will find that, we need to convert this voltage 0 to 5 volt, 0 to 10 volt to the 4 to 20
milliampere or 0 to 20 milliampere. So, that type of, because industry as I told you
several times, that always we need 4 to 20 milliampere of signal conditioning currents.
So, that type of situation can be handled only, if I have, if I have a power I mean current,
because we never transmit the voltage in industries, always in any transmissions, we
need current to transmit. So, in that type of situations we always want, we always want
the current. So, any voltages, we converted it to the current domain, so standard practice
4 to 20 milliampere. So, we can it with a discrete components what; obviously, we can
do better with the IC chip right. So, this all we discuss in this particular lesson, let us first
look at the LVDT signal conditioner. Now, LVDT you know, that it how does it look?
Let us make a brief, I mean re capsulations it will look like this.
I have a primary, I have a 2 secondaries are there and I have a core inside is not, it that is
our ability. So, I put a excitation here and I will get 2 secondary voltages here, This is the
basic diagram of LVDT right. I put a displacement here, please note LVDT is
electromechanical sensors and it basically, a displacement sensors directly, it is
measuring displacement. But it can measure, so many other indirectly, it can measure
pressure it can measure many other parameters right. So, the basic it is displacement,
now, I am giving an excitation. So, all things, so all those things are I mean very, I mean
deeply discussed or very widely discussed in the case of LVDT in the lesson of LVDT,
but here. Actually I mean, I am making brief discussion to have a look re-capsulations of
the LVDT.
Now, if I you see, if I this move this core up and down say there will be a linkage. If I
put this LVDT up this core up there is more linkage, between primary and secondary. If I
put it down, there will be more linkage between and this secondary right. So, 2 voltage I
will find the voltage will be different now, if I put this in opposition in series opposition
like this 1. So, what I will get the output voltage of a function of output voltage of a
function of the displacement x at width positions, geometric width positions. I will get a
null voltage I will get a 0 voltage rather as you get a 0 voltage, because these 2 out
secondary this 2 voltage will cancel out each other. So, I will suppose to get a 0 voltage,
but due to mismatch of the secondaries always. We will get some nonzero voltage right
due to mismatch of the secondary, if I look at the two secondary I mean, looks like this.
Suppose this is 1, 1 current is I s 1 secondary current is I s 1 and 1 secondary current is 2
right. But you will find that the 2 current will exactly, not be the, it is very difficult to
make a 2 identical magnetic circuits right. So, 2 voltages will never cancel out each
other.
So, there will be a certain residual voltages, that voltage now, if I plot this LVDT as we
know, this all things, we have discussed as I told, you several times x this is output
voltage e naught. So, the response will be look like this one, it should be a straight line,
because it is a linear straight line like this one. This is the output voltage, so if I make a
displacement, if is measure the magnitude it will look like this one, because it is a AC
voltage. So, I will get a amplitude, so this will be plotted clear? So, this will give you our
basic LVDT. Now, you see there is always, some null voltage, which will appear, we
have also shown there is an input output voltage. Sorry this will be ex I am sorry, this
will be our, this will be our, so this will be actually ex excitation right.
Now, for this movement, you will always you will find there is a output voltage. And this
excitation voltage, there is a phase difference between input and output, which is
undesirable properties. Now, all these things, we will find the null voltage cancellation
and always wherever you are using a LVDT. I need a separate excitation, voltage right
separate, because it is like a bridge as you know, we use a Wheatstone bridge.
Wheatstone bridge also need an excitation either DC or AC it does not matter, but in the
case of LVDT since we are using transformer. So, always there is an excitation, so this
can be easily avoided, if I use a particular chip available right. So, this excitation also
inherently, you can vary the frequency of excitations right and I am getting output
voltage, which is free from all this problems let us go back.
Now, the input to LVDT, which is linear variable differential transformer, is the
displacement of the core. And it is output is a pair of ac voltages proportional to the core
position, as we know this is basic LVDT.
It is an electromechanical transducer consisting of a primary winding, energized by an

external sine wave reference sources, external sine wave reference sources, please note
always we need this source outside. And two secondary windings are usually, connected
in the series opposition, this we have discussed in details. When we covered the LVDT
the moveable core change the flux linkage between the primary and secondary windings,
which is giving a non zero output voltage e naught.
Now, LVDT signal conditioner circuit is AD598 it is highlights of the chip, what are
those. So, the chip number AD598 developed by the analog devices. AD598 is the
monolithic IC to be used with an LVDT for signal conditioning purpose a complete
transformation of displacement. Or mechanical position to dc voltage both unipolar and
bipolar is possible directly, because you we have seen that to make it sensitive phase
sensitive I need a phase sensitive demodulator. So, those thing also can be, I mean
eliminated, if I use this AD598.
So, this 3 most 3 most important things, which we need to use an LVDT, I need a first of
all I need a source excitation voltage is necessary. I need is a null voltage I mean, null
voltage reduction circuit that will make the null voltage 0. That means, at the geometric
position 0 position there, should be the 0 null voltage thirdly. I need a phase sensitive
demodulation circuits; that means, sensitive modulation circuit is something like, that on
which side of the null position of the core lies. It will be deducted by the phase sensitive
demodulation circuit; that means, if I look at.
If I use an LVDT, you see that we have seen that, if I put a displacement x I will get a I
mean circuit, I mean output like this one so on. You see on both this is a null position,
this is a null position, this is a output voltage here on both side of the null, I am getting a
non zero output voltage. So, it is very difficult by looking at a voltage, it is very difficult
to tell on which side of the null actually the core lies. So, what I can do? I can do like this
one. So, if I use a phase sensitive demodulation circuit, this is the circuit with output
voltage e naught with phase sensitive demodulator phase sensitive, demodulator right if
the output will look like this one clear.
So, for looking at the voltages or the polarity of the voltages, I can tell whether which
side of the null the voltage lies. So, all these is incorporate or incorporated at AD598. So,
it is a great I mean, so many external components will be saved by using a single chip.
So, AD598 is the monolithic IC to be used with an LVDT for signal conditioning
purpose a complete transformation of displacement or mechanical position or dc voltage.
Both unipolar and bipolar is possible with this, it can used also for the RVDT rotational
velocity displacement transducers, this also possible with this, I mean type of device with
this type of chip.
AD598 accepts a wide range of input and output voltages, these are all the features what
it can and frequencies. It can drive on LVDT primary with up to 24 volt rms and accepts
secondary input levels as low as hundred millivolt rms right. Hence, it can be used with
almost all types of LVDT; that means, you please you see that LVDT and the primary
can go up to 24 volt rms.
The excitation frequency is determined by a single external capacitor by using a single

external. Which is outside the chip by excitation frequency of the LVDT can be I mean,
determined by a single external capacitor the range is 20 Hertz to 20 kilo hertz.
Moreover the input signal need not be synchronous with the LVDT, primary drive thus
an external primary excitation of any desirable frequency, can also be used right this is
another most important thing.
The ratiometric decoding scheme used by AD598 is such that, the ill effects caused by
the primary to secondary phase shift. That I told you will be eliminated and transducer
nonzero null voltage, do not influence the overall circuit performance. So, it will be
independent of the non zero null voltage, which is obvious in all LVDT, because you
cannot make 2 LVDT secondaries, exactly the same right. So, say a little I mean, I mean
dissimilarities of the 2 secondaries will give us a non zero null voltage. This can be
measured very easily in LVDT, if we use without any signal conditioning circuitry.
AD598 can be used in telemetry to the interface electronics, when it is far away from the
LVDT. And it can be drive it can drive an LVDT at the end of the 100 meters of capable
and the position output can drive as much as 30 meter 300 meter of cable. I think it will
be a cable if I am not wrong, so it will be a cable right.
The excitation output of AD598 is thermally protected and AD598 can drive the multiple
LVDT in either series or parallel as long as power dissipation limits are not exceeded.
This is a great advantage you see the is same LVDT same chip can be used for several I
mean, LVDT excitation. So, that is a great saving cost, it can be used as a loop integrator
in the design of the electromechanical servo loops.
Theory of operation, let us look at a functional block diagram of AD is shown in the

figure 1 and AD598 energizes the LVDT primary senses. The LVDT secondary output
voltage and produces a dc output voltage, proportional to the core position, that is
actually, ultimate result; that means, for different core positions; I need a dc output
voltage. Why dc? Because I need a dc, because I have to make the phase sensitive on
which side of the core, the on which side of the null position, the core lies that, I must
know. So, that only can tell by the polarity of the dc voltage right.
To drive the LVDT primary AD598 uses a sine wave oscillator and a power amplifier. It
also consists of a decoder, which determines the ratio of the difference between the
LVDT secondary voltage divided by their sum and a filter and an output amplifier. These
are used to produce the dc output voltage proportional to the core position.
You see this is the block diagrams on the LVDT right LVDT our AD598, this is actually
inside the chip. You can see this portion, this portion is inside the chip, this portion is
inside the chip right and there are pin numbers 2 3. So, 2 and 3 will be connected to the
primary of the LVDT and 10 11 and 17 will be connected to the 2 secondaries of the
LVDT 10 and 11 are 2 extreme point of the secondaries. And one will be connected 2
secondaries are will be connected like this, as it is shown and it is grounded and these 2
signals are coming to an signal conditioning circuits which will make 2 signals. That
means, A minus B divided by A plus B means; that means that is; that means, VA
actually, it is VA minus VB divided by VA plus VB ratiometric principles as I told you.
So, this will make when the 2 voltages are equal; obviously, this should be 0 right; that
means, output voltage will be 0. So, output voltage will come out as a 0 here a filters and
amplifiers and this is an oscillator and amplifier to have a stable oscillation. What they
do actually, in the case of LVDT they make first make a triangular generate a triangular
wave. They generate a triangular wave then with the signal processing with they convert
the triangular wave approximate sine wave. Because the triangular wave can be
converted to a sine wave, then it is fit to the amplifier and it is fitting through the primary
of the, this is the primary, this is the primary of the LVDT clear. Now, how you do it?
You see that, if you can look at, it looks like this one.
So, triangular waves are like this. So, with diode, I can what I can I can round of this
value. So, it will look like this one like this one though it is not exactly, sine wave, but
for most of the purposes, it will suffice. It is a continues form, there is error, if I look at
there is a distortions and all these does not matter, it hardly matters, because nobody say
that exactly use. We have to use the sine wave there right and this frequency of this is
determined of this triangular wave is determine or the sine wave approximated sine wave
is determined by an external capacitor clear right. So, this voltage will be divided by a A
plus B, so it will give you the filter, this is the block diagram, but inside there is a much
detail circuit.
This is the chip pin diagrams, we can see it is a 20 pin DIP, there will be line packages.
So, we have all the excitation lines 2 3 we have filter lines 8 9, we have a output filter
then we have signal output. So, this basically, to excitations will go to primary and we
have a dual supply V plus Vs and minus Vs either single ended. If you need single ended
this is to be grounded; that means, if I need only unipolar. So, this is to be grounded and
this output is signal output will come out, you can take the output filter also and signal
differences right this is our pin diagram.

The oscillator consists of multivibrator, because multivibrator as you know multivibrator
frequency. I can use wind bridge oscillators and all these things, but multivibrator
frequency will be more stable. So, that is the reason they use multivibrator, which
produce a triangular wave output the triangular wave drives a sign shaper. Which
produces a low distortion sine wave who is frequency oscillation is determined by a
single capacitor. So, I this one is not exactly, the low distortion side there are several
distortions, it would not be a pure sine wave. So, there will be certain amount of
distortion lies, there as I told you it does not matter much right. Output frequency of the
oscillator or oscillator ranges from 20 Hertz to 20 kilohertz and the amplitude from 2 volt
rms to 24 volt rms, total harmonic distortion is typically minus fifty db.
The output from the LVDT secondaries consists of a pair of sine waves, whose
amplitude difference is VA minus VB and is proportional to core position. This will
proportional to our core position, this always LVDT, we want like that. The previous
LVDT signal conditioners whatever, because we knew already, we did this type of
things. We made a lot of adjustments of the, we need a as I told you, we need some a
some signal conditioners, circuit to kill the null voltage to kill the phase difference
between input and output. Previous LVDT signal conditions a conditions conditioning
circuits, rather used to synchronous used to synchronously detect this amplitude
difference and convert the absolute valve to a voltage proportional to position.
This technique is used the primary excitation voltage as a phase reference to determine
the polarity of the output voltage. There are number of drawbacks associated with this
technique. We had to produce a constant amplitude constant frequency excitation signal,
which is very difficult; you see this is very important. While you are measuring with
LVDT always, we are telling that the amplitude of the input voltage should remain
constant right is not it? If the amplitude of the input voltage excitation, voltage falls then
what will happen? Like this one, suppose if I tell that the it will happen what will
happen? Suppose I have an LVDT.

So, this voltage should remain constant excitation of the ex will remain, amplitude of the
ex will should remain constant. It should have it constant frequency also, otherwise what
will happen? If suppose for some reason this amplitude falls, this output voltage also will
fall. Suppose this is series opposition output volt that is undesirable, because only the
output voltage amplitude will tell me the exact position of the core. So, I will get some
ambiguity, but that is not possible in this particular chip, because we are using
ratiometric principle, they are using some ratiometric principle.
So, it does not matter I mean, if the input voltage frequency falls or rises automatically,
output voltage will be independent of that. So, I do not need constant amplitude constant
frequency signal right. Though, it is internally generated, but some reason or the other, if
the amplitude falls it does not matter. This technique requires compensation for LVDT
primary to secondary phase shifts this technique requires, compensation for LVDT
primary to secondary phase shifts.
Also these shifts varied as a function of temperature and frequency adding to the
complexity of the compensation method. So, this all thing can be avoided AD598
eliminates all of these problems, let us take 1 by 1. It does not require a constant
amplitude, because it works on the ratio of the difference and sum of the LVDT output
signal. So, automatically, it will cancel out right. So, any change any variations of the
input excitation voltage frequency and the it will be automatically cancel out.
A constant frequency signal is not necessary, because the inputs are rectified only the
sine wave carrier magnitude is processed. Input voltage will be rectified and the sine
wave carrier signal will be processed. The phase shifts between the primary excitation
and the secondary output is not important, because the synchronous detection is not
employed right this is another advantage.

The ratiometric principle used in AD598 requires, that the sum of the LVDT secondary
voltage remains constant with LVDT stoke length. Stoke length means, that length
through, which the LVDT travels right that is called the stoke length.
Now, next signal conditioning circuits, I am going to discuss is a thermocouple signal

conditioners using AD594 and AD595. Because some I mean, thermocouple always, we
have seen that we need a cold junction compensation is a headache for thermocouple.
Because ambient temperature may vary, but that does not influence, my reading suppose,
I am reading a thermometer with a thermocouple the temperature of 700 degree
centigrade. If my ambient temperature varies from 5 degree to I mean 40 degree
centigrade, it should not it should not influence my reading. So, that is only possible, if I
use this type of I mean, cold junction compensation. Cold junction compensation chip
was already, there with this improved chip compared to the, what whatever the cold
junction compensation chips.
And it eliminates many additional circuitry which is necessary along with the AD590 or
5 8 0 and features of the chip, it is a pre-trimmed for J type or type K thermocouples. As
you know, industry we always, called the thermocouple by J K SRT depends on the,
what particular alloys they are using. So, it is pre-trimmed for for J type it is AD594 for
K type, which is mostly widely used thermocouple in the industry. It can be used with
type T thermocouples inputs also which is a platinum platinum rhodium thermocouple,
platinum rhodium platinum thermocouple. I should say platinum rhodium is a positive
voltage and negative I mean, platinum is the negative. It has low impedance voltage
output of 10 millivolt per degree centigrade and a wide power supply range of plus 5 volt
to plus minus 15 volt.
It has built in ice point compensation. It has inbuilt devices like, thermocouple failure
alarm, because if the thermocouple junction open, so that some alarm should be given
which is not there previous thermocouple. So, it will give some alarm, some voltage
which can be energy which can be used to some trips of the alarm. So, that the, when the
person the process know, that this thermocouple is out of order. And a self contained
celsius thermometer operation is also possible direct readings of the thermometer in
temperature degree centigrade is possible. There is another advantage, it has a low
power, it is less than 1 millivolt typically and high impedance differential input is there.
So, common mode noise problem can be immediately eliminated.
Basics of thermocouple, one of the most widely, used devices for temperature
measurement is the thermocouple. It works on the basic principle, that when 2 dissimilar
metals are joined at both ends, just have a short or brief recapitulations of the
thermocouple. We are covered in details, but let us have a small discussion, because all
this, chips are related to the thermocouple, this particular 2 chips what we talked about is
related to the thermocouple. So, let us briefly discuss principles of thermocouple, it
works on the basic principle that when two dissimilar metals are joined together at both
ends. One end is heated a current will flow now, if the loop is broken at the center an
open circuit voltage, see that voltage is generated. And is proportional to the difference
in temperature between the 2 junctions we all know these thing right.
Therefore and in determining the temperature of the measuring junction, the reference
junction temperature must be known right, because it is the difference of temperature,
which will actually give us the voltage right? If the 2 thermometer, 2 junctions are the
same temperature output voltage will be 0, right. So, the difference; that means the ice
point or the reference junction temperature, which is not varying, which is supposed to
be at the ambient temperature should be known. An ice bath provides a well defined
temperature of 0 degree centigrade for the reference junction, but that is not possible in
any process industry.

This we can see the thermocouple look, you see we have I have a measuring junction
here right. I have a measuring junctions, let me take this one I have a measuring junction
here, I have a reference junction which is at ice point at 0 degree centigrade degree
centigrade? I have if; it is not if no voltmeter is connected, then what will happen? I will
get a current through it, but I break it open circuits. So, I will get a voltage, so these
voltage is proportional to the difference of the temperature T 1 and T 2. That means, the
VT 1 minus VT 2 is the function of T 1 minus T 2 as the temperature is higher and
higher temperature differences is higher and higher T 1 minus T 2 is higher and higher. I
will get the higher valve of the voltage VT 1 minus VT 2, this is the basic I mean
principles of the thermocouple.
A voltmeter is used to measure the voltage developed as shown in the figure and 2
additional junction J 2 and J 3 are formed at the connection between the thermocouple
connecting wires to the meter. Because that cold junction is not there in industry as I
told. You excuse me these 2 junctions produce opposing voltage within the thermocouple
loop right. So, one in most important things is that we have to this should be isothermal
block I will show you in the next figure.
Using isothermal block at the point of connection keeps, these junctions in thermal
equilibrium and produces the equal and opposing emfs. So, cancelling that, that will not
that will not come in the net or to the I mean, voltages on net voltages. Therefore, the
measured voltage is the difference in potential between the measuring junction and the
isothermal block, which serves as the reference junction.
You see here, fine I have a thermocouple material A right, let me take this one. So, this is
a this is a metal metal A, A I mean, is a alloy A is alloy B right, this is a measuring
junctions. So, I have a, so usually, connecting wires are made of copper. So, if these 2
junctions the, because you see the metal A and copper there is 1 junction and this is
another junction, this is one junction; this is another junction. So, for these 2 junctions
are at the same temperature, that which can achieve this one by an isothermal block. So,
there is no change in the net output voltage net, output voltage, we have measuring it
with this one. Because we are connecting with the different wires so; obviously, if the
different wire are connected dissimilar metal.
So, again thermocouple will be developed, this can be easily managed, if these 2 are put
in a isothermal block, that is we talked about cold junction compensation for industrial
application. It is impractical to use an ice bath for the reference junction it is not
possible. I mean, so it will remain open, by compensating for the voltage developed at
the reference junction. The ice point reference may be eliminated this is achieved by
adding a voltage into the thermocouple loop equal, but opposite to that of the reference
junction.
AD59 4 AD595 are the family of thermocouple signal conditioning ICs. That provide
cold junction compensations as well as output voltage amplification, it also provides an
alarm when the measuring junction is open; that means, the junction reference. I mean
measuring junction is open for cut or some suppose, if it is made by welding. So,
welding is broken, so I need some I need some alarm. So, that can be possible also with
this particular chip.

So, circuit description so; obviously, you can see that, it saves the lot I mean, I do not
additional circuits to make all these to achieve all these extra features. Figure 5 shows a
block diagram of the AD594 AD595 thermocouple signal conditioners. Type J, which is
for I mean AD594 is to be used with the type J and type K is AD595 thermocouple is
connected to the pins 1 and 14 and the inputs to an instrumentation amplifier differential
stage.
The input amplifier is contained in a loop that uses the ambient or room temperature as it
is reference. The IC here is also at the ambient temperature and the cold junction
compensations. Circuits develops a voltage equal to the deficiency due to the non ice
point reference junction. The voltage is then applied to a second preamplifier, whose
output is summed with the output of the input amplifier.
The resultant output is then applied to the input of a main amplifier with feedback to set
the gain of the combined signals, I will show you all this in the circuit. The ice point
compensation voltage is scaled to equal to the voltage that, would be produced by an ice
bath referenced thermocouple measuring the ice temperature IC temperature. So, but this
part is concerned, this is very similar to the, whatever the old chips we have used AD 5 9
0 right, though some additional features are there 580 is for stabilized cause 590 is
actually thermocouple these are later versions.

This is our actual circuits, you can look at this is the thermocouple circuits right. You see
here, so this is we have 1 amplifier here, another amplifier another amplifier here, this
are summed up and coming to another amplifier, and see there this overload detection
circuit. So, this will be give a alarm here, which is alarm transistors, we are using and
some ice resistors are there. So, this is the complete block diagrams, our block diagrams
inside. I should not say the detailed circuit diagram of the AD594 or 595, this is a block
diagram of AD594 this is the factory trim. So, it has advantages; that means, I do not
need additional capacitor I mean, additional resistance. Which are we need in the case of
AD 5 9 0 there is a chart available for particular, Whatever the particular reference I,
resistors we have to use, but this is not necessary here, it is already inside the chip.
The voltage is then added with the locally referenced loop voltages, which will make the
compensation right and though the feedback path through, the feedback path the main
amplifier maintains a balance at it is inputs. In the case of open thermocouple, these
inputs, become unbalanced and the fault is detected. And the overload detection circuit
drives a current limited n-p-n transistor. That gives an output of which can be as an alarm
which can be used as an alarm, which can be used I should say, which can be used which
can be used as an alarm, right.
Output voltages: To produce a temperature proportional to output, you see that further
advantage of this. One that I am getting in amplifications of 10 volt per degree
centigrade, it is a quite high usually, you know that this thermoelectric power usually, I
mean; that means, that sensitivity to thermoelectric is quite low. It varies typically,
suppose from 60 degree 60 micro volt per degree centigrade to 12 micro volt per degree
centigrade in the case of thermal constant. It is coming around 60 micro volt per
centigrade whereas, in the case of platinum rhodium platinum thermocouple. It is 10 to
12 microvolt per centigrade, which is quite low, but here you see at the outputs. I am
getting 10 volt per degree centigrade.
So, sensitivity is extremely, high you can see here 10 millivolt per degree centigrade.
You can see here, this sensitivity is quite high. So, it is already, a signalled process and
we are getting. So, much of sensitivity here, so the resolution; obviously, will increase.
So, provide and to produce a temperature proportional output of 10 millivolt per degree
centigrade provide an accurate reference junction. Over the rated operating temperature
range AD594 AD595 is gained trimmed during the manufacturing to match the transfer
characteristics of type J and K thermocouples at 25 degree centigrade.
At 25 degree centigrade calibration temperature the sensitivity are 51.7 micro volt per
degree centigrade, for type J and 40.44 microvolt per degree centigrade, for type K
thermocouples respectively. This gives the gain of 193.4 or AD594 and 247.3 for AD595
to realize a realize a output of 10 millivolt per degree centigrade output.
Although the device is trimmed to give an output of 25 to 50 millivolt an input offset

error is induced in the output amplifier resulting in offset of 60 micro 16 microvolt and
11 microvolt for the AD594 and AD595 respectively. The actual output can be
calculated, using the following equations now, what are the equations? Output of AD594
voltage J type plus 16 microvolt multiplied by 193.4. This is the actual output voltage I
will get from this circuit. And in the case of AD595, I will get a output AD595 voltage K
type plus 11 microvolt into 247.3.
Where the type J and K voltages are taken from table 1 referred to 0 degree Celsius.
Already, tables are available and for the reference, I am giving this table. So, it is J and K
J is iron constantan and K is the chromel alumel this are the most widely, used
thermocouple in industry. So, maximum temperature range 0 degree to 750 degree
centigrade. So, within that range I am getting 0 to 40 42.283 millivolt and in this case
5.973 to 50.633 millivolt right. So, this is the range, I am getting of voltage swing I am
getting.
Thermocouple output is linear, only over a narrow temperature. That is you know
thermocouple is non-linear devices over a short range only. I will get a linear
characteristics, but however, it is a non-linear. So, you should take a wide range, if it is a
if you can express at the polynomial equation, that I told you earlier also it looks like, if I
take it is e naught sorry.
Output voltage e naught equal to at plus bt square plus ct cube right. So, it is the
temperature created between the 2 junctions. So, it is the non-linear, you can see this one.
So, for only for sometimes, you will find it can be 7 degree polynomial sometimes, you
will find it can be represented by a 8 degree polynomial. So, it is a highly non-linear
devices, only in the case of platinum rhodium platinum thermocouple. You will find that
it is 2 to third degree polynomial, it can be represented by. AD594 AD595 do not
linearization, linearization technique implemented externally, when it is to be used over a
wide temperature range.
Optimizing performance by following certain precautions or adding additional networks

the performance of the chip can be enhanced or optimized. Cold junction error; this one
error, optimal performance from AD594, 5 9 5 is achieved when the thermocouple cold
junctions and the IC chips are at the thermal equilibrium. It should be at the same I mean
equilibrium; that means cold junctions. So, if you put usually, if I put AD594 and AD595
very close to the isothermal junction; that means, where I am taking the connecting the
copper wire from the alloy of the alloy wire of the thermocouple. So, this will satisfy this
statement cold junctions the cold junction errors will not be there.
By avoiding the placement of heat generating devices or components near the chip, this
can be achieved where you cannot. If I put for away any heat generating like,
transformers or that type of things, that can be easily can be achieved this problem, can
be easily solved. Soldering the proper soldering technique is to be implemented to bond
the thermocouple to the printed circuit boards, proper I mean, typical clocks and
whatever the thing; that means, there should not be that type of thing.
Cleaning the thermocouple wire to remove the oxidation, before soldering helps, because
you see that thermocouple some thermocouple like, iron constantants are very prone to
oxidation. So, that, so if I it is prone to oxidations, you must clean it the before soldering.
Because 2 junctions is to be 2 wires is to be soldered either soldered and weld it.
Whatever it may be even, if you can just tie like this 1, 2 wires it will work, but the, I
need for physical strength, it need either soldering or welding. So, soldering if you do,
you should first clean up those junctions, those portions then we weld or sold solder it.
Basic current return the input instrumentation amplifier of the chip requires a return path
for it is input bias current and this should be left floating.
If the thermocouple measuring junction is electrically isolated then the pin 1 of the IC
should be connected to pin 4 to power supply common. In some cases tying the
thermocouple directly to the common is not possible a resistor is connected between pin
4 and pin 1. Will satisfy that bias current return path, but will however, generate an
additional input offset voltage due to the one hundred nano ampere bias current flowing
through it.
If it is the grounded thermocouple or if it is a small common mode potential is present it

is recommended not to connect pin 1 and pin 4 together, right.
Now, last of this particular lessons we will discuss some programmable isolated voltage
to current convertors 1B22. This is also many a places we need, because we cannot
transmit current we cannot transmit voltage, we have to always transmit current. So, I
need this type of convertor in many places whenever I am getting a signals pneumatic
signals or anything which is to be converted into voltage first. Then to transmit to the
actual point of control; that means, which is mostly over the PC.
So, I need a voltage to current convertor. So, all this things can be can be achieved by
this chip 1B22. Let us look at the details of this one. Features; internal isolated loops
supply drives 1000 ohm load. So, it has a large load it can be supplied. Pin
programmable inputs I can have 0 to 5 volt or 0 to 10 volt input. I can have 20 to 5 volt
usually typically you will find in process in industries will have two types of either 0 to 5
volt 0 to 10 volt output. I can have either 0 to 20 or 4 to 20. So, pin programmable
outputs, pin programmable inputs and pin programable outputs 4 to 20 milliampere or 0
to 20 milliampere.
High input to output isolation high input to output isolation 1500 volt rms. So, it is a
large isolation that is very much necessary to I mean to satisfy to protect my other
instruments. Normal mode, output protection that is 240 volt rms. Low non-linearity it is
plus minus 0.02 percent that is also greater advantage high CMRR it is 90 degree. So,
common mode rejection ratio is quite high.
General description 1B22 is an isolated voltage to current convertor that incorporates

transformers isolation to achieve high performance and automated surface mount
manufacturing for low cost and increased reliability. The voltage to current convertor
consists of 4 basic sections we have the input conditioning circuits; we have modulator
demodulator; we have isolated loop and supply and current source this are the basic
blocks.

The 1B22 is a pin programmable for 0 volt to 5 volt or 0 volt to 10 volt inputs and 0 to
20 milliampere or 4 to 20 milliampere outputs using a internal resistor network. It can
also be set by an external resistor to accept 0 volt to 1 volt or 0 volt to 10 volt input. The
transformer coupling provides 1500 volt rms galvanic isolation between the inputs and
the current loop.
Loop power is generated internally through the dc dc convertor and is also isolated from
the input side which is 1500 volt rms. Loop compliance voltage is dependent on the
voltage supplied to the 1B22 that chip and with V loop equal to 28 volt it is sufficient to
drive a one thousand ohm load. It can operate as a wide temperature range it a two type
of I mean chip available once is for commercial, which is minus 25 degree centigrade to
85 centigrade industrial range is minus 40 degree centigrade to 85 degree centigrade.
This is resistor network you see this is our basic block diagrams of our, this isolation you
see the power isolation there is a signal isolation. So, all this isolation are there in this
particular we have oscillators here this is our entire chip of this 1B22. This is an input
input 1 and input 2, you can see which is coming the 4 milliampere is to be keep I mean
this one.
The conditioned the signal is modulated to generate a square wave, that drive the
transformer T 1. The peak to peak amplitude of the signal is proportional to V IN. This is
the transformer T 1 please note this is the transformer T 1 this is the transformer T 1 and
this is the transformer T 2 this is the transformer T 1 and this is the transformer T 2. The
conditioned signal is modulated to generate a square wave that drives the transformer T 1
the peak to peak amplitude of the signal is proportional to V IN right. So, that is actually
we what we want a internal high stability reference with a nominal output voltage of six
6.4 volt is used to develop a 4 milliampere offset for the 4 to 20 milliampere of current
loop output.
After passing to the signal transformer T 1 the amplitude modulated signal is

demodulated and filtered by a single pole filter. That means, first order filter this filtered
output is the control signal for the voltage to current convertor stage. Timing
informations for the demodulator is provided from the power transformer T 2. The 1B22
outputs are protected from the accidental short circuit to ac line voltage up to 240 volt
rms that is already I mean specified.
Now, dc to dc convertor consist of a power drive power transformer T 2 and a full wave
rectifier and a filter right. Now, this dc to dc convertor provides the power for the output
circuitry as well as the isolated compliance voltage for the loop and this voltage is
proportional to V loop on the input side right. So, this will give everything this will give
the our power supplies also. So, a dc to dc convertor is necessary. The 1B22 requires a
plus minus 15 volt supplies to power the input side circuitry and plus 14 volt plus 30 volt
supply for the dc to dc convertor.

So, basic interconnections you can see here the basic interconnections are shown in
figure next slide for 0 to 5 10 10 volt signals either IN 1 or IN 2 can be used for input for
0 volt 5 volt signals jumper IN 1 to IN 2. So, this is to be shorted for 4 to 20 milliampere
the 4 milliampere this is also we call is the, I mean pre programable programable in that
sense. That means, input is programmable and that is output, is programmable here for 4
to 20 milliampere operations. The 4 milliampere offset node, node should be jumped to
the S node while for 0 to 20 milliampere it should be tied to the COM right, because 4
milliampere is not necessary.
The functional diagram of the resistors networks in the 1B22 is shown in figure 8.
This is our functional diagrams you can see here how it works actually input 1 and input
2 these are resistor network 4 milliampere offset, we are using some capacitors here. So,
we have a transformers we are showing like this one this are basic to transformer T 1,
this is our T 2, this is a rectifier loop supply.
You can see here how the resistors networks are connected right? This is resistant
networks in one B I mean this particular chip. What are internal networks? This are all
inside the chip you can see where 6.4 volt reference voltage which can be generated by.
Applications we have lot of applications; we have multi channel process controls; we are
using this particular chip this multichannel process controls we can have a I mean
process controls. We always need this conversion from 0 to 5 volt to 4 to 20 milliampere
of current. So, that is possible A convertor current loop interfaces then we have analog
transmitters and controllers where I have to transmit that signal many places even though
it is digital. But the transmitter is analog, because to the computer when it will go it has a
it I mean data acquisition systems are the input. So, it will convert in the digital domain,
but before that the, from sensor through the transmitter it will go.
So, the transmitter itself this 4 to I mean this 1B22 should be incorporated within the
transmitter. That means, which will converts like Ph suppose Ph whatever the voltage I
will get that convert to the current of 4 to 10 milliampere that can be achieved there.
Remote data acquisition systems; this also can be used, because it has a transmission
capacity of about a large distance of 4 to 20 milliampere even the current is that can be
sensed that can be used as a data acquisitions signal. So, all these features are there in the
case of this particular chip this has several advantages this particular chip. Because
always we are getting a stable 4 to 20 milliampere of current in this cases right. And with
this I come to the end of this particular lesson.

Pdfjoiner

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pdfjoiner

Uploaded by

Copyright:

Available Formats

Industrial Instrumentation

Prof. Alok Barua

(Refer Slide Time: 3:57)

(Refer Slide Time: 4:13)

Now, classification of instruments, so they, I mean instrument can be classified in 4

(Refer Slide Time: 9:02)

(Refer Slide Time: 13:17)

Even though it is not very relevant whether it is monitoring or transmitting instruments,

(Refer Slide Time: 16:51)

(Refer Slide Time: 18:50)

(Refer Slide Time: 25:15)

(Refer Slide Time: 28:45)

(Refer Slide Time: 30:51)

(Refer Slide Time: 31:53)

(Refer Slide Time: 32:29)

(Refer Slide Time: 33:14)

(Refer Slide Time: 34:35)

(Refer Slide Time: 36:14)

(Refer Slide Time: 39:43)

(Refer Slide Time: 40:09)

(Refer Slide Time: 41:26)

Linearity, if the calibration is very important, because in instrumentation you know

(Refer Slide Time: 42:50)

Now, it is normally desirable that the output reading of an instrument is linearly

(Refer Slide Time: 46:14)

(Refer Slide Time: 48:33)

(Refer Slide Time: 50:41)

Now calibration, actually it is very important. It is a procedure that involves a comparison

(Refer Slide Time: 52:29)

(Refer Slide Time: 53:29)

(Refer Slide Time: 54:28)

(Refer Slide Time: 54:45)

(Refer Slide Time: 55:41)

(Refer Slide Time: 56:09)

Now, resolution is another important … The measurement precision of an instrument

Drift is the calibration of an instrument. It is usually performed under some control

Lesson 2 of industrial instrumentation, we will discuss basically the dynamic

(Refer Slide Time: 1:24)

Now, if I look at the dynamic characteristics, the dynamic response of an instrument to a

(Refer Slide Time: 5:46)

(Refer Slide Time: 6:20)

The simplest model of a measurement system is a zero order differential equation. It

(Refer Slide Time: 7:44)

(Refer Slide Time: 8:29)

(Refer Slide Time: 10:12)

(Refer Slide Time: 13:51)

Now, next we will consider the first order instrument.

(Refer Slide Time: 14:19)

(Refer Slide Time: 15:58)

(Refer Slide Time: 16:45)

(Refer Slide Time: 18:25)

(Refer Slide Time: 19:47)

(Refer Slide Time: 21:38)

x-axis is my input, y is, my y-axis is my output y by KX s. Now, interestingly, you see

(Refer Slide Time: 24:34)

(Refer Slide Time: 25:56)

(Refer Slide Time: 27:53)

(Refer Slide Time: 29:02)

(Refer Slide Time: 30:25)

(Refer Slide Time: 30:45)

(Refer Slide Time: 34:23)

(Refer Slide Time: 35:27)