Performance Characteristics of Measurement System

ENGG3201-Measurement Technique
2.0 Introduction
In this chapter we discuss system characteristics that typical measurement elements may posses and their effect on the OVERALL PERFORMANCE of the system.
System Characteristics Statistical Characteristic - Accuracy - Repeatability - Tolerance
Systematic Characteristics
Static Characteristics - Range, Span, ,Sensitivity, Resolution, etc
Dynamic Characteristics -Rise time, settling time, MP, etc
2.1 Systematic Characteristics

Systematic characteristics are those that can be exactly quantified by mathematical or graphical means. These are distinct from statistical characteristics which cannot be exactly quantified.
Static(steady-state) characteristics - are the relationships which occur between the output O and input I of an element when I is either at constant value or changing slowly(Figure 2.1)
Input I Output O
Element
Figure 2.1 Meaning of element Characteristics
Static(steady-state) Characteristics (contd)

Assume I(t) is slowly varying quantity and suppose
Imin - minimum input
Imax - maximum input

Omin - minimum output Omax - maximum output
Range: The input range of an element is specified by the minimum and maximum values of I ,i.e. Imin to Imax ( the same definition hold for O ). E.g., a pressure transducer may have an input range of 0 to 104 Pa and an output range of 4 to 20mA.

Span: - is the maximum variation in input or output, i.e. input span is Imax - Imin , and output span is Omax Omin. Thus in the previous example the pressure transducer has input span of 104Pa and output span of 16mA. Ideal straight line: An element is said to be linear if corresponding values of I and O lie on a straight line defined by eqn.(2.1)

(2.1) where: and Thus the ideal straight line for the pressure transducer (discussed before) is:
Note:
The ideal straight line defines the ideal characteristics of an element. Non-ideal characteristics can be then quantified in terms of deviations from the ideal straight line.

Non-linearity: can be defined(Figure 2.2) in terms of a function N(I) which is the difference between the actual and ideal straight-line behavior, i.e. (2.2)
O Omax Ideal KI+a B(Imax, Omax)
N(I)
Imin
Imax
N(I)
Actual O(I) Imax I
A(Imin, Omin)

The maximum non-linearity, , is expressed as percentage to full-scale deflection(f.s.d), i.e. as a percentage of span. Thus: (2.3) Max. non-linearity as percentage of span Note: In many cases O(I) & therefore N(I) can be expressed as a polynomial in I:

Sensitivity: -Is the slope or gradient of the output versus input characteristics O(I) (2.4) For a non-linear element Hysteresis: For a given value of input I, the output O may depend on whether I is increasing or decreasing. Hysteresis is, thus, the difference between these two values of the output (Figure 2.3)

Hysteresis (contd) (2.5) Hysteresis Hysteresis is usually quantified in terms of the max. hysteresis express as a percentage of f.s.d., i.e. span.
O Figure 2.3 Hysteresis Omax H H(I) Omin
Imin
Imax
Imin
Imax

Resolution: -Is the largest change in I that can occur without any corresponding change in O.
(2.6) Resolution expressed as a percentage of f.s.d
Figure 2.4 Resolution & potentiometer example.

Non-linearity, hysteresis and resolution effects in many modern sensors and transducers are so small that it is difficult and not worthwhile to exactly quantify each individual effect. In these cases the manufacture defines the performance of the element in terms of error bands(Figure 2.5) P(O) O
Figure 2.5 Error band and rectangular PDF OMax
1 2h
Oideal
h h
2h
Omin
IMin
IMax
Oideal
Environmental effects:
In general, the output depends not only on the input signal I but also on the environment inputs such as ambient temperature , atmospheric pressure, relative humidity, supply voltage , etc.
There are two main types of environmental input. [1] Modifying input(IM)- causes change of the linear sensitivity of the element. If IM is the deviation in the modifying input from the standard conditions(IM =0), then this cause the linear sensitivity to shift from
Environmental effects(contd):
Figure 2.6 Modifying & Interfering Inputs
Environmental effects(contd):
[2] Interfering input (II)- causes the straight line intercept ( or zero bias) of the element to change from If both modifying & interfering input differs from the standard value ( i.e., and ) , then eqn.[2.2] becomes
(2.7)
A numerical examples on determining the value of Km, KI, a & K associated with the general model equation will follow in the tutorial section.
2.2 Generalized model of a system element

Figure 2.7 shows eqn.(2.7) in block diagram form to represent the static characteristics of an element. For completeness the diagram also show the transfer function G(s) , which represent dynamic characteristic of an element.
Figure 2.7 General Model of element
Im(Modifying) Km X
input
II(Interfering)
KmImI KI
KI O G(s) Dynamic
N() N(I)
Static
2.3 Dynamic Characteristics:

If the input signal I to an element is changed suddenly, from one value to another, then the output signal O will not instantaneously change to its new value.
The ways in which an element responds to sudden input changes are termed its dynamic characteristics, and these are most conveniently summarized using a transfer function G(s). For example, if the temperature input to a thermocouple is suddenly changed from 250C to 1000C, some time will elapse before the e.m.f. output completes the change from say 1mV to 4mV.
2.3.1 Transfer function G(s) for typical system elements

As an example , let find G(s) of thermocouple inserted in fluid environment.
preliminaries
Heat transfer take place as a result of one or more of three possible types of mechanism; conduction, convection*, radiation

Preliminaries(contd) The temperature of a sensing element @ any instant of time depends on the rate of transfer of heat both to and from the sensor. Returning back to the example, we have from Newtons law of cooling the convective heat flow W watts between a sensor @ T0C and fluid @ TF0C is given by (2.8) Where; U: is the convection heat transfer coefficient and [Wm-2 0C-1] A: is the heat transfer area[m2]

Thermocouple example (contd)
The increase of heat content of the sensor(in joules) is given by
(2.9)
Where; M: is the sensor mass [Kg] C: specific heat of the sensor material [JKg-1 0C-1] Assuming M and C constants: (2.10) Rate of increase of sensor Heat content is
Defining T=T-T(0-) and TF=TF-TF(0-) to be deviations in temperatures from initial steady-state condition, the differential equation describing sensor temperature change is

(2.11) i.e. The quantity MC/UA has the dimensions of time:
And is referred to as the time constant ( ) of the system. The differential equation become;
(2.12)

Taking LT on both sides of eqn.(2.12) gives (2.13) Where; T(0-): is the temperature deviation @ initial condition prior to t=0 and by assumption, T(0-)=0 Therefore, eqn.(2.13) becomes (2.14) Or , equivalently (2.15)

The TF in eqn.(2.15) only relates changes in sensor temperature to changes in fluid temperature. The overall relationship between changes in sensor output signal O and fluid temperature is;
(2.16)
Where; O/T is the steady-state sensitivity of the temperature sensor. (for an ideal element O/T=K).

Example 2.1
For a copper-constantan (Type-T) thermocouple junction, the first four terms in the polynomial relating e.m.f. E(T), expressed in V, and junction temperature T0C are
for the range 0 to 4000C.

Calculate: (a) E/T for small fluctuations in temperature around 1000C. (b) Assuming =10s, find G(s) relating e.m.f & fluid temp0

Example 2.1(contd)
Solution: (a)
Evaluation @ 1000C yields
(b) Using eqn.(2.16) , we get
Exercise2.1: For the same thermocouple find (a) EIDEAL(T) & (b) N(T)

In the general case of an element with static characteristics given by eqn[2.7] and dynamic characteristics defined by G(s), the effect of small, rapid changes in I is evaluated using Figure 2.8, in which steady-state sensitivity (O/I)I0 =K+KMIM+ (dN/dI )I0 , and I0 is the steady-state value of I around which the fluctuations are taking place.
Figure 2.8 Element model for dynamic calculations
2.3.2 Identification of the dynamics of an element

Second-order elements
Transfer function for standard 2nd order element is given by (2.17)
Unit step response of 2nd order element is found as:

Input I (2.18) G(s) Output
Expressing (2.18) in partial fraction we get:

Second-order elements(contd)
(2.19)
Where And this gives
(2.20)

Three different case result depending on the value of Case 1: Critically damped (=1) (2.21) Case 2: Underdamped(<1)
Using Trigonometric relationship (refer the triangle above), we get (2.22)

Figure 2.9 Response of a 2nd order element to a unit step

The time TP @ which the 1st oscillation peak occur is given by: (2.23) The settling time Ts ( the time for the response to settle out approximately within 2% of the final steady-state value) is given by: (2.24) Maximum Overshoot(MP), which is the difference (TP)-1 (for <1), is given by: (2.25) Reading Assignment: Go through the derivations of TP, Ts & MP

Sinusoidal response of 1st and 2nd order element:
Input
KG(s)
output
Figure 2.10 Frequency response of an element with linear dynamic
In the steady-sate , the output O satisfies the following 4 rules: O is also a sine wave The frequency of O is also The amplitude of O is The phase difference between O & I is

Sinusoidal response of 1st and 2nd order element(contd):
Example 2.2 Using the rules on pp 31, find the amplitude ratio & phase relations for a 2nd order element with:
Solution: so that
(2.26a)
Amplitude ratio =
(2.26b)
Phase difference =

Sinusoidal response of 1st and 2nd order element(contd):
Figure 2.11 Frequency response characteristics of 2nd order element with

Sinusoidal response of 1st and 2nd order element(contd): Note:
Both amplitude ratio & phase characteristics are critically dependent on the value of ( enq(2.26))
For < 0.7, |G(j)| has a maximum value which is greater than unity. This maximum value is given by:
(2.27a)
And occurs at the resonant frequency R where:

(2.27b)
(To verify eqns(2.27), first find
to solve R .
By measuring |G(j)|MAX and R ; and n can be found.
2.4 Statistical Characteristics

Statistical variations in the output of a single element with time- repeatability Repeatability- is the ability of an element to give the same output for the same input when repeatedly applied on it. the most common cause of lack of repeatability in the output is
due to random fluctuation of the environmental input( IM,II) with time. If the coupling constants( KM,KI) are non-zero, then there will be corresponding time variation in the output. By making reasonable assumptions about the probability density
functions(PDFs) of the inputs, I,IM,II we can find ( or at least approximate) the probability density function of the output O.
2.4 Statistical Characteristics(contd)

Statistical variations in the output of a single element with time- repeatability(contd) Very often, the PDF of the inputs can be assumed to be the Normal probability distribution or the Gaussian distribution, i.e.,
(2.28)
Where:
= the mean ( specified center of the distribution) = standard deviation ( spread of the distribution).

Statistical variations in the output of a single element with time- repeatability(contd)
Recall the general equation for the output of a measurement system (eqn(2.7))
A small deviation in the output O can be approximated as
(2.29)
Which means O is approximated by a linear combination of the deviations of the inputs, I,IM,II .
It can be shown that, if y is a linear combination of the independent variables x1, x2, x3, i.e.,
(2.30)

And if x1, x2, x3 have normal distributions with standard deviations 1, 2, 3, respectively, then the output will also have a normal distribution with standard deviation
(2.31)
From eqns[2.29] and [2.31] we see that the standard deviation of O, i.e. of O about mean(), is given by:
(2.32)

The corresponding mean of the element output is given by:

(2.33)
and the corresponding PDF is:

(2.34)
Reading Assignment:
Statistical variations amongst a batch of similar elementstolerance
2.5 Identification of static characteristics-Calibration

Is an experimental procedure used to determine most of the static performance parameters.
Figure 2.13 simplified traceability ladder
Figure 2.12 Calibration of an element
2.6 The Accuracy of measured system in the steady-state

Accuracy is quantified using measurement error E where: E = Measured value true Value = System output System input
2.6.1 Measurement error of a system of ideal elements

Let
I=I1
True Value
K1
1
O1=I2
K2
2
O2=I3
K3
3
O3
Ii
Ki
i
Oi
In
Kn
n
On=O
Measured Value
Then the overall output for the above system become:
If the measured system is complete, then
2.6.1 Measurement error of a system of ideal elements()
Thus if
We have E = 0 and the system is perfectly accurate!

Example 2.3 consider the simple temperature measurement system depicted below:
Thermocouple TT
0C
E(T) V e.m.f
Amplifier
True temperature
K1=40V/0C
K2= 1000V/V
V
volts
Indicator
K2= 250C/V
TM 0C Measured temperature
For this example,
2.6.1 Measurement error of a system of ideal elements()

However, in practice the system in Example 2.3 cant be accurate because of the following limiting factors.
the thermocouple is non-linear, and hence sensitivity is dependent on temperature to be measured.
the output voltage of the amplifier is also affected by changes in ambient temperature ( Chapter 4/ or Section 9 )
the sensitivity K3 of the indicator depends on the stiffness of the restoring spring in the moving coil assembly( affected by changes in environmental temperature and wear) In general the error of any measurement system depends on the non-ideal characteristics e.g. non-linearity, environmental & statistical effects of every elements in the system
2.6.2 The error PDF of a system of non-ideal elements

Objective: to calculate the system error PDF , p(E) ( using Eqns(2.32)-(2.34)
Table 2.1: General calculation of system p(E)

I=I1 1 O1=I2 2 O2 = I3 Ii i Oi In n On =O
Mean values of element outputs
(2.35)
:
:
:
:
2.6.2 The error PDF of a system of non-ideal elements()

Table 2.1: General calculation of system p(E) (contd) Mean value of system error
(2.36)
Standard deviation of element outputs
(2.37)

Table 2.1: General calculation of system p(E) (contd) Standard deviation of element outputs(contd)
(2.37)
Standard deviation of system error

(2.38)

Table 2.1: General calculation of system p(E) (contd) Error PDF
(2.39)
Standard deviation of system error

Example 2.4 For the temperature measurement system shown below calculate and

Table 2.2: Model for temp0 measurement system element(e.g.2.4) (a) Platinum resistance temp0 detector Model eqn. Individual mean values Individual stand. deviations
(between 100 to 1300C)
Partial derivatives
Overall mean value Overall standard deviation

Table 2.2: Model for temp0 measurement system element(e.g.2.4) (b) Current transmitter Model eqn.
4 to 20mA output for 138.5 to 149.8 input ( 100 to 1300C) Ta = deviations of ambient temperature from 200C
Individual mean values
Individual stand. deviations

Table 2.2: Model for temp0 measurement system element(e.g.2.4) (b) Current transmitter(contd) Partial derivatives Overall mean value Overall standard deviation

Table 2.2: Model for temp0 measurement system element(e.g.2.4) (b) Current transmitter Model eqn. Individual mean values Individual stand. deviations
(100 to 1300C record for 4 to 20mA input)
Partial derivatives
Overall mean value Overall standard deviation

Table 2.3: Summary of calculation of Mean and for Example 2.4)
Standard deviation
2.7 Error reduction techniques

Compensation methods for non-linear and environmental effects.
Compensating non-linear element
U(I)
Uncompensated Non-linear element
C(U)
compensated Non-linear element

Voltage ETh V
Temperature k Thermistor
Resistance R k
Deflection bridge
Figure 2.14 Compensating non-linear element
2.7 Error reduction techniques(contd)

Opposing environmental inputs e.g. Variations in temperature T2 of the reference junction of a thermocouple.
II
+ +
KI
KI
U
+
C=KI if KI=KI
Uncompensated element
Compensating element
(a) Using opposing environmental inputs

Opposing environmental inputs(contd)
(b) Using a differential system Figure 2.15 Compensation for interfering inputs

Isolation* Zero environmental sensitivity High-gain negative feedback*

Performance Characteristics of Measurement System

Uploaded by

Copyright:

Available Formats

You might also like

Performance Characteristics of Measurement System

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

Performance Characteristics of Measurement System

Uploaded by

Copyright:

Available Formats

ENGG3201-Measurement Technique

Static Characteristics - Range, Span, ,Sensitivity, Resolution, etc

Dynamic Characteristics -Rise time, settling time, MP, etc

2.1 Systematic Characteristics

Figure 2.1 Meaning of element Characteristics

Static(steady-state) Characteristics (contd)

Imin - minimum input

Imax - maximum input

Static(steady-state) Characteristics (contd)

Static(steady-state) Characteristics (contd)

Static(steady-state) Characteristics (contd)

Actual O(I) Imax I

Static(steady-state) Characteristics (contd)

Static(steady-state) Characteristics (contd)

Static(steady-state) Characteristics (contd)

Static(steady-state) Characteristics (contd)

Figure 2.4 Resolution & potentiometer example.

Static(steady-state) Characteristics (contd)

Figure 2.6 Modifying & Interfering Inputs

2.2 Generalized model of a system element

2.3 Dynamic Characteristics:

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

2.3.1 Transfer function G(s) for typical system elements

for the range 0 to 4000C.

2.3.1 Transfer function G(s) for typical system elements

Evaluation @ 1000C yields

(b) Using eqn.(2.16) , we get

2.3.1 Transfer function G(s) for typical system elements

Figure 2.8 Element model for dynamic calculations

2.3.2 Identification of the dynamics of an element

Unit step response of 2nd order element is found as:

Expressing (2.18) in partial fraction we get:

2.3.2 Identification of the dynamics of an element

Where And this gives

2.3.2 Identification of the dynamics of an element

Using Trigonometric relationship (refer the triangle above), we get (2.22)

2.3.2 Identification of the dynamics of an element

2.3.2 Identification of the dynamics of an element

2.3.2 Identification of the dynamics of an element

Figure 2.10 Frequency response of an element with linear dynamic

2.3.2 Identification of the dynamics of an element

2.3.2 Identification of the dynamics of an element

2.3.2 Identification of the dynamics of an element

And occurs at the resonant frequency R where:

(To verify eqns(2.27), first find

By measuring |G(j)|MAX and R ; and n can be found.

2.4 Statistical Characteristics

2.4 Statistical Characteristics(contd)

2.4 Statistical Characteristics(contd)

2.4 Statistical Characteristics(contd)

2.4 Statistical Characteristics(contd)

The corresponding mean of the element output is given by: