Professional Documents
Culture Documents
EE320 Electronic Notes
EE320 Electronic Notes
Lecture Notes
Keith W. Whites
Fall 2009
Course Notes:
EE 320 Electronics I
Table of Contents
Lecture
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Title
Ideal Diode.
Physical Operation of Diodes.
DC Analysis of Diode Circuits.
Small-Signal Diode Model and Its Application.
Introduction to B2 Spice from Beige Bag Software.
Zener Diodes.
Diode Rectifier Circuits (Half Cycle, Full Cycle, and Bridge).
Peak Rectifiers.
Limiting and Clamping Diode Circuits. Voltage Doubler. Special Diode Types.
Bipolar Junction Transistor Construction. NPN Physical Operation.
PNP Bipolar Junction Transistor Physical Operation. BJT Examples.
DC Analysis of BJT Circuits.
The BJT as a Signal Amplifier.
BJT Small-Signal Equivalent Circuit Models.
BJT Small-Signal Amplifier Examples.
Graphical Analysis of a BJT Small-Signal Amplifier.
BJT Biasing. Current Mirror.
Common Emitter Amplifier.
Common Emitter Amplifier with Emitter Degeneration.
Common Base Amplifier.
Common Collector (Emitter Follower) Amplifier.
BJT Internal Capacitances. High Frequency Circuit Model.
Common Emitter Amplifier Frequency Response. Millers Theorem.
BJT as an Electronic Switch.
Enhancement Type MOSFET Operation, P-Channel, and CMOS.
MOSFET Circuit Symbols, iD-vDS Characteristics.
MOSFET Circuits at DC.
MOSFET as an Amplifier. Small-Signal Equivalent Circuit Models.
MOSFET Small-Signal Amplifier Examples.
Biasing MOSFET Amplifiers. MOSFET Current Mirrors.
Common Source Amplifier.
Common Source Amplifier with Source Degeneration.
CMOS Common Source Amplifier.
MOSFET Common Gate Amplifier.
CMOS Common Gate Amplifier.
MOSFET Common Drain (Source Follower) Amplifier.
CMOS Digital Logic Inverter.
Whites, EE 320
Lecture 1
Page 1 of 7
(2)
Whites, EE 320
Lecture 1
Page 2 of 7
+
v
-
Capacitors: i = C
v
R
dv
.
dt
i
C
+
v
-
Ideal Diodes
You will now learn about a new electrical circuit element, the
diode. Diodes are made from two different types of
semiconducting materials that come together to form a
junction:
Whites, EE 320
Lecture 1
Page 3 of 7
It is apparent from this i-v characteristic curve that there are two
distinct regions of operation of the ideal diode:
v < 0 i = 0 . In this region, the diode is off.
i > 0 v = 0 . In this region, the diode is on.
The ideal diode acts as an electronic valve allowing current in
only one direction through the diode: in the direction of the
arrow in the circuit symbol.
Whites, EE 320
Lecture 1
Page 4 of 7
Applications of Diodes
Now we will briefly consider a couple of applications for
diodes. Well cover these in much more detail later.
Signal rectifier.
Think of this R as the Thevenin
resistance of additional circuitry
attached to the rectifier circuit.
+ vD +
vI
D
R
+
vO
-
Whites, EE 320
Lecture 1
Page 5 of 7
vI
A
t
vO
A
t
D1
D2
vY
R
Whites, EE 320
Lecture 1
Page 6 of 7
Whites, EE 320
Lecture 1
Page 7 of 7
D1
1V
D2
3V
V
I
1k
-5 V
Whites, EE 320
Lecture 2
Page 1 of 10
(Fig. 3.8)
The diode has three distinct regions of operation:
1. Forward bias note that when the diode is on, the voltage
drop is approximately 0.6 V to 0.7 V for a silicon diode.
2. Reverse bias in this region i = I S , where IS is called the
saturation current. For small signal diodes, IS is often on
the order of fA (10-15 A).
3. Breakdown in this region v VZK for all I, where VZK is
called the breakdown knee voltage. This region of operation
is useful in certain applications.
In the forward bias region of operation, it can be shown from
first principals that
2009 Keith W. Whites
Whites, EE 320
Lecture 2
nVv
T
i = I S e 1
Page 2 of 10
(3.1),(1)
where
n = emission constant. Typically between 1 and 2.
VT = kT q 25 mV at room temperature (20C). Called
the thermal voltage.
Notice the highly non-linear relationship between i and v in this
equation. (Youll learn where this mathematical expression
comes from in EE 362.)
When v << 0 in (1), then
nVv
i = I S e T 1 I S
(2)
PN Junction
Semiconductor junction diodes are made by joining two
semiconductors together. A pn junction diode is formed by
joining a p-type semiconductor to an n-type semiconductor:
Whites, EE 320
Lecture 2
p type
silicon
lattice
Page 3 of 10
n type
silicon
lattice
For a silicon diode, both the p and n regions are silicon, but in
each of these regions, small amounts of impurities have been
added through a process called doping.
To make p and n regions, we begin with a silicon crystal as
shown in Fig. 3.40. These atoms are held together by covalent
bonds (sharing pairs of electrons).
(Fig. 3.40)
At T = 0, the outermost electron (e-) of each atom is held in
covalent bonds. No current is possible since no electrons are
available to contribute to conduction.
Whites, EE 320
Lecture 2
Page 4 of 10
(Fig. 3.41)
Holes
When electrons are thermally excited out of covalent bonds,
they leave a vacancy at the bond site, as illustrated above in
Fig. 3.41. This is called a hole.
Interestingly, holes can also contribute to conduction current in a
semiconductor material (see the figure below). This movement
is usually much slower than e- so the mobility of holes is
smaller.
Whites, EE 320
Lecture 2
Page 5 of 10
Whites, EE 320
Lecture 2
Page 6 of 10
(Fig. 3.44)
(2) To create free electrons, add donor dopants (see Fig. 3.43).
For such n-type semiconductors, the silicon is doped with
pentavalent impurity elements such as phosphorus. These
impurities displace silicon atoms with phosphorous atoms
(having five electrons). Consequently, one extra electron is
available to move through the silicon lattice.
(Fig. 3.43)
Be aware that the entire p-type and n-type regions remain charge
neutral at all times! The dopant atoms are also charge neutral.
Whites, EE 320
Lecture 2
Page 7 of 10
Depletion Region
Something very special occurs when we place p-type material in
contact with n-type material. There now appears to be an
excess of holes in the p-type material and an excess of free
electrons in the n type.
p type
silicon
lattice
n type
silicon
lattice
Whites, EE 320
Lecture 2
Page 8 of 10
(1) Holes diffuse across the junction (diffuse because the hole
concentration is higher in p type) into the n-type region and
recombine with majority electrons.
Diffusion of
majority carrier.
Recombination
+
+ -
p type
n type
p type
++
++
++
++
++
+ + n type
++
Whites, EE 320
Lecture 2
Page 9 of 10
E-
+++
+++
+++
+++
+++
+++
+++
width of depletion
region increases
V
Whites, EE 320
Lecture 2
Page 10 of 10
+
p
-
E -
+
+
+
+
+
+
+
+
n
-
width of depletion
region decreases
V
Whites, EE 320
Lecture 3
Page 1 of 10
T
I = I S e 1
(3.1),(2)
T
VD = VDD I S R e 1
(3)
Whites, EE 320
Lecture 3
Page 2 of 10
Slope = -1/R
v
VDD
Whites, EE 320
Lecture 3
Page 3 of 10
(Fig. 3.11)
The point where these two curves intersect is the
simultaneous solution to the two equations (1) and (2).
This graphical method is an impractical solution method for
all but the simplest circuits. However, it is useful for a
qualitative understanding of these circuits. For example, what
happens when:
(a) VDD increases?
VDD
(b) R increases?
Whites, EE 320
Lecture 3
Page 4 of 10
V_DC
SRC1
Vdc=2.0 V
R
R1
R=1 kOhm
1.43 mA
ap_dio_1N4148_1_19930601
D1
Whites, EE 320
Lecture 3
Page 5 of 10
Whites, EE 320
Lecture 3
Page 6 of 10
(Fig. 3.15)
In words, this model says that if the diode is forward biased,
then the voltage drop across the diode is VD. If not forward
biased, the diode is then reversed biased and the current is zero
and VD can be any value < VD.
VD is often set to 0.7 V for silicon diodes, as shown above, while
set to 0.2 V for Schottky diodes, for example.
The CVD circuit model for diodes is
(Fig. 3.16b)
This is probably the most commonly used diode model for hand
calculations.
Whites, EE 320
Lecture 3
Page 7 of 10
Whites, EE 320
Lecture 3
Page 8 of 10
(Fig. 3.12)
The finite slope to this curve means that the diode has a nonzero internal resistance, which we will label as rD. The
equivalent circuit for the PWL diode model is then
(Fig. 3.13b)
Whites, EE 320
Lecture 3
Page 9 of 10
From Fig. 3.12, we can determine VD0 and rD for the particular
diode whose characteristic equation is shown:
VD0 = 0.65 V
run 0.9 0.65
rD =
=
= 20.8 .
rise 12 103
The equivalent circuit using the PWL model of the diode is then
I
R=1 k
+
VDD=
2V
VD
-
+
Ideal VIdeal
VD0=0.65 V
rD=20.8
I = 1.32 mA.
Whites, EE 320
Lecture 3
Page 10 of 10
current indicates that we made the correct choice that the diode
is on.
Whats the forward voltage drop across the diode?
VD = 2 1000 1.32 103 = 0.68 V
Is this enough to turn the diode on? Yes, referring to the
equivalent circuit above
VIdeal = VD 0.65 20.8 1.32 103 0+ V
Whites, EE 320
Lecture 4
Page 1 of 9
Whites, EE 320
Lecture 4
Page 2 of 9
For example:
(Fig. 3.17a)
where vd(t) is some time varying waveform, perhaps periodic
such as a sinusoid or triangle signal.
The purpose of VD in this circuit is to set the operation of the
diode about a point on the forward bias i-v characteristic curve
of the diode. This is called the quiescent point, or Q point, and
the process of setting these DC values is called biasing the
diode.
(Fig. 3.17b)
Whites, EE 320
Lecture 4
Page 3 of 9
VD
nVT
vd ( t )
iD ( t ) I S e nVT = I S e e nVT
N
=ID
vd ( t )
or
iD ( t ) = I D e nVT
(3.12),(2)
Whites, EE 320
Lecture 4
Page 4 of 9
So, if vd(t) is small enough we can see from this last equation
that iD is the sum (or superposition) of two components: DC and
AC signals. What weve done is to linearize the problem by
limiting the AC portion of vD to small values.
The term nVT I D has units of ohms. It is called the diode smallsignal resistance:
nV
rd T []
(3.18),(5)
ID
From a physical viewpoint, rd is the inverse slope of the tangent
line at a particular bias point along the characteristic curve of the
diode. Note that rd changes depending on the (DC) bias:
Whites, EE 320
Lecture 4
Page 5 of 9
vs
-
DC (bias)
ID
VDD
id
+
VDD
+
vD=VD+vd
-
VD
-
Ideal
VD0
R
+
+
rd
vs
-
vd =
-
rd
vs
rd + R
rD
AC only: rides on VD.
Whites, EE 320
Lecture 4
Page 6 of 9
Example N4.1 (Text example 3.6). For the circuit shown below,
determine vD when V + = 10 + 1 cos ( 2 60t ) V.
V+
2V
10 V
Called ripple if one
desires purely DC.
t
T=1/f=1/60 s
Whites, EE 320
Lecture 4
Page 7 of 9
Whites, EE 320
Lecture 4
Page 8 of 9
p
-
+
+
+
+
+
+
+
n
-
Cj
VS
vs
Whites, EE 320
Lecture 4
Page 9 of 9
Cj
Cd
Whites, EE 320
Lecture 5
Page 1 of 10
Whites, EE 320
Lecture 5
Page 2 of 10
Whites, EE 320
Lecture 5
Page 3 of 10
You can double click the component to change its values, such
as the resistance of the resistor in this case. However, the default
value of 1 k is the value we need here.
To add the DC voltage source, we select Devices | Voltage
Source (V) then click once to place the voltage source on the
schematic. Selecting Ctrl + r once rotates the source vertically
giving:
Whites, EE 320
Lecture 5
Page 4 of 10
Next, well add the diode. In this case, well add the generic
diode available in B2 Spice in Devices | Diode (D):
Whites, EE 320
Lecture 5
Page 5 of 10
only two: the generic diode and the 1N4007 diode, which can be
found in Categories | Diode .
The next steps are to connect the circuit components together
using the wire tool and then add a circuit ground from Devices |
Ground (0). The wire tool is accessed from the toolbar as
illustrated below:
Whites, EE 320
Lecture 5
Page 6 of 10
Whites, EE 320
Lecture 5
Page 7 of 10
Whites, EE 320
Lecture 5
Page 8 of 10
Whites, EE 320
Lecture 5
Page 9 of 10
Whites, EE 320
Lecture 5
Page 10 of 10
Whites, EE 320
Lecture 6
Page 1 of 11
Whites, EE 320
Lecture 6
Page 2 of 11
Provided that the power dissipated in the diode is less than the
maximum rated, the diode is not damaged when operating in the
breakdown region. In fact, Zener diodes are designed to operate
in this region.
The circuit symbol for the Zener diode is
(Fig. 3.20)
These diodes are usually operated in the reverse bias regime
(i.e., breakdown region) so that IZ > 0 and VZ > 0.
An enlargement of this breakdown region is shown in text
Figure 3.21:
(Fig. 3.21)
Whites, EE 320
Lecture 6
Page 3 of 11
The manufacturer specifies the VZ0 and test current IZT. One can
design Zeners with a wide range of voltages.
The rated VZ at the specified IZT is listed for these Zener diodes.
The circled component, for example, has VZ = 8.2 V at IZT = 31
mA. The maximum rated power is 1 W for this device.
Whites, EE 320
Lecture 6
Page 4 of 11
As the current deviates from the specified value IZT, the voltage
VZ also changes, though perhaps only by a small amount. The
change in voltage VZ is related to the change in the current IZ
as
VZ = rz I Z
(1)
where rz is the incremental or dynamic resistance at the Q point
and is usually a few Ohms to tens of Ohms. See the datasheet for
the particular device you are working with.
Because of the nearly linear relationships in the breakdown
region, the reverse bias model of the Zener diode is
(Fig. 3.22)
VZ = VZ 0 + rz I Z
where
as is apparent from Fig. 3.21.
(3.20),(2)
Whites, EE 320
Lecture 6
Page 5 of 11
Whites, EE 320
Lecture 6
Page 6 of 11
(Fig. 3.23a)
With these ratings
VZ = VZ 0 + rz I Z VZ 0 = VZ rz I Z
or
VZ 0 = 6.8 20 5 103 = 6.7 V
Note that the supply voltage can fluctuate by 1 V. Imagine this
fluctuation is a random process rather than a time periodic
variation.
Determine the following quantities:
(b) Find VO with no load and V+ at the nominal value. The
equivalent circuit for the reverse bias operation of the
Zener diode is
Whites, EE 320
Lecture 6
Page 7 of 11
or
Whites, EE 320
Lecture 6
Page 8 of 11
then
IL =
6.8
= 3.4 mA.
2000
Whites, EE 320
Lecture 6
Page 9 of 11
VO = 20 ( 3.4 103 ) = 68 mV
The ratio of the change in output voltage to the change in
the load current ( VO I L ) is called the load regulation of
the regulator circuit. Its often expressed in units of
mV/mA. For this example,
V
mV
77 mV
= 22.6
Load Regulation O =
I L 3.4 mA
mA
Whites, EE 320
Lecture 6
Page 10 of 11
10 V
500
+
VO
-
Whites, EE 320
Lecture 6
Page 11 of 11
If V+ = 9 V:
9 6.7
= 4.6 mA.
500
Therefore, IL = 4.6 mA-0.2 mA = 4.4 mA, so that
V
6.7
RL = L =
= 1,522
I L 4.4 103
IS =
If V+ = 11 V:
11 6.7
= 8.6 mA.
500
Therefore, IL = 8.6 mA-0.2 mA = 8.4 mA, so that
V
6.7
RL = L =
= 798
3
I L 8.4 10
IS =
Whites, EE 320
Lecture 7
Page 1 of 9
Whites, EE 320
Lecture 7
Page 2 of 9
N2
vp
(1)
N1
This occurs even though there is no direct contact between the
input and output sections. This magic is described by
Faradays law:
d
E
dl
B ds
=
v
dt
C(S )
S (C )
vs =
or
emf =
d m
dt
(2)
=
or 8 :1 ratio ( N1 : N 2 ).
N1 120 8
We choose vs 15 VDC for a margin.
Whites, EE 320
Lecture 7
Page 3 of 9
Diode Rectification
We will discuss three methods for diode rectification:
1. Half-cycle rectification.
2. Full-cycle rectification.
3. Bridge rectification. (This is probably the most widely
used.)
1. Half-Cycle Rectification
Weve actually already seen this circuit before in this class!
(Fig. 3.25a)
We will use the PWL model for the diode to construct the
equivalent circuit for the rectifier:
Whites, EE 320
Lecture 7
Page 4 of 9
(Fig. 3.25b)
From this circuit, the output voltage will be zero if vS ( t ) < VD 0 .
Conversely, if vS ( t ) > VD 0 we can determine vO by superposition
of the two sources (DC and AC) in the circuit sources since we
have linearized the diode:
R
DC: vO ( t ) = VD 0
R + rD
R
AC: vO ( t ) = vS ( t )
R + rD
Notice that were not making a small AC signal assumption
here. Rather, we have used the assumption of the PWL model to
completely linearize this problem when vS ( t ) > VD 0 and then
used superposition of the two sources, which just happen to be
DC and AC sources. (Consequently, we should not use rd here.)
The total voltage is the sum of the DC and AC components:
R
vS ( t ) > VD 0
vO ( t ) = vO ( t ) + vO ( t ) = vS ( t ) VD 0
R + rD
(3.21),(3)
For vS ( t ) < VD 0 , vO ( t ) = 0 .
Whites, EE 320
Lecture 7
Page 5 of 9
(Fig. 3.25d)
There are two important device parameters that must be
considered when selecting rectifier diodes:
1. Diode current carrying capacity.
2. Peak inverse voltage (PIV). This is the largest reverse
voltage across the diode. The diode must be able to
withstand this voltage without shifting into breakdown.
For the half-cycle rectifier with a periodic waveform input
having a zero average value
PIV = Vs
(5)
where Vs is the amplitude of vS.
Whites, EE 320
Lecture 7
Page 6 of 9
2. Full-Cycle Rectification
One disadvantage of half-cycle rectification is that one half of
the source waveform is not utilized. No power from the source
will be converted to DC during these half cycles when the input
waveform is negative in Fig. 3.25d.
The full-cycle rectifier, on the other hand, utilizes both the
positive and negative portions of the input waveform. An
example of a full-cycle rectifier circuit is:
(Fig. 3.26a)
Notice that the transformer has a center tap that is connected to
ground.
On the positive half of the input cycle vS > 0 , which implies that
D1 is on and D2 is off. Conversely, on the negative half of
the input cycle, vS < 0 which implies that D1 is off and D2 is
on.
Whites, EE 320
Lecture 7
Page 7 of 9
(Fig. 3.26c)
While this full-cycle rectifier is a big improvement over the halfcycle, there are a couple of disadvantages:
PIV = 2Vs VD 0 , which is about twice that of the half-cycle
rectifier. This fact may require expensive or hard-to-find
diodes.
Requires twice as many transformer windings on the
secondary as does the half-cycle rectifier.
3. Bridge Rectification
The bridge rectifier uses four diodes connected in the famous
bridge pattern:
Whites, EE 320
Lecture 7
Page 8 of 9
Is
+
vp
vs
D4
D1
- vO +
D2
D3
Is
D3
Whites, EE 320
Lecture 7
Page 9 of 9
(Fig. 3.27b)
The bridge rectifier is the most popular rectifier circuit.
Advantages include:
PIV = Vs VD 0 , which is approximately the same as the
half-cycle rectifier.
No center tapped transformer is required, as with the halfcycle rectifier.
Whites, EE 320
Lecture 8
Page 1 of 10
(Fig. 3.29a)
We expect that as soon as we turn on the source, the capacitor
will charge up on + cycles of vI and discharge on the -
cycles.
To smooth out the voltage, we need this discharge to occur
slowly in time. This means we need to choose C large enough to
make this happen, presuming that R is a given quantity (the
Thvenin resistance of the rest of the circuit).
Whites, EE 320
Lecture 8
Page 2 of 10
(Fig. 3.29)
Notice the diode current and the capacitor voltage. They display
behavior much different than what one would find in an AC
circuit.
Whites, EE 320
Lecture 8
Page 3 of 10
td
vI
Diode on
Not sketched to scale.
(1)
At the end of the discharge time, td, the output voltage equals
vO ( td ) = V p Vr
(2)
Substituting for vO from (1) at this time td leads to
Vr
V p e td = V p Vr or
= 1 e td
Vp
(3)
Whites, EE 320
Lecture 8
Page 4 of 10
(5)
( T )
(7)
Vp
This simple equation gives the ratio of the ripple voltage to the
peak voltage of the input sinusoidal signal for the half-cycle
rectifier. Its worth memorizing, or knowing how to derive.
Often R and T are fixed quantities. So from (7)
T
Vr V p
( T )
(3.28),(8)
RC
to obtain a small ripple voltage we need a large C in this case.
Whites, EE 320
Lecture 8
Page 5 of 10
Conduction Interval
Lastly, the conduction interval t is defined as the time interval
in which the diode is actually conducting current. This time
period is sketched in the preceding two figures.
The diode conducts current beginning at time td and ending at T,
within each period. Using equation (4) at time td
V p cos (T td ) = V p Vr or V p cos (t ) = V p Vr (9)
We expect the conduction interval to be small. So truncating the
series expansion of cosine to two terms, (9) gives
2Vr
t
(3.30),(10)
Vp
The factor t is sometimes called the conduction angle, . For
Vr V p this conduction angle (and conduction interval) will be
small, as expected.
Discussion
To reiterate, the objective of the peak rectifier is to charge the
shunt C when D is on, and slowly discharge it during those
times when D is off.
Whites, EE 320
Lecture 8
Page 6 of 10
R
Vr
Whites, EE 320
iD
or
Lecture 8
max
Page 7 of 10
Vp
2V p
100
2 100
1
2
1 + 2
+
=
R
Vr 10,000
2
iD max 638 mA.
Whites, EE 320
Lecture 8
Page 8 of 10
vO
Vr
Vp
vI
t
t
Diode on
Whites, EE 320
Lecture 8
Page 9 of 10
The output voltage has less ripple than from a half-cycle peak
rectifier (actually one half less ripple).
vO
vI
t
T
( T )
(3.33),(13)
V p 2
Whites, EE 320
Lecture 8
Page 10 of 10
Lastly, it can be shown that the iD max for the full-cycle peak
rectifier:
Vp
Vp
(3.35),(14)
iD max 1 + 2
[A]
2Vr
R
is approximately one-half that of the half-cycle peak rectifier
when Vr V p .
Whites, EE 320
Lecture 9
Page 1 of 8
Whites, EE 320
Lecture 9
Page 2 of 8
Then one would see this output voltage vO for this particular
input voltage vI:
vO
vI
-1
-3
Ideal
vI
5V
+
Ideal vO
Whites, EE 320
Lecture 9
Page 3 of 8
(Fig. 3.36b)
There are three important things to note about this circuit:
1. The ideal D keeps vO 0 .
2. C charges only when vI < 0 . Without a load, there is no
other path for current.
3. The vC polarity is positive as shown above.
With these insights, lets look at a specific example to illustrate
the operation of this circuit. Consider this input voltage:
(Fig. 3.36a)
C in Fig. 3.36b will eventually charge completely so that vC = +6
V. In that case, the lowest output voltage will be clamped to
zero. The output voltage will appear as:
Whites, EE 320
Lecture 9
Page 4 of 8
(Fig. 3.36c)
Hence, this is called a clamped capacitor circuit. Without the
diode present in this circuit, the capacitor would not retain any
net charge per period so it would never charge up to 6 V.
Note that here we are looking at the steady state response. It
may take a few periods for the capacitor to completely charge.
Were not looking at the transient response.
There are two applications of the clamped capacitor circuit
discussed in the text.
(a) Pulse width modulation detector. PWM is used for
motor speed control, for example. The width of the pulse
contains the information.
Whites, EE 320
Lecture 9
Page 5 of 8
C1
Vpcos( t)
D1
-
D2
+
vD1 C2
-
+
vO
-
Clamped
capacitor
Half-cycle peak
rectifier
vD1
Whites, EE 320
Lecture 9
Page 6 of 8
Whites, EE 320
Lecture 9
Page 7 of 8
+++
+++
+++
+++
+++
+++
+++
Whites, EE 320
Lecture 9
Page 8 of 8
+
+
+
+
+
+
+
Whites, EE 320
Lecture 10
Page 1 of 9
Whites, EE 320
Lecture 10
Page 2 of 9
(Fig. 5.1)
As can be seen, the BJT is formed from two back-to-back pn
junctions:
Emitter-base junction (EBJ)
Collector-base junction (CBJ).
This specific way of drawing the BJT has been around from the
very beginning of these transistors. This figure:
Whites, EE 320
Lecture 10
Page 3 of 9
BJTs can also be fabricated from two p-type regions and one ntype. This is called a pnp transistor:
(Fig. 5.2)
While the BJT might appear to be symmetrical by looking at
Fig. 5.1, the actual devices are not.
For example, the cross section below of an npn transistor clearly
shows that the EBJ and CBJ, for example, have very differently
sized surface contact areas, which will greatly change their
relative behaviors.
(Fig. 5.6)
There are four basic modes of operation for a BJT depending on
the states of the two pn junctions of the transistor:
Whites, EE 320
Lecture 10
Mode
Cutoff
Active
Saturation
Reverse Active
Emitter-Base Jct.
Reverse
Forward
Forward
Reverse
Page 4 of 9
Collector-Base Jct.
Reverse
Reverse
Forward
Forward
Whites, EE 320
Lecture 10
Page 5 of 9
the CBJ (notice that the emitter and collectors have swapped
positions from Fig. 5.1 shown earlier):
(Fig. 1)
The overall objective of this circuit is to create a current flowing
from the collector to the emitter terminals in the transistor that is
controlled, so to speak, by the base voltage VBB.
How does this transistor operate in this circuit?
Because of the forward bias on the EBJ, charges can flow
across this junction giving rise to iE. This current is
primarily electrons that are injected from n to p.
The electrons injected in the base diffuse across the thin
base region towards the collector. Some of the e- recombine
in the base, but this region is manufactured to be thin and
lightly doped compared to the emitter so this recombination
is kept small. Otherwise, the BJT would just operate as two
back-to-back diodes and no current would flow.
Whites, EE 320
Lecture 10
Page 6 of 9
(Fig. 5.4)
The e- that reach the reverse-biased CBJ encounter a large
electric field. This E sweeps them into the collector
forming the collector current iC as shown in Fig. 1 above.
A small base current iB is present largely due to
recombination in the base with the small amount of injected
holes from the base to the emitter. This is an important
current, though.
Whites, EE 320
Lecture 10
Page 7 of 9
iC
or iC = iE
(5.16),(1)
iE
Typically it has values of near 0.99. Note that is called the
common-base current gain in the text.
(5.13),(2)
(3)
=
iB 1
Equating this to (4) we find
1
and solving this equation for
=
+1
=
(5.19),(5)
(5.17),(6)
Whites, EE 320
Lecture 10
Page 8 of 9
Whites, EE 320
Lecture 10
(Fig. 5.14a)
Page 9 of 9
Whites, EE 320
Lecture 11
Page 1 of 8
(Fig. 5.11)
Differences between pnp and npn BJTs are:
Biasing voltages are applied oppositely to the npn, though
still forward biasing EBJ and reverse biasing the CBJ for
active mode operation, for example.
Current is primarily composed of holes (in the p type
regions) rather than electrons as in the npn BJT.
The current direction conventions are iE into the emitter
while iC and iB are out from the device.
The circuit symbol for the pnp BJT is
Whites, EE 320
Lecture 11
Page 2 of 8
Once again, the filled arrow is always located on the emitter and
helps us to remember the direction of the emitter current. Notice
that the currents are pointed in opposite directions compared to
the npn BJT.
For biasing in the active mode, a possible circuit is
(Fig. 5.14b)
As with the npn, for the pnp BJT in the active mode and with the
current convention shown above
iC = iE
(5.16),(1)
iB = (1 ) iE
iC = iB
=
+1
=
(2)
(5.10),(3)
(5.19),(4)
(5.17),(5)
Whites, EE 320
Lecture 11
Page 3 of 8
Examples
Well now consider a few examples of the DC analysis of npn
and pnp BJT circuits.
(Fig. 5.15)
The design of this circuit is to determine the RC and RE that
provide the specified IC and VC.
For IC = 2 mA, then
15 VC
15 5
= 5 k.
= 2 mA or RC =
3
2 10
RC
Whites, EE 320
Lecture 11
Page 4 of 8
Were assuming that the transistor is in the active mode with the
EBJ forward biased and the CBJ reversed biased.
For the forward biased EBJ junction,
vBE
VT
iC = I S e
(5.3),(6)
Its given that at IC = 1 mA, VBE = 0.7 V. What is VBE when IC = 2
mA? Using (6) for two different iC and vBE we find that
vBE 1 vBE 2
i
iC1
v v
= e VT
or BE1 BE 2 = ln C1
iC 2
VT
iC 2
Therefore,
i
vBE 2 = vBE1 + VT ln C1
(7)
iC 2
For this particular case,
2
VBE 2 = 0.7 + 25 103 ln = 0.717 V
1
This is not much of an increase from 0.7 V, which is what we
typically observe when the BJT is in the active mode.
VE = 0.717 V
Consequently,
Next,
then
iC = iE iE =
IE =
iC
+1
i
C
100 + 1
2 mA or I E = 2.02 mA
100
Whites, EE 320
Lecture 11
Page 5 of 8
We can use this emitter current to select the proper resistor RE:
V ( 15 V )
IE = E
RE
0.717 + 15
or
RE =
= 7.07 k
2.02 103
Example N11.2 (text exercise 5.10). Determine IE, IB, IC, and VC
in the circuit below if = 50 and VE = -0.7 V.
(Fig. E5.10)
Whites, EE 320
Lecture 11
Page 6 of 8
IE =
(ii) Compute IC.
IC = I E =
0.7 ( 10 )
= 0.93 mA
10,000
+1
IE =
50
0.93 mA=0.91 mA
51
IC
0.91 mA
= 18.2 A
50
VC = 10 5,000 I C = 5.45 V
Whites, EE 320
Lecture 11
Page 7 of 8
(Fig. E5.11)
Because VEB = VE VB = 0.7 V, the pnp transistor may be
operating in the active mode, which is what we will assume.
(i) Determine and . Well use the relationships iC = iE and
iC = iB to determine and .
VB
1.0
=
= 10 A
100 103 100 103
10 1.7
IE =
= 1.66 mA
5,000
IB =
Using KCL:
I C = I E I B = 1.66 103 10 106 = 1.65 mA.
Therefore,
Whites, EE 320
Lecture 11
Page 8 of 8
I C 1.65 103
= =
= 165
6
IB
10 10
I C 1.65 103
= =
= 0.994
I E 1.66 103
and
Alternatively,
+1
= 0.994
or
Note that this VC means that the CBJ is reversed biased by the
voltage 1.0 ( 1.75 ) = 2.75 V. Hence, the active mode
operation for the pnp BJT is the proper assumption since weve
already determined that the EBJ is forward biased.
Whites, EE 320
Lecture 12
Page 1 of 9
(Fig. 5.34a)
Well assume the device is operating in the active mode, then
well check this assumption at the end of the problem by
calculating the bias of the EBJ and CBJ.
If the BJT is in the active mode, VBE = 0.7 V then
V
3.3
VE = 4 VBE = 3.3 V and I E = E =
= 1 mA.
3
RE 3.3 10
With I C = I E then
2009 Keith W. Whites
Whites, EE 320
Lecture 12
IC =
+1
Page 2 of 9
1 mA=0.99 mA
(Fig. 5.34c)
Whites, EE 320
Lecture 12
Page 3 of 9
10 I C 4.7 k = 2.57 V
6 0.7 = 5.3 V
IE =
5.3 V
= 1.6 mA
3.3 k
Whites, EE 320
Lecture 12
IC =
Page 4 of 9
10 5.5
=0.96 mA
4700
VCE
sat
= 0.2 V
6 0.7 = 5.3 V
IE =
5.3 V
= 1.6 mA
3.3 k
Notice that
I C 0.96
=
= 1.5
I B 0.64
This ratio is often called forced . Observe that its not equal to
100, as this ratio would be if the transistor were operating in the
active mode (see Section 5.3.4).
Whites, EE 320
Lecture 12
IE =
I B = I E IC
= 0.05 mA
Page 5 of 9
10 0.7
=4.65 mA
2000
0.7 V
10 + 4.6 mA 1 k = 5.4 V
Whites, EE 320
Lecture 12
Page 6 of 9
I E = 4.65 mA
0.7 V
I C = I E = 4.6 mA
10 + RC I C = 0
or
10
10
=
= 2,174
3
I C 4.6 10
This value of RC and smaller is required for the BJT to operate
in the active mode.
RC =
Whites, EE 320
Lecture 12
Page 7 of 9
100 k
+
50 k
RTH
VTH
-
Whites, EE 320
Lecture 12
Page 8 of 9
15 V
IC
5k
5V
VC
33.3 k
IB
VB
VE
KVL
3k
IE
To find the emitter current, well apply KVL over the loop
shown giving
5 = 33.3 103 I B + 0.7 + 3,000 I E
The quantity of interest is IB. With I C = I B and I C = I E for a
BJT in the active mode, we find
I
IE = C = IB =
IB
( + 1)
I E = ( + 1) I B
or
Using this in the KVL equation
5 0.7 = 33.3 103 + 3,000 ( + 1) I B
With = 100 then solving this equation we find
I B = 12.8 A I E = ( + 1) I B = 1.29 mA.
Next, by KCL
I C = I E I B = 1.29 m 12.8 A = 1.28 mA
The node voltages are then
Whites, EE 320
Lecture 12
Page 9 of 9
VC = 15 I C 5 k = 8.6 V
VE = I E 3 k = 3.87 V
VB = 5 I B 33.3 k = 4.57 V
Whites, EE 320
Lecture 13
Page 1 of 6
(Fig. 5.48a)
The DC voltages provide the biasing. The input signal is vbe and
the output signal is vc.
Whites, EE 320
Lecture 13
Page 2 of 6
(1)
AC
(2)
IC
or using (5.53)
iC = I C evbe VT
(5.82),(3)
For small vbe such that vbe 2VT (i.e., the small-signal
approximation), then (3) can be approximated by
v
I
iC I C 1 + be = I C + C vbe
(5.84),(4)
N
T
VT DC V
N
AC
IC
vbe
VT
(5.85),(5)
ic = g m vbe
(5.86),(6)
ic =
can be written as
Whites, EE 320
Lecture 13
Page 3 of 6
IC
[S]
(5.87),(7)
VT
is defined as the transistor small-signal transconductance. Its
units are Siemens. Note that g m I C .
gm
where
(Fig. 5.49)
With iC = I S evBE
vT
Whites, EE 320
Lecture 13
iC
I
= S evBE
vBE VT
Page 4 of 6
iC
(3) VT
(9)
IC
VT
(10)
=
N
VT
Therefore
gm =
iC
vBE
=
iC = I C
as we defined in (6).
Observe that:
= VCC I C RC ic RC
=VC
or
vC = VC ic RC
N N
DC
AC
(5.101),(11)
Whites, EE 320
Lecture 13
Page 5 of 6
Output (vC):
VC
Input (vBE):
vbe
VBE
t
Whites, EE 320
Lecture 13
Page 6 of 6
Whites, EE 320
Lecture 14
Page 1 of 9
rd =
nVT
ID
Whites, EE 320
Lecture 14
Page 2 of 9
Determine r
Assuming the transistor in this circuit
(Fig. 5.48a)
is operating in the active mode, then
i
1
IC
iB = C =
I
vbe
+
N
NC
(5.84) DC V
T
N
AC
I
g
so that
ib = C vbe = m vbe
VT
(1)
(5.90),(5.91),(2)
Whites, EE 320
Lecture 14
Page 3 of 9
(Fig. 1)
Notice the AC ground in the circuit. This is an extremely
important concept. Since the voltage at this terminal is held
constant at VCC, there is no time variation of the voltage.
Consequently, we can set this terminal to be an AC ground in
the small-signal circuit.
For AC grounds, we kill the DC sources at that terminal: short
circuit voltage sources and open circuit current sources.
So, from the small-signal equivalent circuit above:
(Fig. 2)
we see that
Whites, EE 320
Lecture 14
Page 4 of 9
vbe
ib
(5.92),(3)
r =
Hence, using (2) in (3)
r =
[]
(5.93),(4)
gm
This r is the BJT active mode small-signal input resistance of
the BJT between the base and the emitter as seen looking into
the base terminal. (Similar to a Thvenin resistance, this
statement means we are fictitiously separating the source from
the base of the BJT and observing the input resistance, as
indicated by the dashed line in Fig. 2.)
Determine re
Well determine re following a similar procedure as for r, but
beginning with
i
I
i
(5)
iE = C = C + c
N
N
DC
AC
(5.96),(6)
or with I E = I C ,
ie =
IE
vbe
VT
(5.96),(7)
Whites, EE 320
Lecture 14
Page 5 of 9
ve
(8)
ie
Assuming an ideal signal voltage source, then ve = vbe and
v
re be
(5.97),(9)
ie
Using (7) in this equation we find
V
(5.98),(10)
re = T
IE
But from (5.87)
I
I
V
gm = C = E T =
I E gm
VT
VT
re
Whites, EE 320
Lecture 14
Page 6 of 9
1
[]
(5.99),(11)
gm gm
This is the BJT active mode small-signal resistance between the
base and emitter seen looking into the emitter.
re =
r = ( + 1) re []
(5.100),(12)
Hybrid- Model
Version A.
Whites, EE 320
Lecture 14
Page 7 of 9
(Fig. 5.51a)
Lets verify that this circuit incorporates all of the necessary
small-signal characteristics of the BJT:
9 ib = vbe r as required by (3).
9 ic = g m vbe as required by (5.86), which we saw in the last
lecture.
9 ib + ic = ie as required by KCL.
We can also show from these relationships that ie = vbe re .
(Fig. 5.51b)
Whites, EE 320
Lecture 14
Page 8 of 9
T Model
The hybrid- model is definitely the most popular small-signal
model for the BJT. The alternative is the T model, which is
useful in certain situations.
The T model also has two versions:
Version A.
(Fig. 5.52a)
Version B.
Whites, EE 320
Lecture 14
Page 9 of 9
(Fig. 5.52b)
The small-signal models for pnp BJTs are identically the same
as those shown here for the npn transistors. It is important to
note that there is no change in any polarities (voltage or current)
for the pnp models relative to the npn models. Again, these
small-signal models are identically the same.
Whites, EE 320
Lecture 15
Page 1 of 10
r =
gm
, or
(5.93),(2)
VT
(5.99),(3)
gm I E
3. Rewrite the small-signal circuit: short out DC sources and
open DC current sources. Use the small-signal model for
the BJT.
4. Analyze the small-signal circuit for the desired quantities
such as voltage, small-signal voltage gain, etc.
re =
Whites, EE 320
Lecture 15
Page 2 of 10
(Fig. 5.53a)
The first step in the solution is to determine the Q point through
DC analysis. By superposition, well force vi = 0 for this
analysis.
Assuming the BJT is in the active mode, the results of the DC
analysis are:
(Fig. 5.53b)
We see that the CBJ is reversed biased so this npn BJT is in the
active mode because of this and the EBJ is forward biased.
Next, we determine the BJT small-signal model parameters for
the hybrid- model:
Whites, EE 320
Lecture 15
Page 3 of 10
I C 2.3 103
From (1), g m =
=
= 0.092 S
3
VT 25 10
100
From (2), r =
=
= 1,087
g m 0.092
Now, we insert a small-signal equivalent model of the BJT into
the circuit of Fig. 5.53(a) after shorting the DC voltage sources
(VBB and VCC). This gives the small-signal equivalent circuit:
(Fig. 5.53c)
Notice the AC ground at RC. This is an AC ground because
the voltage at this node does not vary with time. For the
purposes of the AC signal analysis, we can set this node to an
AC ground. (As a side note, in the lab power supplies have a
finite internal resistance. This Thvenin equivalent resistance
must be included in the AC circuit for analysis purposes.)
Next, we perform the small-signal analysis referring to Fig.
5.53c. At the input
r
(5.105),(4)
vbe =
vi
r + RBB
while at the output
vo = RC ic = RC g m vbe
(5)
Substituting for vbe from (4) gives
Whites, EE 320
Lecture 15
r
vi
r + RBB
Therefore, the small-signal AC voltage gain, Av, is
v
r
Av = o = RC g m
vi
r + RBB
vo = RC g m
Page 4 of 10
(6)
(7)
Av = 3,000 0.092
or
Whites, EE 320
Lecture 15
Page 5 of 10
Whites, EE 320
Lecture 15
Page 6 of 10
Whites, EE 320
Lecture 15
Page 7 of 10
(Fig. 5.55a)
The two capacitors in this circuit serve as DC blocks. They have
a large enough C so that X C 0 at the operating frequency.
With these capacitors, the DC bias is unchanged by the source or
load attachments. We call this capacitively coupled input and
output.
As always, we first determine the DC bias. Well assume the
BJT is in the active mode and that = 100:
(Fig. 5.55b)
From this result
Whites, EE 320
Lecture 15
Page 8 of 10
I C = 0.92 mA VC = 10 + I C RC = 5.4 V
Since VC < VB (and VEB = 0.7 V), the pnp BJT is operating in the
active mode.
(Fig. 5.55c)
Notice the two AC grounds in this circuit: one at RE and the
other at RC.
Also notice this is the first small-signal model of the pnp
transistor we have used. The small-signal model of the pnp
transistor is exactly the same as that for the npn with no change
in the polarities of the currents or voltages. This can be a little
confusing. Here, for example, ie is a negative quantity.
Using (3) for the small-signal equivalent model of the BJT
VT
25 103
re =
=
= 26.9
3
I E 0.93 10
Whites, EE 320
Lecture 15
Page 9 of 10
(8)
(9)
(10)
Whites, EE 320
Lecture 15
Page 10 of 10
Were assuming the output remains linear and the BJT in the
active mode at all times for the entire voltage swing in vC.
If this input voltage were set to a larger value, this would no
longer be the case and the BJT would first encounter nonlinear
behavior and eventually saturate. Both of these effects would
distort the output voltage and it would no longer be an amplified
copy of the input voltage.
Whites, EE 320
Lecture 16
Page 1 of 7
(Fig. 5.27)
Similar to the analytical solution, there are two primary steps to
the graphical solution of such small-signal amplifiers:
1. DC basis analysis
2. AC small-signal analysis.
DC Bias
The first step in the bias calculations is to determine IB. This is
done with the iB-vBE characteristic curve and the load line:
Whites, EE 320
Lecture 16
VBB = I B RB + VBE
Page 2 of 7
IB =
VBE VBB
+
RB RB
(Fig. 5.28)
Once IB has been determined we can compute IC knowing that
I C = I B for a BJT in the active mode. With this IC value and
the iC-vCE characteristic curve of the transistor, we can determine
VCE.
We havent yet seen the iC-vCE characteristic curve of the BJT.
This can be measured using the circuit in Fig. 5.19(a) below. vBE
is fixed at some value, then vCE is swept while measuring iC. The
results are shown below for different values of vBE.
(Fig. 5.19)
Whites, EE 320
Lecture 16
Page 3 of 7
When vCE is very small, iC is nearly zero. This is the cutoff mode
of the BJT. As vCE increases, the CBJ is forward biased and the
BJT is in the saturation mode. When vCE becomes large enough,
the CBJ becomes reversed biased and the BJT enters the active
mode.
The slopes of the lines in Fig. 5.19 in the active mode are quite
exaggerated in this figure.
So, back to the graphical solution. With the I C = I B value from
Fig. 5.28 and the iC-vCE characteristic curve of the transistor
from Fig. 5.19, we can determine VCE:
VCC = I C RC + VCE
IC =
VCE VCC
+
RC RC
(Fig. 5.29)
Curve tracers are pieces of equipment that will measure and
display families of iC-vCE characteristic curves for transistors.
Whites, EE 320
Lecture 16
Page 4 of 7
AC Small-Signal Analysis
The first step in the AC small-signal analysis is to determine ib.
This is performed using a slightly complicated interaction of the
input waveform vi, the subsequent time variation of the load
line, and the iB-vBE characteristic curve of the BJT:
(Fig. 5.30a)
From this comes the small-signal quantities vbe and ib.
With ib known and ic = ib , then we use these values on the iCvCE characteristic curve to determine vce:
Whites, EE 320
Lecture 16
Page 5 of 7
(Fig. 5.30b)
Whites, EE 320
Lecture 16
Page 6 of 7
(Fig. 5.30b)
Because of these limits on vCE, it is important to choose the Q
point properly to all for the desired swing in the signal voltage
(vce).
Whites, EE 320
Lecture 16
Better
compromise
Page 7 of 7
(Fig. 5.31)
Whites, EE 320
Lecture 17
Page 1 of 10
Whites, EE 320
Lecture 17
Page 2 of 10
(Fig. 5.44a)
RE is part of this biasing method as well. When used as an
amplifier, the input signal would be capacitively coupled to the
base of the BJT while the output would be taken (through
capacitive coupling) at the collector or emitter of the transistor,
depending on the specific requirements for the amplifier.
We analyzed a specific example of this type of circuit in Lecture
12 employing Thvenins theorem to simplify the analysis:
IE
(Fig. 5.44b)
where VBB and RB are given in (5.68) and (5.69) in the text.
Whites, EE 320
Lecture 17
Page 3 of 10
R
VBB = VBE + B I E + RE I E = VBE + B + RE I E
+1
+1
Consequently,
IE =
VBB VBE
R
RE + B
+1
(1)
(2)
(5.70),(3)
(5.71),(5)
Whites, EE 320
Lecture 17
Page 4 of 10
Whites, EE 320
Lecture 17
Page 5 of 10
VCC
3
0.2 I E
VCC
3
0.5 mA
Whites, EE 320
Lecture 17
Page 6 of 10
For the design with = 100 it can be shown that I E = 0.48 mA.
(This is only a -4% change from 0.5 mA with = .)
(Fig. 5.45)
Using KVL around the loop L gives
IE
R + V + I R = +VEE
(6)
+ 1 B BE E E
V VBE
(5.73),(7)
or
I E = EE
RB
RE +
+1
This is the same result as (3), but with VBB replaced by VEE.
Consequently, the - and temperature-invariant design equations
for this circuit are the same as those given earlier in (4) and (5)
with VBB replaced by VEE.
Whites, EE 320
Lecture 17
Page 7 of 10
(Fig. 5.46a)
As shown in the text, for IE to be insensitive to variations,
choose
R
RC B
(8)
+1
and for VBE to be insensitive to temperature variations, choose
VCC VBE
(9)
This latter requirement is most often very easy to meet!
Whites, EE 320
Lecture 17
Page 8 of 10
(Fig. 5.47a)
In this circuit, I E = I . If we are using a good current source,
then IE will not depend on . Very nice.
However, what weve done in this approach is to push the
technical problem to the design of a good current source.
Current Mirror
Simple biasing methods often fail to provide constant collector
currents if the supply voltage or ambient temperature change.
This is a problem with mobile telephones, for example, where
the battery voltage changes with use and the device operates in a
range of temperatures.
There are sophisticated circuits consisting of tens of devices that
can produce golden currents that are supply voltage and
temperature independent. These golden currents are replicated
throughout a device using a current mirror:
Whites, EE 320
Lecture 17
Page 9 of 10
0
0 0
(Fig. 5.47b)
There are better and more sophisticated approaches than this, of
course. This is just a simple example.
In this current mirror, Q1 is called a diode-connected BJT
because the collector and base terminals are connected together.
For proper operation of this circuit, it is very important that the
BJTs be matched, meaning they having the same ,
characteristic curves, etc. Usually this means that the BJTs must
be fabricated at the same time on the same substrate.
For the analysis of this circuit, we assume that is very large
and that Q1 and Q2 operate in the active mode. Because of this,
we ignore the base currents in Q1 and Q2.
Therefore, the collector (and emitter) current through Q1 is
approximately equal to IREF. By KVL,
VCC = I REF R + VBE VEE
1
I REF = (VCC VBE + VEE )
(5.76),(10)
or
R
Whites, EE 320
Lecture 17
Page 10 of 10
Now, since Q1 and Q2 are matched and they have the same VBE,
then the collector currents must be the same. This implies that
V + VEE VBE
I = I REF = CC
(5.77),(11)
R
This current mirror circuit will supply this current I as long as
Q2 operates in the active region:
V > VBE VEE
Notice that the diode-connected Q1 cannot saturate since the
base and collector terminals are shorted together. Hence, Q1
operates in the active mode or is simply cutoff.
Whites, EE 320
Lecture 18
Page 1 of 7
(Fig. 5.60a)
The capacitor CE is called a bypass capacitor. At the operating
frequency, its purpose is to shunt out the effects of the DC
current source from the time varying signal. In other words, CE
sets an AC ground at this node at the frequency of operation.
2009 Keith W. Whites
Whites, EE 320
Lecture 18
Page 2 of 7
There are a number of ways to bias this amplifier, other than that
shown above. What were primarily interested in here is the
small-signal characteristics.
(Fig. 5.60b)
Notice that weve included ro in this small-signal model. This is
the finite output resistance of the BJT. This accounts for the
finite slope of the characteristic curves of iC versus vCE
mentioned briefly in Lecture 16:
Whites, EE 320
Lecture 18
Page 3 of 7
(Fig. 5.19b)
where VA is called the Early voltage. Usually ro is fairly large, on
the order of many tens of k.
Our quest in the small-signal analysis of this amplifier is to
determine these quantities: input resistance Rin, the overall
small-signal voltage gain Gv = vo vsig , the partial small-signal
voltage gain Av = vo vi , the overall small-signal current gain
Gi = io ii , the short circuit small-signal current gain Ais = ios ii ,
and the output resistance Rout.
Input resistance, Rin. Directly from the small-signal equivalent
circuit, we see that
Rin = RB || r
(5.109),(1)
Oftentimes we select RB r so that
Rin r
r will often be a few k, which means this CE amplifier
presents a moderately large value of input impedance.
Whites, EE 320
Lecture 18
Page 4 of 7
r || RB
vsig
r || RB + Rsig
(5.113),(4)
(5.122),(6)
Whites, EE 320
Lecture 18
vo
vi
At the input, vi = v while at the output,
vo = g m v ( ro || RC || RL )
Av
Page 5 of 7
(7)
(8)
Gi =
g m ( r || RB )( ro || RC || RL )
RL
(11)
Short circuit small-signal current gain, Ais. This is the smallsignal current gain of the amplifier but with a short circuited
load ( RL = 0 ):
Whites, EE 320
Lecture 18
Page 6 of 7
ios
ii
(5.124),(12)
Ais
Equivalently,
Ais = Gi
(13)
RL = 0
(5.124),(14)
Rout = RC || ro
which is generally fairly large.
(16)
Whites, EE 320
Lecture 18
Page 7 of 7
Whites, EE 320
Lecture 19
Page 1 of 10
(Fig. 5.61a)
This is called emitter degeneration and has the effect of
greatly enhancing the usefulness of the CE amplifier.
Well calculate similar amplifier quantities for this circuit as
those in the previous lecture for the CE amplifier.
In contrast to the previous lecture, well use a T small-signal
model (Re in series with re) for the BJT and well also drop ro: it
turns out to have little effect here but complicates the analysis.
Whites, EE 320
Lecture 19
Page 2 of 10
(Fig. 5.61b)
Input resistance, Rin. From this circuit, we see directly that the
input resistance at the base Rib is defined as
v
(1)
Rib i
ib
Notice that here vi v , unlike the CE amplifier w/o emitter
degeneration. Referring to the small-signal circuit we see that
vi = ie ( re + Re )
(2)
i
(3)
ib = e
and
+1
Substituting these into (1) gives
Rib = ( + 1) ( re + Re )
(5.127),(4)
We see from this expression that the base input resistance is
+1 times the total resistance in the emitter circuit. This is
called the resistance reflection rule.
[In the previous lecture, we see in Fig. 5.60(b) that
Whites, EE 320
Lecture 19
Page 3 of 10
Rib = r
but r = ( + 1) re , which obeys this resistance reflection rule
since there is no Re in that circuit.]
This base input resistance can be much larger than without the
emitter resistance. Thats often a good thing. The designer can
change Re to achieve a desired input resistance [> (+1)re].
The total input resistance to this CE amplifier with emitter
degeneration is then
Rin = RB || Rib = RB || ( + 1) ( re + Re ) (5.125),(5)
(5.129),(8)
Whites, EE 320
Lecture 19
Page 4 of 10
(10)
(14)
Whites, EE 320
Lecture 19
Page 5 of 10
Whites, EE 320
Lecture 19
Page 6 of 10
(Fig. E5.41a)
The small-signal equivalent circuit for this CE amplifier with
emitter degeneration is the same as that shown in the previous
lecture:
Whites, EE 320
Lecture 19
Page 7 of 10
(Fig. 5.61b)
With I E = 1 mA, then re = VT I E = 25 mV 1 mA = 25 .
Find the value of Re that gives Rin = 4 Rsig = 20 k. From (5)
Rin = RB || Rib , which implies that Rib = 25 k. Using (4)
R
Rib = ( + 1) ( re + Re ) re + Re = ib
+1
R
25,000
25 = 222.5
Re = ib re =
or
101
+1
Determine the output resistance. From (20),
Rout = RC = 8 k
Compute the overall small-signal voltage gain. Using (12)
( RC || RL ) Rin
v
Gv = o =
vsig
re + Re
Rin + Rsig
Gv =
0.99 ( 3,080 )
20,000
= 9.86 V/V
25 + 222.5 20,000 + 5,000
Whites, EE 320
Lecture 19
Page 8 of 10
RC
re + Re
(5.131),(22)
0.99 8,000
= 32 V/V
25 + 222.5
Can you physically explain why Gvo and Avo are different
values?
Avo =
Compute the overall current gain and the short circuit current
gain. Using (17)
Whites, EE 320
Gi =
=
Lecture 19
Page 9 of 10
RB RC
( RC + RL )( RB + Rib )
100 100,000 8,000
= 49.2 A/A
(8,000 + 5,000 )(100,000 + 25,000 )
Whites, EE 320
Lecture 19
Page 10 of 10
Whites, EE 320
Lecture 20
Page 1 of 5
(Fig. 5.62a)
The small-signal equivalent circuit for this amplifier is shown in
Fig. 5.62b (ignoring ro):
(Fig. 5.62b)
2009 Keith W. Whites
Whites, EE 320
Lecture 20
Page 2 of 5
Av =
re
( RC || RL ) = g m ( RC || RL )
(5.138),(5)
Whites, EE 320
Lecture 20
Page 3 of 5
(7)
(5.142),(10)
Whites, EE 320
Lecture 20
Page 4 of 5
gain can be fairly large, though if Rsig is nearly the same size
as the total emitter resistance the gain will be small. In other
words, if this amplifier is connected to a high output
impedance stage, it will be difficult to realize high gain.
Overall small-signal current gain, Gi. By definition
i
(11)
Gi o
ii
Using current division at the output of the small-signal
equivalent circuit above
RC
RC
(12)
io =
ic =
ie
RC + RL
RC + RL
Because ii = ie this expression gives
i
RC
(13)
Gi = o =
ii RC + RL
Short circuit current gain, Ais. In the case of a short circuit
load (RL = 0), Gi in (13) reduces to the short circuit current
gain:
i
(5.140),(14)
Ais = os =
ii
Output resistance, Rout. Referring to the small-signal
equivalent circuit above and shorting out the input vsig = 0
Rout = RC
(15)
which is the same as the CE amplifier (when ignoring ro).
Whites, EE 320
Lecture 20
Page 5 of 5
Summary
Summary of the CB small-signal amplifier:
1. Low input resistance.
2. Gv can be very large, though critically dependent on Rsig.
3. Ais=
4. Potentially large output resistance (dependent on RC).
One very important use of the CB amplifier is as a unity-gain
current amplifier, which is also called a current buffer amplifier.
This type of amplifier accepts an input signal current at a low
impedance level and outputs nearly the same current amplitude,
but at a high output impedance level. Even though this is a
buffer amplifier, there is still power gain.
Whites, EE 320
Lecture 21
Page 1 of 9
(Fig. 5.63a)
The small-signal equivalent circuit is shown in Fig. 5.63b:
(Fig. 5.63b)
Weve included ro in this model since it can have an appreciable
effect on the operation of this amplifier.
Notice that ro is connected from the emitter to an AC ground.
We can simplify the AC small-signal analysis of this circuit by
2009 Keith W. Whites
Whites, EE 320
Lecture 21
Page 2 of 9
ii
io
Rib
(Fig. 5.63c)
(1)
(2)
(3)
Whites, EE 320
Lecture 21
Page 3 of 9
(4)
Whites, EE 320
Lecture 21
Page 4 of 9
(10)
RB || ( + 1) ( re + ro || RL )
ro || RL
Gv =
ro || RL + re RB || ( + 1) ( re + ro || RL ) + Rsig
(12)
Whites, EE 320
Lecture 21
Gv
Page 5 of 9
RL
re + RL +
Rsig
(5.146),(14)
+1
+
1
(19)
(
)
ro + RL
RB + Rib
from which we find that
Whites, EE 320
Lecture 21
Gi =
or
Gi =
( + 1) ro RB
io
=
ii ( ro + RL )( RB + Rib )
( + 1) ro RB
( ro + RL ) RB + ( + 1) ( re + ro || RL )
Page 6 of 9
(20)
(21)
So even though the amplifier has a voltage gain less than one
(and approaching one in certain circumstances), it has a very
large small-signal current gain. Overall, the amplifier does
provide power gain to the AC signal.
Output resistance, Rout. With vsig = 0 in the small-signal
equivalent circuit, were left with
Whites, EE 320
Lecture 21
Page 7 of 9
(1 ) ie
ie
B
RB||Rsig
ie
re
E
ix
+
ro
vx
Rout
= ie (1 ) ( Rsig || RB ) + re
Using KCL at the output
v
(26)
ix = x ie
ro
Substituting (26) into (25)
Whites, EE 320
Lecture 21
Page 8 of 9
vx
vx = ix (1 ) ( Rsig || RB ) + re
ro
(1 ) ( Rsig || RB ) + re
vx 1 +
= ix (1 ) ( Rsig || RB ) + re (27)
ro
Rout = ro || (1 ) ( Rsig || RB ) + re
This is equivalent to
Rout
Rsig || RB
= ro ||
+ re
+1
(5.148),(28)
Rsig || RB
+1
+ re
(5.149),(29)
Summary
Summary of the CC (emitter follower) small-signal amplifier:
Whites, EE 320
Lecture 21
Page 9 of 9
Whites, EE 320
Lecture 22
Page 1 of 12
Whites, EE 320
Lecture 22
Page 2 of 12
(Fig. 5.71)
The roll off in gain near fL and lower is due to effects of the DC
blocking capacitors CC1 and CC2, and the bypass capacitor CE.
Its not possible to eliminate this effect, though fL can be moved
about by choosing different values for these capacitors. But
large capacitors take up lots of space and can be expensive.
Whites, EE 320
Lecture 22
Page 3 of 12
Capacitance of pn Junctions
There are basically two types of capacitances associated with pn
junctions:
1. Junction capacitance. This is related to the space charge
that exists in the depletion region of the pn junction.
2. Diffusion capacitance, or charge storage capacitance. This
is a new phenomenon we havent yet considered in this
course.
The junction capacitance effect was briefly mentioned earlier in
this course in Lecture 4. The width of the depletion region will
change depending on the applied voltage and whether the
junction is reversed or forward biased:
Whites, EE 320
Lecture 22
Page 4 of 12
(Fig. 3.49)
In this state, current will flow across the junction, of course.
Because of the current source in Fig. 3.49 and the voltage drop
V, holes are injected across the junction into the n region while
electrons are injected across the junction into the p region.
Whites, EE 320
Lecture 22
Page 5 of 12
(Fig. 3.50)
The concentrations of these electrons and holes decrease in
value away from the junction, as shown in Fig. 3.50, due to
recombination effects.
The important point here is that these concentrations of charges
create an electric field across the pn junction that will vary with
time when a signal source is connected to this device. This
electric field is directed from the n to p region, and the overall
effect can be modeled by what is called the charge storage
capacitance, or diffusion capacitance.
To summarize, the capacitive effects of a reversed biased pn
junction are described by the junction capacitance while those of
a forward biased pn junction are described by both a junction
Whites, EE 320
Lecture 22
Page 6 of 12
Whites, EE 320
Lecture 22
Page 7 of 12
(Fig. 5.67)
Note the use of the V notation in this small-signal model. Your
textbook has switched to sinusoidal steady state notation for this
high frequency discussion.
The high frequency small-signal model in Fig. 5.67 also
includes the resistance rx, which is mostly important at high
frequencies. Its there to approximately model the resistance of
the base region from the terminal to a point somewhere directly
below the emitter:
rx
(Fig. 5.6)
C is sometimes referred to as Cob (or Cobo) in datasheets. This
designation reflects the fact that C can be the output resistance
when the BJT is used as a common base amplifier.
Whites, EE 320
Lecture 22
Page 8 of 12
Whites, EE 320
Lecture 22
Page 9 of 12
Unity-Gain Bandwidth
An important high frequency characteristic of transistors that is
usually specified is the unity-gain bandwidth, fT. This is defined
as the frequency at which the short-circuit current gain
I
h fe c
(2)
I b s.c. load
has decreased to a value of one.
A test circuit for this measurement would look something like:
Whites, EE 320
Lecture 22
Page 10 of 12
I c = g mV jCV = ( g m jC )V
At the input terminal B
V = I b Z in = I b r || Z C + Z C
or
V = I b + j ( C + C )
r
(5.157),(3)
(5.158),(4)
I c = ( g m jC ) I b + j ( C + C )
r
Using the definition of hfe from (2) we find from this last
equation that
g m jC
I
h fe = c
=
(5)
1
I b s.c. load
+ j ( C + C )
r
It turns out that C is typically quite small and for the purposes
of determining the unity-gain bandwidth, gm is | jC | for the
frequencies of interest here. In other words, the frequency at
which C is important relative to gm is much higher than what
is of interest here.
Consequently, from (5)
Whites, EE 320
Lecture 22
h fe
gm
1
+ j ( C + C )
r
Page 11 of 12
g m r
1 + j r ( C + C )
(6)
(Fig. 5.69)
or
such that
1 + jT r ( C + C )
0 = 1 + jT r ( C + C ) = 1 + j
Whites, EE 320
Lecture 22
T
2
0 = 1 +
Page 12 of 12
T
for T .
Therefore
T 0 =
g m r
gm
=
( C + C ) r C + C
(5.162),(8)
so that
fT
gm
2 ( C + C )
(5.163),(9)
Whites, EE 320
Lecture 23
Page 1 of 17
(Fig. 5.71)
As we discussed in the previous lecture, there are three distinct
region of frequency operation for this and most transistor
amplifier circuits. Well examine the operation of this CE
Whites, EE 320
Lecture 23
Page 2 of 17
Well define
RL = ro || RC || RL
(1)
Vo = g m RLV
(2)
Whites, EE 320
V =
Lecture 23
Page 3 of 17
r
r
RB
VTH =
Vsig
r + rx + RTH
r + rx + RB || Rsig RB + Rsig
(3)
( ro || RC || RL ) (4)
Am o =
Vsig r + rx + RB || Rsig RB + Rsig
(Fig. 5.72a)
Whites, EE 320
Lecture 23
Page 4 of 17
(Fig. 5.72b)
where it can be easily shown that Vsig is V given in (3)
r
RB
Vsig =
Vsig (5.167),(5)
r + rx + RB || Rsig RB + Rsig
while
Rsig = r || rx + ( RB || Rsig )
(5.168),(6)
Millers Theorem
We can analyze the circuit in Fig. 5.72b through traditional
methods, but if we apply Millers theorem we can greatly
simplify the effort. Plus, it will be easier to apply an
approximation that will arise if we use Millers theorem.
You may have seen Millers theorem previously in circuit
analysis. It is another equivalent circuit theorem for linear
Whites, EE 320
Lecture 23
Page 5 of 17
(Fig. 1)
The equivalent Millers theorem circuit is
(Fig. 2)
where
ZA =
Zx
v
1 B
vA
and Z B =
Zx
v
1 A
vB
(7),(8)
Whites, EE 320
Lecture 23
Page 6 of 17
vA
ZA
(10)
iA =
Whites, EE 320
Lecture 23
Page 7 of 17
(Fig. 3)
where, using (14) and (15),
V
V
C A = C 1 o and CB = C 1
V
Vo
(16),(17)
Whites, EE 320
and
Lecture 23
Page 8 of 17
I L g mV
Vo I L RL = g m RLV
(19)
(5.169),(20)
V
1
C B C 1 +
and
= C 1 +
g R V
g R
m L
m L
(21)
(22)
(Fig. 5.72c)
Whites, EE 320
where
Lecture 23
Page 9 of 17
Cin C + C A = C + C 1 + g m RL
(5.173),(22)
Z Cin + Rsig
while at the output
Vo = g m RLV
(24)
Vo = g m RL
Z Cin
Z Cin
+ Rsig
Vsig
Vo = g m RL
Vsig =
Vsig
1
1 + jCin Rsig
+ Rsig
jCin
If we define
1
H =
Cin Rsig
1
f
Vsig 1 + j
1+ j
H
fH
(25)
(26)
(27)
(28)
Whites, EE 320
where
Lecture 23
fH =
H
1
=
2 2 Cin Rsig
Page 10 of 17
(5.176),(29)
You should recognize this transfer function (28) as that for a low
pass circuit with a cut-off frequency (or 3-dB frequency) of H.
This is the response of a single time constant circuit, which is
what we have in the circuit of Fig. 5.72c.
What were ultimately interested in is the overall transfer
function Vo Vsig from input to output. This can be easily derived
from the work weve already done here. Since
Vo
Vo Vsig
=
(30)
Vsig Vsig Vsig
We can use (28) for the first term in the RHS of (30), and use (5)
for the second giving
Vo
r
g m RL
RB
=
(31)
Vsig 1 + j f r + rx + RB || Rsig RB + Rsig
fH
We can recognize Am from (4) in this expression giving
Vo
Am
=
(5.175),(32)
Vsig 1 + j f
fH
Once again, this is the frequency response of a low pass circuit,
as shown below:
Whites, EE 320
Lecture 23
Page 11 of 17
(Fig. 5.72d)
Whites, EE 320
Lecture 23
Page 12 of 17
Whites, EE 320
Lecture 23
Page 13 of 17
Vo
j
j
j
Am
(5.183),(33)
Vsig
j + p1 j + p 2 j + p 3
(Fig. 5.73e)
So there isnt a single fL as suggested by Fig. 5.71b but rather a
more complicated response at low frequencies as we see in Fig.
5.73e above. Computer simulation is perhaps the best predictor
for this complicated frequency response, but an approximate
formula for fL is given in the text as
1 1
1
1
+
+
f L f p1 + f p 2 + f p 3 =
2 CC1RC1 CE RE CC 2 RC 2
(5.184),(5.185),(34)
where RC1 , RE , and RC 2 are the resistances seen by CC1 , CE , and
CC 2 , respectively, with the signal source Vsig = 0 and the other
two capacitors replaced by short circuits.
Whites, EE 320
Lecture 23
Page 14 of 17
V_DC
SRC2
Vdc=10.0 V
AC
AC
AC1
Start=100 Hz
Stop=1.0 MHz
Step=100 Hz
R
R2
R=8 kOhm
vo
C
C2
C=10 uF
vi
V_AC
R
SRC1
R3
Vac=polar(1,0) V R=5 kOhm
Freq=freq
C
C1
C=10 uF
R
R1
R=100 kOhm ap_npn_2N2222A_19930601
Q1
C
I_DC
C3
SRC4
Idc=1 mA C=10 uF
V_DC
SRC3
Vdc=-10.0 V
R
R4
R=5 kOhm
Whites, EE 320
Lecture 23
Page 15 of 17
gm =
IC
1 mA
=
= 0.04 S
VT 25 mV
From (5.163),
fT
gm
0.04
=
= 246.8 MHz
2 ( C + C ) 2 ( 20 pF + 5.8 pF )
This value agrees fairly with the datasheet value of 300 MHz.
0 265 from the ADS parts list for this 2N2222A transistor.
Therefore,
r =
0
gm
265
= 6,625
0.04
Whites, EE 320
Lecture 23
Am =
Page 16 of 17
g m r
RB
( ro || RC || RL )
r + rx + RB || Rsig RB + Rsig
2,898.6 = RL
0.02327
0.9524
V
V
Am ) = 36.2 dB
Am = 64.24
Therefore,
or in decibels
Am = 20 log10 (
From ADS:
m3
freq= 400.0 Hz
dB(vo)=33.632
40
dB(vo)
35
m3
m1
m2
freq= 6.300kHz freq= 84.40kHz
dB(vo)=36.053 dB(vo)=33.046
m1
m2
30
25
20
15
10
1E2
1E3
1E4
1E5
1E6
freq, Hz
fH
1
2 Cin Rsig
Whites, EE 320
Lecture 23
Page 17 of 17
Therefore,
f H 2 698.3 1012 2,771 = 82.25 kHz
This agrees very closely with the value of 84.40 kHz predicted
by the ADS simulation shown above.
Whites, EE 320
Lecture 24
Page 1 of 5
(Fig. 5.74)
Cutoff Region. If vI 3 0.5 or so, the EBJ will conduct
negligible current. Also, the CBJ will be reversed biased with
a large VCC.
Consequently,
iB 0 , iC 0 , and iE 0
(1)
which means
vO = VCC
(2)
These are the cutoff conditions and the BJT is in the off
state.
Whites, EE 320
Lecture 24
Page 2 of 5
VCE
sat
0.2 V
VEC
sat
0.2 V
VBE 0.7 V
VCE
sat
0.2 V
VEB 0.7 V
VEC
iC
RC
vO VCE
RB
+
vI
iB
iE
sat
sat
0.2 V
Whites, EE 320
Lecture 24
vO = VCE
With
Page 3 of 5
sat
(3)
then
VCC VCE sat
vI 0.7
, iC sat =
, iE = iB + iC sat
(4)
RB
RC
Remember that because the BJT is no longer operating in the
active region, iC iB .
iB =
(5)
VO 0.2 V
Whites, EE 320
Lecture 24
Page 4 of 5
Therefore, since
IB =
5 0.7
4.3
RB =
= 2.2 k
RB
IB
Now, with this design and the transistor saturated, what is the
forced ?
I
9.8 mA
forced = C sat =
=5
IB
1.96 mA
Whites, EE 320
Lecture 24
Page 5 of 5
forced =
ODF
(7)
forced =
50
=5
10
Whites, EE 320
Lecture 25
Page 1 of 10
Whites, EE 320
Lecture 25
Page 2 of 10
(Fig. 4.1a)
Four Terminals:
n+ means
heavily doped
(Fig. 4.1b)
Whites, EE 320
Lecture 25
Page 3 of 10
Whites, EE 320
Lecture 25
Page 4 of 10
(Fig. 4.2)
In effect, the gate and the channel region form a parallel plate
capacitor of sorts. Two things happen when vGS is applied:
1. Free holes in the p-type substrate are repelled from the
region under the gate. This process uncovers bound
negative charge.
2. Electrons from the heavily doped n+ regions (the drain and
source) are attracted under the gate.
These effects create an n-type channel. Notice that this bias
voltage vGS is required in order to create the channel: no vGS,
no channel.
Now, if a voltage is applied between the drain and source we
will have a flow of electrons from source to drain (i.e., a
current). This is the origin of the names source and drain.
Whites, EE 320
Lecture 25
Page 5 of 10
(Fig. 4.3)
The vGS required to accumulate sufficient numbers of mobile
electrons in the channel is called the threshold voltage Vt. For an
n-channel MOSFET, Vt 1 3 V (note that this is a positive
voltage).
A family of iD-vDS characteristic curves for the MOSFET with a
small vDS is shown in Fig. 4.4 with vGS as the parameter:
(Fig. 4.4)
Whites, EE 320
Lecture 25
Page 6 of 10
(Fig. 4.5)
In these circumstances, an additional electric field is created
from drain to source that is large enough to alter the shape of the
channel. With the electric field from vDS directed as shown
above, there exists more negative charges near the source end of
Whites, EE 320
Lecture 25
Page 7 of 10
(Fig. 4.6)
Note that it is possible to increase vDS large enough to reduce the
channel thickness to zero at the drain end.
(Fig. 4.7)
This is called pinch off ( vDS vGS Vt ).
Whites, EE 320
Lecture 25
Page 8 of 10
- - - - -
n+
E ++
+
+
Depletion region
p type
(4.1),(1)
Whites, EE 320
Lecture 25
Page 9 of 10
sat
(2)
vDS
W
i
k
v
V
v
=
Triode region: D
(4.5a),(3)
( GS t ) DS
n
L
2
1 W
2
iD = kn ( vGS Vt )
(4.6a),(4)
2
L
where kn is the process transconductance parameter [A/V2] and
is equal to
Saturation region:
kn = nCox = n ox
(4.7),(5)
tox
Here, n is the mobility of electrons in the channel [cm2/(V-s)],
Cox is the capacitance per unit gate area [F/m2], and ox and tox
are the permittivity and thickness of the gate oxide layer,
respectively.
Whites, EE 320
Lecture 25
Page 10 of 10
IS
p+
p channel
ID
ISD
p+
n type
(Fig. 4.9)
Whites, EE 320
Lecture 26
Page 1 of 8
(Fig. 4.10b)
Referring to this circuit symbol:
9 The arrowed terminal indicates the source,
9 This arrow direction indicates n-type (direction of current)
9 The gap at the gate indicates the oxide layer.
However, the body is often connected to the source. This leads
to a more common circuit symbol:
(Fig. 4.10c)
Similar circuit symbols are used for p-channel enhancement type
MOSFETS:
Whites, EE 320
Lecture 26
Page 2 of 8
(Fig. 4.18b,c)
(Fig. 4.11)
There are three regions of operation:
(1) Cutoff. To operate an enhancement type MOSFET, we first
must induce the channel. For NMOS, this means that
vGS Vt (induce)
(4.8),(1)
Whites, EE 320
Lecture 26
Page 3 of 8
we see that
Whites, EE 320
Lecture 26
Page 4 of 8
Therefore,
vDS < vGS Vt (continuous)
(4.10),(4)
We can use either (2) or (4) to check for triode operation of
the MOSFET.
(4.11),(5)
(4.12),(6)
1
rDS
Whites, EE 320
Lecture 26
1 W
2
iD = kn ( vGS Vt )
2
L
and is not dependent on vDS.
Page 5 of 8
(4.20),(10)
(Fig. 4.12)
In the saturation mode, this device behaves as an ideal
current source controlled by vGS:
(Fig. 4.13)
In reality, though, there is a finite output resistance (ro) that
should be added to this model:
Whites, EE 320
Lecture 26
iG = 0
vGS
Page 6 of 8
iD
kn W
2
( vGS Vt )
2 L
vDS
VA
(4.26),(11)
ID
This finite output resistance gives a slope to the iDvDS
characteristic curves:
where
ro =
(Fig. 4.16)
Whites, EE 320
Lecture 26
Page 7 of 8
VD
D
G
3V
Whites, EE 320
Lecture 26
vGD
vDS
Page 8 of 8
Whites, EE 320
Lecture 27
Page 1 of 8
I D = 0.4 mA
VD = 1 V
5 1
= 10 k
0.4 mA
Whites, EE 320
Lecture 27
Page 2 of 8
From this circuit we can see that VGD = 1 V, which is less than
Vt. Consequently, the channel is pinched off at the drain end.
Therefore, the MOSFET is operating in the saturation or cutoff
modes (not the triode).
Well assume operation in the saturation mode. In this mode
1 W
1
W
2
2
I D = kn (VGS Vt ) = nCox (VGS Vt )
L
L
2
2
Substituting
1
A 400
2
0.4 mA = 20 106 2
(VGS 2 )
2
V 10
Therefore
2
(VGS 2 ) = 1 VGS 2 = 1
or
VGS = +1 V or +3 V
The first solution is not consistent with our initial assumption of
operation in the saturation mode since it is less than Vt.
Therefore,
VGS = 3 V VS = 3 V
Finally,
RS =
VS ( 5 ) VS + 5 3 + 5
=
=
= 5 k
IS
ID
0.4 mA
Whites, EE 320
Lecture 27
Page 3 of 8
VD
In saturation,
Whites, EE 320
Lecture 27
Page 4 of 8
Example N27.3 (text example 4.4). Design the circuit below for
a drain voltage of 0.1 V. Determine rDS. The MOSFET has Vt = 1
V and kn W L = 1 mA/V2. Neglect ro.
(Fig. 4.22)
With VGS = 5 V and greater than Vt, the MOSFET has an
induced channel and is not cutoff.
Next, lets check to see if the channel is pinched off at the drain
end. We can do this two (equivalent) ways. First, with VD = 0.1
V then
VGD = 5 0.1 = 4.9 V
which is greater than Vt (= 1 V), so the channel is not pinched
off at the drain. Alternatively, we can compute
VGS Vt = 5 1 = 4 V
which is greater than VDS (= 0.1 V). So again we find that the
channel is not pinched off at the drain.
Either of these two results means the MOSFET is operating in
the triode mode (continuous channel).
Whites, EE 320
Lecture 27
Page 5 of 8
so that
Then
and
W
1 2
I D = kn (VGS Vt )VDS VDS
L
2
1
I D = 0.395 mA
5 0.1
k = 12.41 k
0.395
V
0.1
= DS =
k = 253
I D 0.395
RD =
rDS
We could also use (4.13) for this last result, but the work was
already done here. From the text,
1
vDS
W
rDS
= kn
(4.13)
(VGS Vt )
iD vDS small L
vGS =VGS
Whites, EE 320
Lecture 27
Page 6 of 8
(Fig. 4.24)
For saturation in an enhancement type PMOS device requires
VGS Vt (induced) or VSG Vt (induced)
(4.27)
and
VDS VGS Vt (pinched off)
(4.31)
In words, this last equation states that the drain-to-source
voltage must be less than the gate-to-source voltage plus |Vt|.
Whites, EE 320
Lecture 27
Page 7 of 8
VSG Vt
VDG < Vt
Whites, EE 320
Lecture 27
RDmax =
VDmax
ID
4V
= 8 k.
0.5 mA
Page 8 of 8
Whites, EE 320
Lecture 28
Page 1 of 7
iD
vDS
(Fig. 4.34)
This is only a conceptual amplifier for two primary reasons:
1. The bias with VGS is impractical. (Will consider others
later.)
2. In ICs, resistors take up too much room. (Would use
another triode-region biased MOSFET in lieu of RD.)
To operate as a small-signal amplifier, we bias the MOSFET in
the saturation region. For the analysis of the DC operating point,
we set vgs = 0 so that from (4.22) with = 0
1 W
2
(4.20),(1)
iD = kn ( vGS Vt )
2
L
From the circuit
VDS = VDD I D RD
(4.55),(2)
Whites, EE 320
Lecture 28
Page 2 of 7
(4.18),(3)
AC
(4.56),(4)
(5)
1 W
W
1 W
2 2
= kn (VGS Vt ) + kn (VGS Vt ) vgs + kn vgs2 (4.57),(6)
2
L
2
L
2
L
= I D (DC)
(time varying)
The last term in (6) is nonlinear in vgs, which is undesirable for a
linear amplifier. Consequently, for linear operation we will
require that the last term be small:
1 W 2
W
kn
vgs << kn (VGS Vt ) vgs
L
L
2
vgs << 2 (VGS Vt )
(4.58),(7)
or
Whites, EE 320
Lecture 28
Page 3 of 7
where
AC
W
id = kn (VGS Vt ) vgs .
L
(9)
GS
Whites, EE 320
Lecture 28
Page 4 of 7
(Fig. 4.35)
Lastly, it can be easily show that for this conceptual amplifier in
Fig. 4.34,
vd
(4.65),(13)
= g m RD
vgs
Consequently, Av g m , which is the same result we found for a
similar BJT conceptual amplifier [see (5.103)].
Whites, EE 320
Lecture 28
Page 5 of 7
id
VA
1
=
ID ID
is
(Fig. 4.37b)
Things to note from this small-signal model include:
1. ig = 0 and vgs 0 infinite input impedance.
2. ro models the finite output resistance. Practically speaking,
it will range from 10 k 1 M. Note that it depends
on the bias current ID.
3. From (10) we found
W
(14)
g m = kn (VGS Vt )
L
Alternatively, it can be shown that
I
ID
gm = D =
(4.71),(15)
Veff (VGS Vt ) / 2
which is similar to g m = I C VT for BJTs.
Whites, EE 320
Lecture 28
Page 6 of 7
ig = 0
is
(Fig. 4.40a)
Notice the direct connection between the gate and both the
dependent current source and 1/gm. While this model is correct,
weve added the explicit boundary condition that ig = 0 to
this small-signal model.
It isnt necessary to do this because the currents in the two
vertical branches are both equal to g m vgs , which means ig = 0 .
But adding this condition ig = 0 to the small-signal model in Fig.
4.40a makes this explicit in the circuit calculations. (The T
model usually shows this direct connection while the model
usually doesnt.)
MOSFETs have many advantages over BJTs including:
1. High input resistance
2. Small physical size
3. Low power dissipation
Whites, EE 320
Lecture 28
Page 7 of 7
(Fig. 6.46a)
Such a combination provides a very large input resistance from
the MOSFET and a large output impedance from the BJT.
Whites, EE 320
Lecture 29
Page 1 of 8
(Fig. 4.38a)
The first step is to determine the DC operating point. The DC
equivalent circuit is:
Whites, EE 320
Lecture 29
=0
Page 2 of 8
ID
Also, by KVL
Whites, EE 320
Lecture 29
Page 3 of 8
VDS = 15 RD I D = 15 10,000 I D
(4.74),(2)
(Fig. 4.38b)
W
(VGS Vt ) = 0.25 103 ( 4.4 1.5) = 0.725 mS
L
V
50
= 47.2 k
9 ro = A =
I D 1.06 mA
9 g m = k n
Whites, EE 320
Lecture 29
Page 4 of 8
Whites, EE 320
Lecture 29
Page 5 of 8
0.2sin (t ) V =
2V=
(Fig. 4.34)
The MOSFET characteristics are Vt = 1 V, kn = 20 A/V2, W/L
= 20, and = 0.
Whites, EE 320
Lecture 29
Page 6 of 8
(a) Determine ID and VD. We see from the circuit that VGS > Vt .
Therefore, the MOSFET is operating in the saturation or
triode mode. Well assume saturation. In that case
1 W
1
2
I D = kn (VGS Vt ) = 20 106 20(2 1) 2 = 0.2 mA
L
2
2
VD = VDD I D RD = 3 V
and
Lets check if the MOSFET is operating in the saturation
mode:
VGD = 2 3 = 1 < Vt
Therefore, the MOSFET is indeed saturated, as assumed.
(b) Determine gm. Using (4.61)
W
g m = kn (VGS Vt ) = 20 106 20 (2 1) = 4.0 mS
L
(c) Determine the voltage gain Av. We begin by first
constructing the small-signal equivalent circuit
vgs
g m vgs
vo
Whites, EE 320
so Av =
Lecture 29
Page 7 of 8
vo
= g m RD = 0.4 103 10 103 = 4 V/V
vgs
Whites, EE 320
Lecture 29
Page 8 of 8
Whites, EE 320
Lecture 30
Page 1 of 8
(Fig. 4.30d)
2009 Keith W. Whites
Whites, EE 320
Lecture 30
(Fig. 4.32)
Page 2 of 8
(Fig. 4.33a)
vi
vo
Whites, EE 320
Lecture 30
Page 3 of 8
Whites, EE 320
Lecture 30
vgs
Page 4 of 8
g m vgs
vo
we see that
Rin = RG RG = 1 M
Whites, EE 320
Lecture 30
Page 5 of 8
(Fig. 4.33b)
Q1 has the drain and gate terminals connected together. This
forces Q1 to operate in the saturation mode in this particular
circuit if I D1 0 . In this mode
1
W
2
I D1 = kn1 1 (VGS Vt1 )
(4.50),(1)
2
L1
With a zero gate current,
I REF = I D1
(2)
where we can easily see from the above circuit that
V V ( VSS )
I REF = DD GS
(4.51),(3)
R
Now, well assume the two MOSFETs in the circuit have the
same VGS. Consequently, the drain current in the second
transistor is
1
W
2
I D 2 = kn 2 2 (VGS Vt 2 )
(4)
2
L2
Whites, EE 320
Lecture 30
Page 6 of 8
Q2
Q1
0
IO = I D 2
R
I REF
To amplifier circuit
Whites, EE 320
Lecture 30
Page 7 of 8
Now, by KVL
52
VDD VGS
=
= 30 k
100 A
I REF
Whites, EE 320
Hence
Therefore,
Lecture 30
Page 8 of 8
VGD 2 Vt VG 2 VD 2 Vt
VO = VD 2 VG 2 Vt
or VO VGS Vt = 2 1 = 1 V
VO min = 1 V
Whites, EE 320
Lecture 31
Page 1 of 5
Whites, EE 320
Lecture 31
Page 2 of 5
(Fig. 4.43a)
Assuming sufficiently large values for the coupling capacitors
(CC1 and CC2) and the bypass capacitor (CS) so that their
reactances are very small at the frequency of operation the
equivalent small-signal circuit for this amplifier is shown in Fig.
4.43(b):
(Fig. 4.43b)
The text mentions performing the small-signal analysis directly
on the amplifier circuit, as illustrated in Fig. 4.43(c). We do not
recommend this approach. It is better to take the time and
construct the small-signal equivalent circuit, as were doing
here.
Whites, EE 320
Lecture 31
Page 3 of 5
vi = vgs
(3)
Substituting (3) into (2), we find that the partial voltage gain
as
v
Av o = g m ( ro || RD || RL )
(4.80),(4)
vi
Whites, EE 320
Lecture 31
Gv
vo
v v
v
= i o = i Av
vsig vsig vi vsig
N
Page 4 of 5
(5)
= Av
(4.82),(7)
Whites, EE 320
Lecture 31
Page 5 of 5
Summary
In summary, we find for the CS small-signal amplifier that it has
a
o High input resistance [see (1)].
o Relatively high small-signal voltage gain [see (7)].
o Very high small-signal current gain [see (10)].
o Relatively high output resistance [see (11)].
Whites, EE 320
Lecture 32
Page 1 of 9
vO
(Fig. 4.44a)
We have a choice of small-signal models to use for the
MOSFET. A T model will simplify the analysis, on one hand, by
allowing us to incorporate the effects of RS by simply adding
this value to 1/gm in the small-signal model, if we ignore ro.
Whites, EE 320
Lecture 32
Page 2 of 9
(Fig. 4.44b)
On the other hand, using the T model makes the analysis more
difficult when ro is included. (The hybrid model is better at
easily including the effects of ro.) However, ro in the MOSFET
amplifier is large so we can reasonably ignore its effects for now
in the expectation of making the analysis more tractable.
Whites, EE 320
Lecture 32
Page 3 of 9
Whites, EE 320
Lecture 32
Page 4 of 9
Whites, EE 320
Lecture 32
Page 5 of 9
Discussion
Adding RS has a number of effects on the CS amplifier. (Notice,
though, that it doesnt affect the input and output resistances.)
First, observe from (3)
vi
(3)
1 + g m RS
that we can employ RS as a tool to lower vgs relative to vi and
lessen the effects of nonlinear distortion.
vgs =
Whites, EE 320
Lecture 32
Gv
RG RD || RL
RG + Rsig RS
Page 6 of 9
(12)
Whites, EE 320
Lecture 32
Page 7 of 9
W
mA
kn = 1 2
L
V
Vt = 1.5 V
vO
Whites, EE 320
Lecture 32
Page 8 of 9
4.7M
V
103 (14k ||14k ) = 6.85
4.7M + 100k
V
For an input sinusoid with 0.4-Vpp amplitude, then
Vo = Gv Vsig = 6.85 0.4 Vpp = 2.74 Vpp
Gv =
Will the MOSFET remain in the saturation mode for the entire
cycle of this output voltage? For operation in the saturation
mode, vDG = vD > Vt = 1.5 V. On the negative swing of the output
voltage,
v
2.74
vD min = VD o ,pp = 3
= 1.63 V
2
2
which is greater than Vt, so the MOSFET will not leave the
saturation mode on the negative swings of the output voltage.
On the positive swings,
v
2.74
vD max = VD + o ,pp = 3 +
= 4.37 V
2
2
which is less than VDD = 10 V so the MOSFET will not cutoff
and leave the saturation mode.
(Interestingly, the MOSFET does leave the saturation mode on
the negative swings for RD = RL = 15 k, as used in the text
exercises 4.32 and 4.33.)
Lastly, imagine that for some reason the input voltage is
increased by a factor of 3 (to 1.2 Vpp). What value of RS can be
used to keep the output voltage unchanged?
Whites, EE 320
Lecture 32
Page 9 of 9
3 1
2
= 3 = 2 k.
g m 10
With RS = 2 k the new overall small-signal AC voltage gain is
from (7)
V
6.85
6.85
Gv =
=
= 2.28
1 + g m RS
3
V
The overall small-signal voltage gain has gone down, but the
amplitude of the output voltage has stayed the same since the
input voltage amplitude was increased.
1 + g m RS = 3 RS =
Whites, EE 320
Lecture 33
Page 1 of 8
I
vO
vI
vO
vO
vI
vI
Common source
Common gate
-VSS
Common drain
(source follower)
Whites, EE 320
Lecture 33
Page 2 of 8
(Fig. 6.18a)
In this circuit, Q2 and Q3 form a PMOS current mirror. Because
both PMOS and NMOS devices are used in this circuit, it is
called a complementary MOS (CMOS) circuit.
In addition to forming part of the current mirror, Q2 also
functions as the current source load (aka active load) for Q1.
Whites, EE 320
Lecture 33
Page 3 of 8
(Fig. 6.18b)
Referring to the CS amplifier circuit above in Fig. 6.18(a), when
i = I REF then VGD 2 = 0 (by symmetry with Q1). This implies that
v = VSG , which is the Q point shown in Fig. 6.18(b).
Furthermore, it is useful to observe the graphical construction of
the transfer function vO/vI for this amplifier, as illustrated in
Figs. 6.18(c) and (d) shown below. The drain currents of Q1 and
Q2 are the same. The operating point of the amplifier is found
Whites, EE 320
Lecture 33
Page 4 of 8
= iD 2
(Fig. 6.18c)
Collecting these intersections from this figure as vGS1 ( = vI )
changes, we can construct point-by-point the transfer
characteristic curve for this amplifier:
(Fig. 6.18d)
From this plot, we can see that Region III shows a linear
relationship between vO and vI. This is the region where the
circuit of Fig. 6.18(a) can be used as a linear amplifier.
Whites, EE 320
Lecture 33
Page 5 of 8
vgs1
g m1vgs1
vo
Rout
vo = g m1vgs1 ( ro1 || ro 2 )
(2)
vgs1 = vi
(3)
Whites, EE 320
Lecture 33
W
2 k n
L 1
Avo =
1
1
+
VA1 VA 2
Page 6 of 8
1
I REF
(5)
Since ro1 and ro2 are usually large, this Avo gain is typically
relatively large (approximately -20 to -100, or so).
Neat! We have incorporated the effects of relatively large
resistance for this amplifier without having to actually construct
a large resistor.
From the small-signal model we see from inspection that
Rout = ro1 || ro 2
Summary for CMOS CS amplifier:
1. Very large input resistance.
2. Very large output resistance.
3. Potentially large small-signal voltage gain.
Whites, EE 320
Lecture 33
Page 7 of 8
ro1 =
| VA |1 VAn
8 1.6
=
=
= 128 k
I D1
I REF 100 106
Whites, EE 320
Lecture 33
Page 8 of 8
V
V
This value represents the largest gain. The gain will be
reduced when an actual load is attached to the amplifier.
Avo = 81.4
Whites, EE 320
Lecture 34
Page 1 of 9
(Fig. 4.45a)
Whites, EE 320
Lecture 34
Page 2 of 9
common gate amplifier: Rin, Av, Avo, Gv, Gi, Ais, and Rout. To
begin, we construct the small-signal equivalent circuit:
(Fig. 4.45b)
The T model was used since we ignored ro while Rsig appears in
series with 1/gm.
Input resistance, Rin. Because the gate is grounded, we can see
directly from this small-signal equivalent circuit that
1
Rin =
(4.91),(1)
gm
Whites, EE 320
Lecture 34
Page 3 of 9
i = g m vgs
Whites, EE 320
Lecture 34
vi = vgs
Page 4 of 9
(3)
vo
v v
v
= i o = i Av
vsig vsig vi vsig
N
(7)
= Av
(4.96b),(9)
Whites, EE 320
Lecture 34
Page 5 of 9
Summary
In summary, we find for the CG small-signal amplifier:
o A non-inverting amplifier.
Whites, EE 320
Lecture 34
Page 6 of 9
Whites, EE 320
Lecture 34
Page 7 of 9
(Fig. E4.30)
Using (4.71)
gm =
2 I D 2 0.5 m
=
= 1 mS
VOV 2.5 1.5
vO
Whites, EE 320
Lecture 34
Page 8 of 9
g m vgs
G
4.7 M
vo
RD=
15 k
RL=
15 k
ig=0
1/gm
Rout
Rsig
+
vsig
-
Rin
1
1
= 3 = 1 k.
g m 10
V
V
V
V
Whites, EE 320
Lecture 34
g m ( RD || RL )
Av
7.5
=
=
N
1 + g m Rsig ( 4) 1 + g m Rsig 1 + 103 50
V
= 7.14
V
RD
15
1 A
=
=
From (12), Gi =
RD + RL 15 + 15 2 A
From (9), Gv =
Page 9 of 9
(15)
Whites, EE 320
Lecture 35
Page 1 of 12
Whites, EE 320
Lecture 35
Page 2 of 12
D
n-channel
n+
etio
Depl
ion
n reg
n+
(Fig. 1)
Whites, EE 320
Lecture 35
Page 3 of 12
where
Vt 0 is the threshold voltage for VSB = 0
2 f is the material dependent Fermi potential (often 2 f 0.6
V for NMOS)
Whites, EE 320
Lecture 35
Page 4 of 12
Whites, EE 320
Lecture 35
g mb
iD
vBS
Page 5 of 12
(4.75),(4)
vGS = constant
vDS = constant
g mb = g m [S]
(4.76),(5)
where is dimensionless and typically lies in value between 0.1
and 0.3.
vgs
g m vgs
g mb vbs
vbs
(Fig. 4.41b)
Alternatively, the small-signal T model for the MOSFET
amplifier including the body effect is:
g m vgs + g mb vbs
( g m + g mb )
Whites, EE 320
Lecture 35
Page 6 of 12
(Fig. 6.27a)
Note here that the body terminal is not connected to the source
terminal, but rather is connected to the lowest voltage in the
circuit (ground). Because of this we need to account for the body
effect in the small-signal T model of this amplifier:
Whites, EE 320
Lecture 35
Page 7 of 12
io
g m vgs
vgs , vbs
vo
g mb vbs
iro
ii
vi
(Fig. 2)
Notice that weve incorporated the body effect into the T smallsignal model of the MOSFET. Because the gate and body are
both grounded in Fig. 2, then
vgs = vbs
(6)
Consequently, the body effect in this CG amplifier can be
completely accounted for by simply replacing gm of the
MOSFET with g m + g mb . That is,
g m g m + g mb = (1 + ) g m
(7)
Whites, EE 320
Lecture 35
Page 8 of 12
Whites, EE 320
Lecture 35
vo = io RL =
N ii RL
Page 9 of 12
(6.89),(13)
(9)
vi = ii Rin
(6.90),(14)
Dividing these two equations we arrive at the partial smallsignal voltage gain
RL 1 + ( g m + g mb ) ro
vo RL
(6.91),(15)
Av =
=
N
vi Rin (12)
ro + RL
(6.93),(17)
Since the body effect tends to reduce Rin, we see from this
expression that it tends to increase Gv.
Whites, EE 320
Lecture 35
Page 10 of 12
Output resistance, Rout. To determine Rout from the smallsignal circuit above we set vsig = 0 and apply a fictitious AC
voltage source vx at the output as shown:
ix
vgs
i1
iro
vs
(Fig. 3)
We can see that with vx attached, the voltage vgs will not
usually be zero. This means the current in the dependent
current source is also not zero. Consequently, we need to
analyze this circuit including the effects of the dependent
current source to determine the output resistance.
Employing KCL at the drain terminal
ix + iro = ( g m + g mb ) vgs
It is easy to see in Fig. 3 that
(19)
vs vx
(20)
ro
Applying KCL to the supernode indicated in Fig. 3, we find
i1 = ix
(21)
Because
iro =
Whites, EE 320
Lecture 35
Page 11 of 12
vs = i1Rsig =
N ix Rsig
(22)
(21)
(23)
(24)
Whites, EE 320
Lecture 35
Page 12 of 12
Example N35.1.
Whites, EE 320
Lecture 36
Page 1 of 10
(Fig. 4.46a)
Whites, EE 320
Lecture 36
Page 2 of 10
g m vgs
Rsig
+
vsig
-
ii
+ ig=0
vi
RG vgs
Rin
1/gm
S
vo
Rout
RL
ro
io
(Fig. 4.46b)
(3)
Whites, EE 320
Lecture 36
Page 3 of 10
vo
v v
v
= i o = i Av
vsig vsig vi vsig
N
(7)
= Av
and using (1) and (4) gives the overall small-signal voltage
gain of this common drain amplifier to be
v
RG
RL || ro
Gv o =
(4.104),(8)
vsig RG + Rsig RL || ro + 1 g m
Again, notice that if ro RL and RL 1 g m , as well as
RG Rsig , then
Whites, EE 320
Lecture 36
Page 4 of 10
Gv 3 1
(9)
(13)
If ro RL and g m RL 1, then
Gi
which likely is quite large.
RG
RL
(14)
Whites, EE 320
Lecture 36
Page 5 of 10
Output resistance, Rout. To determine Rout from the smallsignal circuit above we set vsig = 0 and apply a fictitious AC
voltage source vx at the output as shown:
g m vgs
(Fig. 1)
Notice that the gate terminal has zero voltage because vsig = 0
and isig = 0 .
By definition
Rout
vx
ix
(16)
Whites, EE 320
Lecture 36
Page 6 of 10
We can see that with vx attached, the voltage vgs will not
usually be zero. This means the current in the dependent
current source is also not zero.
In such instances, we would normally need to analyze this
circuit to find the voltage vx in terms of ix, and then apply (16)
to determine the output resistance of this amplifier.
However, in this case both terminals of the dependent current
source are grounded so it makes no contribution to the output
resistance. By inspection, the output resistance is simply
1
Rout = ro ||
(17)
gm
Whites, EE 320
Lecture 36
Page 7 of 10
Whites, EE 320
Lecture 36
Page 8 of 10
(Fig. E4.30)
Using (4.71)
gm =
2 I D 2 0.5 m
=
= 1 mS
VOV 2.5 1.5
Whites, EE 320
Lecture 36
Page 9 of 10
g m vgs
1
= 150k ||103 = 0.993 k (w/ ro), or
gm
1
= 1 k (w/o ro).
gm
ro
150k
V
=
=
(w/ ro),
0.9993
ro + 1 g m 150k + 1 103
V
V
(w/o ro).
V
Whites, EE 320
Lecture 36
Page 10 of 10
RL || ro
15k ||150k
V
=
=
0.932
RL || ro + 1 g m 15k ||150k + 103
V
RL || ro
15k
V
=
=
(w/o ro).
(w/ ro), or Av =
0.938
RL || ro + 1 g m 15k + 103
V
From (4), Av =
From (8),
RG
RL || ro
4.7M
15k ||150k
Gv =
=
RG + Rsig RL || ro + 1 g m 4.7M + 1M 15k ||150k + 103
RG
RL
V
= 0.768
(w/ ro), or Gv =
RG + Rsig RL + 1 g m
V
4.7M
15k
V
=
=
0.
77
3
(w/o ro).
3
4.7M + 1M 15k + 10
V
From (13),
g m ( RL || ro ) RG
103 (15k ||150k ) 4.7M
A
Gi =
313.3
=
=
1 + g m ( RL || ro ) RL 1 + 103 (15k ||150k ) 15k
A
g m RL RG
103 15k 4.7M
A
(w/ ro), or Gi =
=
=
29
3.8
1 + g m RL RL 1 + 103 15k 15k
A
(w/o ro).
Whites, EE 320
Lecture 37
Page 1 of 10
(Fig. 4.53)
The operation of this circuit can be summarized as:
When vI is low, QN is off, QP is on vO is high
When vI is high, QN is on, QP is off vO is low
Conceptually, this CMOS circuit is intended to function as
complementary switches shown in Fig. 1.32(a):
Whites, EE 320
Lecture 37
Page 2 of 10
(Fig. 1.32a)
The directions of the arrows indicate the complementary nature
of the two switches: when one is closed the other is open, and
vice versa.
Whites, EE 320
Lecture 37
Page 3 of 10
The vDS values are different for the two transistors, of course. By
KVL,
(1)
VDD vSDP = vDSN
How we interpret this equation can give us the answers on how
to plot the QP characteristic curve in the iDN-vDSN plane:
The minus sign in (1) tells us to flip the QP characteristic
curve about the vertical axis.
The VDD term tells us to shift this flipped curve VDD units to
the right.
(This is the same process we used for the graphical solution of
the CMOS common source amplifier discussed in Lecture 33.)
Whites, EE 320
Lecture 37
Page 4 of 10
Whites, EE 320
Lecture 37
Page 5 of 10
Whites, EE 320
Lecture 37
Page 6 of 10
Whites, EE 320
Lecture 37
Page 7 of 10
1
W
iDN = kn vI Vtn vO vO2
N
2N
L n vN
2
vDS
vDS
GS
iDN
1 W
= kn vI Vtn
2 L n vN
GS
vO vI Vtn
(4.142),(2)
vO vI Vtn
(4.143),(3)
= k p VDD vI Vtp
L p
vGS
iDP
1 W
= k p VDD vI Vtp
2 L p
vGS
(VDD vO ) 1 VDD vO
2
vSD
vSD
v v +V
I
tp
O
(4.144),(4)
vO vI + Vtp
(4.145),(5)
For the digital logic inverter circuit, these two currents must be
equal:
iDN = iDP
(6)
To construct the complete characteristic curve for this logic
inverter, we solve (6) for vO using (2) through (5) as vI varies
from 0 to VDD.
In the case of symmetrical MOSFETs in which Vtn = Vtp and
kn (W L )n = k p (W L ) p , the voltage transfer characteristic
curve will also be symmetrical, as shown below in Fig. 4.56.
Whites, EE 320
Lecture 37
Page 8 of 10
Noise Margins
We can observe from this characteristic curve that there is a
range of input values (from 0 to VIL) that produce a high output
Whites, EE 320
Lecture 37
Page 9 of 10
Whites, EE 320
Lecture 37
Page 10 of 10