FREQUENCY RESPONSE OF AMPLIFIERS

* Effects of capacitances within transistors and in amplifiers

* Build on previous analysis of amplifiers from EENG 341
z Transistor DC biasing
z Small signal amplification, i.e. voltage and current gain
z Transistor small signal equivalent circuit
* Use in Bipolar and FET Transistors Amplifiers and their analysis
* Build on previous analysis of single time constant circuits
z Review simple RC, LC and RLC circuits
z Recall frequency dependent impedances for C and L
z Review frequency dependence in transfer functions
* Magnitude and phase
* Examine origins of frequency dependence in amplifier gain
z Identify capacitors and their origins; find the dominant C
z Determine equivalent R and determine RC time constant
z Use to describe approximately the amplifier’s frequency behavior
z Examine effects of other capacitors
* GOAL: Use results of analysis to modify circuit design to improve performance.

1
 Analysis of Amplifier Performance
* Previously analyzed
z DC bias point
z AC analysis (midband gain)
* Neglected all capacitances in
the transistor and circuit
* Gain at middle frequencies,
i.e. not too high or too low
in frequency
iB iC

DC bias or
quiescent point

vBE vCE

2
 Frequency Response of Amplifiers
* In reality, all amplifiers have a limited
range of frequencies of operation
z Called the bandwidth of the amplifier
z Falloff at low frequencies
* At ~ 100 Hz to a few kHz
* Due to coupling capacitors at the
input or output, e.g. CC1 or CC2
z Falloff at high frequencies
* At ~ 100’s MHz or few GHz
20 logT(ω) * Due to capacitances within the
transistors themselves.

Midband Gain
Equivalent circuit for bipolar transistor

3
 Frequency Response of Amplifiers
* First approximation – describe the amplifier’s high Low-Pass Network
and low frequency responses in terms of that of
single time constant (STC) circuits
z High frequency falloff –
Æ Like that of a low pass filter Vi
* Simple RC equivalent circuit
* Shunting capacitor shorts signal at the
output at high frequencies
z Low frequency falloff
High-Pass Network
Æ Like that of a high pass filter
* Simple RC equivalent circuit
* Series capacitor blocks output signal at Vi
low frequencies (acts like open circuit)
* Amplifier frequency analysis
z Determine equivalent R for each C
z Compare and find the most important 1 1 ⎧0 (short ) as ω → ∞
(dominant RC) combination ZC = = →⎨
sC jωC ⎩ ∞ (open) as ω → 0
ÆFind the dominant one (RC)
at high frequencies
ÆFind the dominant one (RC)
at low frequencies
4
 Review of Complex Numbers
* Complex numbers
z General form a + bj where
a = real part, b = imaginary part and j = − 1

z Magnitude of complex number M = a + bj = a 2 + b 2

z Phase of complex number ⎛b⎞
θ = tan −1 ⎜ ⎟
⎝a⎠
z Phasor form a + bj = Me jθ

* Complex number math

z Multiplication of two complex numbers a + bj = Me jθ and c + dj =Ne jφ
(a + bj )(c + dj ) = (ac − bd ) + j (bc + ad )
OR
(a + bj )(c + dj ) = (Me jθ )(Ne jϕ ) = MNe j (θ +ϕ )
z Reciprocal of a complex number
1 1 a − bj a − bj a b
= = 2 = 2 −j 2
a + bj a + bj a − bj a + b 2
a +b 2
a + b2
OR
1 1 1 − jθ
= jθ
= e
a + bj Me M

5
 Amplifier Transfer Function (Gain) - General Form
* A (s) = Gain Function (general form of amplifier transfer function)

A( s ) = AM FH ( s ) FL ( s )
z AM = midband gain (independent of frequency)
1
z FH(s) = high frequency function (acts like low pass filter) FH ( s ) =
1 + s / ωH

1
z FL(s) = low frequency function (acts like high pass filter) FL ( s ) =
1 + ωL / s
Magnitude

AM
FL(s) FH(s)

ωL ωH
6
Amplifier Transfer Function (Gain) - General Form

A( s ) = AM FH ( s ) FL ( s )

AM
FL(s) FH(s)

Coupling Transistor’s
Capacitors Capacitors

ωL ωH
7
 Frequency Response of MOSFET vs BJT Amplifiers
Equivalent circuit for MOSFET Equivalent circuit for bipolar transistor

Similar equivalent circuits

Common Source Common Emitter

Amplifier Amplifier
Corresponding
amplifier circuits

Gain Gain

Similar frequency
performance

8
 Amplifier Transfer Function (Gain) - General Form

A( s ) = AM FH ( s ) FL ( s )

Now we consider the

low frequency behavior.

Coupling Transistor’s
Capacitors Capacitors

ωL ωH
9
 Summary
* Examined origin of falloff in amplifier gain at low and high frequencies.
z Degradation in magnitude of the gain.
z Shift in phase of output relative to input.
* Due to presence of capacitors within the amplifier (Create poles and zeros).
z Coupling capacitors limit gain at low frequencies.
z Transistor’s capacitances limit gain at high frequencies.
* Examined and quantified the falloff due to single and multiple poles and zeroes.
z Bode plots of gain and phase shift with frequency.
* Next, we will apply this method of analysis to transistor amplifiers.
z Multiple capacitors so multiple RC combinations.
z Investigate how to determine which capacitors are most important and limit the
bandwidth.
z Examine how to change the amplifier to get better frequency performance.
Higher frequency operation before falloff (improved bandwidth).
Better low frequency behavior.

10
 Analysis of Bipolar Transistor Amplifiers
* Single stage amplifiers
z Common Emitter (CE)
z Common Base (CB)
z Emitter Follower (EF) (Common Collector)
* DC biasing
z Calculate IC, IB, VCE
z Determine related small signal equivalent circuit parameters
Transconductance gm
Input resistance rπ
* Low frequency analysis
z Gray-Searle (Short Circuit) Technique
Determine the pole frequencies ωPL1, ωPL2, ... ωPLn
z Determine the zero frequencies ωZL1, ωZL2, ... ωZLn
* High frequency analysis
z Gray-Searle (Open Circuit) Technique
Determine the pole frequencies ωPH1, ωPH2, ... ωPHn
z Determine the zero frequencies ωZH1, ωZH2, ... ωZHn

11
CE Amplifier Frequency Analysis - Long and Difficult Way
* DC analysis: IC , IB , VCE ; gm , rπ
* Draw ac equivalent circuit
* Substitute hybrid-pi model for
transistor
* Obtain KVL equations (at least one
for each capacitor in circuit (5))
* Solve set of 5 simultaneous equations
to obtain voltage gain AV = Vo /Vs
* Put expression in standard form for
gain AV(ω) = AVo FH(ω) FL(ω)
* Identify midband gain AVo
* Determine FH(ω) part and factor to
z Determine high frequency poles
ωPH1 and ωPH2
z Determine high frequency zeros
ωZH1 and ωZH2
* Determine FL(ω) part and factor to
z Determine low frequency poles
ωPL1, ωPL2 and ωPL3
z Determine high frequency zeros
FL ( s ) =
(s + ωZ 1 )(s + ωZ 1 )(s + ωZ 3 ) ωZL1, ωZL2 and ωZL3
⎛
⎜⎜1 +
s ⎞⎛
⎟⎟⎜⎜1 +
s ⎞
⎟
(s + ωP1 )(s + ωP 2 )(s + ωP 3 )
⎝ ωZ 1 ⎠⎝ ωZ 2 ⎟⎠ ⎛ ωZ 1 ⎞⎛ ωZ 2 ⎞⎛ ωZ 3 ⎞
FH ( s ) =
⎛ s ⎞⎛ s ⎞
⎜1 + ⎟⎜1 + ⎟⎜1 + ⎟ * Is there an easier way ? YES
=⎝
⎜⎜1 + ⎟⎟⎜⎜1 + ⎟⎟ s ⎠⎝ s ⎠⎝ s ⎠
⎝ ω P1 ⎠⎝ ω P 2 ⎠ ⎛ ω P1 ⎞⎛ ω P 2 ⎞⎛ ω P 3 ⎞
⎜1 + ⎟⎜1 + ⎟⎜1 + ⎟
⎝ s ⎠⎝ s ⎠⎝ s ⎠

12
 CE Amplifier - Starting Point is DC Analysis

* Q is quiescent point (DC bias point)

* Q needs to be in the active region
z I C = β IB
* If Q is in saturation (VCE < 0.3 V) , then
IC < β IB and there is little or no gain
from the transistor amplifier
* If the transistor is in the cutoff mode,
there is virtually no IC so there is no
gain, i.e. gm ¡ 0.
* Q depends on the choice of R1 and R2
since they determine the size of IB.
Active * Q point determines the size of the
region small signal parameters
Saturation Q point
z Transconductance gm = IC / VT
region VT = kBT/q = 26 mV
z Input resistance rπ = β / gm

Cutoff region

13
 Example of CE Amplifier - DC Analysis
* GIVEN: Transistor parameters:
z Current gain β = 200
z Base resistance rx = 65 Ω
z Base-emitter voltage VBE,active = 0.7 V
z Resistors: R1=10K, R2=2.5K, RC=1.2K, RE=0.33K

* Form Thevenin equivalent for base; given VCC = 12.5V

z RTh = RB = R1||R2 = 10K||2.5K = 2K
z VTh = VBB = VCC R2 / [R1+R2] = 2.5V
z DC Base Current (use KVL base loop)
IB = [VTh-VBE,active] / [RTh+(β +1)RE]
IB = 26 µA
* DC collector current IC = β IB
IC = 200(26 µ A) = 5.27 mA
* Transconductance gm = IC / VT ; VT = kBT/q = 26 mV
gm = 5.27 mA/26 mV = 206 mA/V
* Input resistance
rπ = β / gm = 200/[206 mA/V]= 0.97 K
R1 = 10K * Check on transistor region of operation (Find VCE)
R2 = 2.5K z KVL collector loop
RC = 1.2K z VCE = VCC - IC RC - (β +1) IB RE = 4.4 V
RE = 0.33K (okay since not close to zero volts, i.e. > 0.2V).

14
 CE Amplifier - Midband Gain Analysis
* Construct amplifier’s small signal ac equivalent circuit
(set DC supply to ground)
* Substitute small signal equivalent circuit
(hybrid-pi model) for transistor
* Neglect all capacitances
z Coupling and emitter bypass capacitors become shorts
at midband frequencies (~ 105 rad/s)
Why? Their impedances are negligibly small, e.g.
few ohms because CC1, CC2, CE are large,
e.g.~ few µF (10-6F)
1 1
ZC = ~ = 10 Ω
ωC (105 rad / s )(1µF )

Hybrid-Pi Model for BJT z Transistor capacitances become open circuits at

midband frequencies
Why? Their impedances are very large, e.g. ~
10’s M Ω because Cπ , Cµ are very small, e.g. ~ pF
(10-12 F)
1 1
ZC = ~ = 107 Ω
ωC (10 rad / s )(1 pF )
5

* Calculate small signal voltage gain AVo = Vo /Vs

15
 CE Amplifier - Midband Gain Analysis

rx Io
RL = 9 K
Vo
+ RC = 1.2 K
+ RE = 0.33K
Vs V
Vii rπ Vπ gmVπ
R1 = 10 K
-- R2 = 2.5 K
RL||RC RS = 5 K

V V V V
AVo = o = o π i Break voltage gain into a series of voltage ratios
Vs Vπ Vi Vs
Vo − g mVπ RL RC
=
( ) ( )
= − g m RL RC = − (206 mA / V ) 1.2 K 9 K = −218 ( )
Vπ Vπ
Vπ rπ 0.97 K
= = = 0.94
Vi rx + rπ 0.97 K + 0.065 K
Vi
=
[(rx + rπ ) RB ]=
[
(0.97 K + 0.065 K ) 2.0 K
=
]
0.68 K
= 0.12
[
Vs Rs + (rx + rπ ) RB ] [
5 K + (0.97 K + 0.065 K ) 2.0 K ]
5.68 K
AVo = (− 218 )(0.94 )(0.12 ) = −24.6 V / V Negative sign means output signal is
180o out of phase with the input signal.
AVo (dB ) = 20 log( − 24.6 ) = 27.7 dB

16
 Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
* Draw AC equivalent circuit at low frequency
z Include coupling and emitter bypass capacitors CC1,CC2, CE
z Substitute AC equivalent circuit for transistor
(hybrid-pi for bipolar transistor)
z Ignore (remove) all transistor capacitances Cπ , Cµ
* Turn off signal source, i.e. set Vs= 0
z Keep source resistance RS in circuit (do not remove)
* Analyze the circuit one capacitor Cx at a time
z Replace all other capacitors with short circuits
z Solve remaining circuit for equivalent resistance Rx seen by
the selected capacitor
1
z Calculate pole frequency using ω Px =
Rx C x
z Repeat process for each capacitor finding equivalent
resistance seen by it and find corresponding pole frequency
* Determine which is dominant (largest) low frequency pole
* Calculate the final, low 3dB frequency using
ω LP = ∑ ω Px = ω P1 + ω P 2 ...ω Pn = ∑ 1
RxC x
17
 Common Emitter - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
AC equivalent circuit
RS CC1 rX CC2
at low frequency Vo

VS RB rπ Vπ gmVπ RL
RC

RE CE

* Input coupling capacitor CC1 = 2 µF * Output coupling capacitor CC2 = 3 µF

VX rX VX
IX IX Vo
RS RB rπ RC RL
Vπ

VX
RC 2 = = RL + RC = 9 K + 1.2 K = 10.2 K
= RS + RB (rx + rπ ) = 5 K + 2 K (0.065 K + 0.97 K ) ) = 5.7 K
VX IX
RC1 =
IX
1 1
1 1 ωPL 2 = = = 33 rad / s
ω PL1 = = = 88 rad / s RC 2CC 2 10.2 K (3µF )
RC1CC1 5.7 K (2 µF )

18
 Common Emitter - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
KCL at node E
V
REx = x =? I x = I e − I π − g mVπ
RS CC1 rX CC2
Ix Vo

Vx − Vx VS RB rπ Vπ gmVπ RC RL
Ie = Iπ =
RE rx + rπ + Rs RB
− rπ Vx
Vπ = I π rπ =
rx + rπ + Rs RB RE CE
Vx Vx g m rπ Vx
Ix = + +
RE rx + rπ + Rs RB rx + rπ + Rs RB
1 I 1 1 g m rπ Emitter bypass capacitor CE = 12 µF
= x = + + rX
REx Vx RE rx + rπ + Rs RB rx + rπ + Rs RB
Iπ
1 1 + g m rπ +
= + Recall β = gmrπ RS
RE rx + rπ + Rs RB RB rπ Vπ gmVπ
⎛ r + r + Rs RB ⎞
REx = RE ⎜⎜ x π ⎟
⎟
Ix
β +1 REx
⎝ ⎠
Ie VX
⎛ 0.065 K + 0.97 K + 5 K 2 K ⎞ RE
= 0.33K ⎜⎜ ⎟ = 0.016 K
⎟
⎝ 201 ⎠ Low 3db frequency
1 1
ω PL 3 = = = 5,342 rad / s ω PL = ω PL1 + ω PL 2 + ω PL3 = 88 + 33 + 5342 = 5463 rad / s
REx C E 0.016 K (12 µF )
19
 Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
* Draw AC equivalent circuit at high frequency
z Substitute AC equivalent circuit for
transistor (hybrid-pi model for transistor
with Cπ, Cµ )
z Remove coupling and emitter bypass
capacitors CC1, CC2, CE (consider as shorts).
z Turn off signal source, i.e. set Vs = 0
z Keep source resistance RS in circuit
z Neglect transistor’s output resistance ro
* Consider the circuit one capacitor Cx at a time
z Replace all other transistor capacitors with
Cπ comes from Emitter-Base p-n junction. open circuits
z Solve remaining circuit for the equivalent
Cµ comes from Base-Collector p-n junction.
resistance Rx seen by the selected capacitor
1
z Calculate the pole frequency using ω PHx =
Rx C x
z Repeat process for each capacitor

* Calculate the final high 3dB frequency using

−1 −1

∑
⎡ 1 ⎤ ⎡ 1 1 1 ⎤ 1
ω PH = ⎢ ⎥ =⎢ + ... + ⎥ =
⎣ ω Px ⎦ ⎣ ω PH 1 ω PH 2 ω PHn ⎦
∑R C x x

20
 Common Emitter - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

Vo

VS
CE shorts RE
at high frequencies.

AC equivalent circuit at High frequency

VS Vπ Vo

21
 Common Emitter - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

VS Vπ Vo

* Equivalent circuit for Capacitor Cπ = 17 pF

IX
= rπ (rx + RB RS ) = 0.97 K (0.065 K + 2 K 5 K ) = 0.59 K
VX
Rπx =
IX
Rπx Cπ = 0.59 K (17 pF ) = 1.0 x10 −8 sec
VX 1 1
ω PH 1 = = −8
= 1.0 x108 rad / s
Rπx Cπ 1.0 x10 sec

Note: This frequency is very high due to the

very small size of the capacitor.

22
 Common Emitter - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

Vπ Vπ ⎡1 1 ⎤
At node B' I x = + so Vπ = I x ⎢ + ⎥
rπ rx + RB RS ⎣ rπ rx + RB RS ⎦
Vπ V V
At node C I x = − g mVπ − o − o
RC RL
Using − Vπ + Vx + Vo = 0 or Vo = Vπ − Vx
Vπ − Vx Vπ − Vx ⎡ 1 1⎤ ⎡1 1⎤
I x = − g mVπ − − = −Vπ ⎢ gm + + ⎥ + Vx ⎢ + ⎥
RC RL ⎣ RC RL ⎦ ⎣ RC RL ⎦
* Equivalent circuit for Capacitor Cµ = 1.3 pF V
Substituting for Vπ and solving for Rµx = x we get
Ix
VX
B/ IX C ⎡ ⎛
Rµx = (RC RL )⎢1 + ⎜⎜ g m +
1 1 ⎞
( )⎤
+ ⎟⎟ rπ (rx + RB RS ) ⎥
⎣ ⎝ RC RL ⎠ ⎦

Vπ
⎡ ⎛
= (1.2K 9K )⎢1 + ⎜ 206mA / V +
1
+
1 ⎞
( ⎤
⎟ 0.97K (0.065K + 2K 5K ) ⎥
1.2K 9K ⎠
)
⎣ ⎝ ⎦
= 130K
RµxCµ = 130K (1.3 pF ) = 1.7 x10− 7 sec
1 1
ωPH 2 = = −7
= 5.9 x106 rad / s
RµxCµ 1.7 x10 sec
VX
Need to solve for RµX = Final high frequency 3dB frequency is
IX 1 1
ωPH = = −8 −7
= 5.6 x106 rad / s
RπxCπ + RµxCµ 1.0 x10 + 1.7 x10 sec

23
 Common-Base (CB) Amplifier
* DC biasing
z Calculate IC, IB, VCE
z Determine related small signal
equivalent circuit parameters
Transconductance gm
Input resistance rπ
* Midband gain analysis
* Low frequency analysis
z Gray-Searle (Short Circuit)
Technique
Determine pole frequencies
ωPL1, ωPL2, ... ωPLn
z Determine zero frequencies
ωZL1, ωZL2, ... ωZLn
* High frequency analysis
z Gray-Searle (Open Circuit)
Technique
Determine pole frequencies
Input at emitter, output at collector. ωPH1, ωPH2, ... ωPHn
z Determine zero frequencies
ωZH1, ωZH2, ... ωZHn

24
 CB Amplifier - DC Analysis (Same as CE Amplifier)
* GIVEN: Transistor parameters:
z Current gain β = 200
z Base resistance rx = 65 Ω
z Base-emitter voltage VBE,active = 0.7 V
z Resistors: R1=10K, R2=2.5K, RC=1.2K, RE=0.33K

* Form Thevenin equivalent for base; given VCC = 12.5V

z RTh = RB = R1||R2 = 10K||2.5K = 2K
z VTh = VBB = VCC R2 / [R1+R2] = 2.5V
z KVL base loop
IB = [VTh-VBE,active] / [RTh+(β +1)RE]
IB = 26 µA
* DC collector current IC = β IB
IC = 200(26 µ A) = 5.27 mA
* Transconductance gm = IC / VT ; VT = kBT/q = 26 mV
gm = 5.27 mA/26 mV = 206 mA/V
* Input resistance
rπ = β / gm = 200/[206 mA/V]= 0.97 K
* Check on transistor region of operation
R1 = 10K
z KVL collector loop
R2 = 2.5K
z VCE = VCC - IC RC - (β +1) IB RE = 4.4 V
RC = 1.2K (okay since not close to zero volts).
RE = 0.33K

25
 CB Amplifier - Midband Gain Analysis
* Construct small signal ac equivalent circuit
(set DC supply to ground)
* Substitute small signal equivalent circuit
(hybrid-pi model) for transistor
* Neglect all capacitances
z Coupling and emitter bypass capacitors become shorts
at midband frequencies (~ 105 rad/s)
Why? Impedances are negligibly small, e.g. few
-6
ohms because CC1, CC2, CE ~ few µF (10 F)
1 1
ZC = = = 10 Ω
ωC (105 rad / s )(1µF )
High and Low Frequency AC Equivalent Circuit
z Transistor capacitances become open circuits at
midband frequencies
Why? Impedances are very large, e.g. ~ 10’s M Ω
because Cπ , Cµ ~ pF (10-12 F)
1 1
ZC = = = 10 7 Ω
ωC (10 rad / s )(1 pF )
5

* Calculate small signal voltage gain AVo = Vo /Vs

26
 CB Amplifier - Midband Gain Analysis
RL = 9 K
V − Ve
Iπ = π = RC = 1.2 K
rπ rx + rπ Iπ
RE = 0.33K
Vπ − rπ βIπ
= R1 = 10 K
Ve rx + rπ
R2 = 2.5 K
re + RS = 5 K
Ve
_

V V V V
Equivalent resistance re
AVo = o = o π e Ve
Vs Vπ Ve Vs re =
Vo − gmVπ (RL RC )
Ie
= = −gm (RL RC ) = −(206mA/ V )(1.2K 9K ) = −218 KCL at node E
Vπ Vπ Vπ
I e + g mVπ + =0
Vπ rπ Iπ − 0.97K rπ
= = = −0.94
Ve − Iπ (rx + rπ ) 0.97K + 0.065K ⎛ ⎞ 1 ⎛ 1 + g m rπ ⎞
I e = −Vπ ⎜⎜ g m +
⎟⎟ = −Vπ ⎜⎜ ⎟⎟
Ve
=
[RE re ] = [0.33K 0.0051K ] = 0.0050K = 0.001 ⎝ ⎠rπ ⎝ rπ ⎠
Vs Rs + [RE re ] 5K + [0.33K 0.0051K ] 5.0050K V V
re = e = − e
rπ ⎛ r + r ⎞ rπ
= −⎜⎜ − x π ⎟⎟
Vπ 1 + g m rπ rπ ⎠ 1 + g m rπ
AVo = (− 218)(− 0.94)(0.001) = 0.20 V / V Voltage gain is
Ie ⎝
rx + rπ r +r 0.065K + 0.97 K
AVo (dB) = 20log(0.20) = −14dB less than one ! =
1 + g m rπ
= x π =
1+ β 1 + 200
= 0.0051 K

27
 What Happened to the CB Amplifier’s Midband Gain?

* Source resistance Rs = 5K is
killing the gain.
z Why? Rs >> re = 0.0051 K
so Ve/Vs<<1
re + * Need to use a different signal
Ve
_ source with a very low source
resistance Rs , i.e. ~ few ohms
AVo = (−218 )(−0.94 )(0.001) = 0.20 V / V
Ve [
RE re ] [
0.33 K 0.0051K ]
0.0050 K
* Why is re so low?
= = = = 0.001
[
Vs Rs + RE re ] [
5 K + 0.33 K 0.0051K 5.0050 K ] z Vs drives formation of Ve
z Ve creates Vπ across rπ
New signal source with low resistance z Vπ turns on dependent
For Rs = 5Ω current source
Ve
=
[
RE re
=
] [
0.33 K 0.0051K
=
]
0.0050 K
= 0.5 z Get large Ie for small Ve
[
Vs Rs + RE re ] [
0.005 K + 0.33K 0.0051K ]
0.005 K + 0.005 K
so re =Ve/Ie is very small.
and
AVo = (− 218 )(− 0.94 )(0.5) = 102.5 V / V Voltage gain is
AVo ( dB ) = 20 log(102.5) = 40.2dB now much bigger than one !

28
 Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
* Draw low frequency AC circuit
z Substitute AC equivalent circuit for transistor
(hybrid-pi for bipolar transistor)
z Include coupling and base capacitors CC1, CC2, CB
z Ignore (remove) all transistor capacitances Cπ , Cµ
* Turn off signal source, i.e. set Vs= 0
z Keep source resistance RS in circuit (do not remove)
* Consider the circuit one capacitor Cx at a time
z Replace all other capacitors with short circuits
z Solve remaining circuit for equivalent resistance Rx seen
by the selected capacitor
1
z Calculate pole frequency using ω Px = R C
x x
z Repeat process for each capacitor finding equivalent
resistance seen and the corresponding pole frequency
* Determine the dominant (largest) pole frequency
* Calculate the final low pole frequency using
ω LP = ∑ ω Px = ω P1 + ω P 2 ...ω Pn = ∑ 1
RxC x
29
 Common Base - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
Low Frequency AC Equivalent Circuit
RxCB = RB (rx + Ri )
Vi Vi V Vπ
Ri = = = rπ i sin ce Iπ =
Iπ Vπ / rπ Vπ rπ
⎡ ⎛1 ⎞ ⎤
Vi = Vπ + (Iπ + g mVπ )(RE RS ) = Vπ ⎢1 + ⎜⎜ + g m ⎟⎟(RE RS )⎥
⎣ ⎝ rπ ⎠ ⎦
⎡ ⎛1 ⎞ ⎤
Vπ ⎢1 + ⎜⎜ + g m ⎟⎟(RE RS )⎥
⎝ rπ ⎠ ⎦ = r + (1 + g r )(R R )
Ri = rπ ⎣ π m π E S
Vπ
* Base capacitor CB = 12 µF
RxCB = RB [rx + rπ + (1 + g m rπ )(RE RS )]
Vx Ix Iπ
= 2K [0.065K + 0.97K + (201)(0.33K 0.005K )]
+ Vo
Vπ = 2K 2.03K = 1.0K

Vi CB RxCB = 12µF (1.0K ) = 1.2 x10−2 sec

1 1
ωPL1 = = = 83 rad / s
RxCB CB 1.2 x10−2
RxCB _
Ri
30
 Common Base - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
Input coupling capacitor CC1 = 2 µF

V V
RxCC1 = x = Rs + RE re re = e
Ix Ie
I e = − Iπ − g mVπ = −(1 + g m rπ )Iπ
Ve = − Iπ (rx + rπ )
V − Iπ (rx + rπ ) (r + r )
re = e = = x π
I e − (1 + g m rπ )Iπ (1 + g m rπ )
⎡ r +r ⎤
RxCC1 = Rs + RE ⎢ x π ⎥
⎣ 1 + g m rπ ⎦ Iπ
⎡ 0.065 K + 0.97 K ⎤ Vo
= 0.005 K + 0.33K ⎢ ⎥
⎣ 201 ⎦ Vπ
Ve
= 0.005 K + 0.0051K = 0.010 K
Vx
Ie
CC1RxCC1 = 2µF (0.010 K ) = 2.0 x10 −5 sec
1 1
re +
ω PL 2 = = = 5.0 x10 4 rad / s Ve Ix Rs
− 5
CC1RxCC1 2.0 x10 sec _

31
 Common Base - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique

* Output coupling capacitor CC2 = 3 µF * Low 3dB frequency

VX ω PL = ω PL1 + ω PL 2 + ω PL3
Vo
= 83 + 50,000 + 33 = 50,116 rad / s
RC RL

RC 2 = RL + RC = 9 K + 1.2 K = 10.2 K Dominant low frequency

ω PL3 =
1
=
1
= 33 rad / s pole is due to CC1 !
RC 2CC 2 10.2 K (3µF )

32
 Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
* Draw high frequency AC equivalent circuit
z Substitute AC equivalent circuit for transistor
(hybrid-pi model for transistor with Cπ, Cµ)
z Consider coupling and emitter bypass
capacitors CC1, CC2, CB as shorts
z Turn off signal source, i.e. set Vs = 0
z Keep source resistance RS in circuit
z Neglect transistor’s output resistance ro
* Consider the circuit one capacitor Cx at a time
z Replace all other transistor capacitors with
open circuits
z Solve remaining circuit for equivalent
resistance Rx seen by the selected capacitor
z Calculate pole frequency using ω PHx =
1
RxC x
z Repeat process for each capacitor

* Calculate the final high frequency pole using

−1 −1

∑
⎡ 1 ⎤ ⎡ 1 1 1 ⎤ 1
ω PH = ⎢ ⎥ =⎢ + ... + ⎥ =
⎣ ω Px ⎦ ⎣ ω PH 1 ω PH 2 ω PHn ⎦
∑R C x x

33
 Common Base - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

High frequency AC equivalent circuit

NOTE: We neglect rx here

since the base is grounded.
This simplifies our analysis,
but doesn’t change the
results appreciably.
34
 Common Base - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
Ve
Ze = Ve = −Vπ KCL at node E gives
Ie
Ve Ve ⎡1 ⎤
Ie = + − g mVπ = Ve ⎢ + sCπ + g m ⎥
rπ ⎛ 1 ⎞ ⎣ rπ ⎦
⎜ sC ⎟
Ze ⎝ π ⎠
+ ⎡1 + g m rπ ⎤
Ve = Ve ⎢ + sCπ ⎥
_ ⎣ rπ ⎦
Ve Ve 1
Ze = = =
Replace this Ie ⎡1 + g m rπ ⎤ ⎡1 + g m rπ ⎤
Ve ⎢ + sCπ ⎥ ⎢ + sCπ ⎥
with this. ⎣ rπ ⎦ ⎣ rπ ⎦
1 ⎡ rπ ⎤ ⎛ 1 ⎞
* Equivalent circuit for Ze = =⎢ ⎥ ⎜ sC ⎟
⎡ ⎤ ⎣1 + g m rπ ⎦ ⎝ π ⎠
⎢ ⎥
⎢ 1 1 ⎥
+
⎢ ⎛ rπ ⎞ ⎛ 1 ⎞⎥
⎢ ⎜⎜ ⎟⎟ ⎜ sC ⎟ ⎥
⎝ π ⎠
⎣⎢ ⎝ 1 + g m rπ ⎠ ⎦⎥ Parallel combination
of a resistor and
So Z e = re Z Cπ capacitor.

where
rπ r
Ze re = = π
1 + gmr 1 + β
35
Common Base - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

Turn off signal source

when finding resistance
seen by capacitor.

* Pole frequency for Cπ =17pF

RxCπ = re RE Rs
rπ 0.97 K
re = = = 0.0048 K = 4.8Ω
1 + g m rπ 1 + 200
RxCπ = 0.0048 K 0.33K 0.005 K = 0.0024 K = 2.4Ω
1 1 1
ω PH 1 = = = = 2.5 x1010 rad / s
RxCπ Cπ 2.4Ω(17 pF ) 4.1x10 −11 s

36
 Common Base - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

* Pole frequency for Cµ =1.3pF

RxCµ = RC RL
RxCµ = 1.2 K 9 K = 1.05 K
1 1 1
ω PH 2 = = = = 7.1x108 rad / s
RxCµ Cµ 1.05 K (1.3 pF ) 1.4 x10 s
− 9

* Equivalent circuit for Capacitor Cµ = 1.3 pF * High 3 dB frequency

1 1
ω PH = =
RxCπ Cπ + RxCµ C µ 4.1x10 −11 + 1.4 x10 −9
1
=0 ω PH = −9
= 6. 9 x10 8
rad / s
1.44 x10 s

Rs || RE || r π Dominant high frequency

pole is due to Cµ !

37
 Comparison of CB to CE Amplifier
CE (with RS = 5K) CB (with RS = 5Ω)
⎡ r ⎤⎡ (rx + rπ ) RB [ ] ) [ [ ] ]
⎤ ⎛ − rπ ⎞
V V V V
[
AVo = o = o π i = − gm RL RC ⎢ π ⎥⎢ ( )] [ ] ⎥
V V V
AVo = o π e = − g m RL RC ( RE re
⎜ ⎟
⎣ rx + rπ ⎦⎢⎣ Rs + (rx + rπ ) RB Rs + RE re ⎜⎝ rx + rπ ⎟
Midband Gain Vs Vπ Vi Vs ⎥⎦ Vπ Ve Vs ⎠
AVo = (− 218)(0.94)(0.12) = −24.6 V / V AVo = (− 218)(− 0.94 )(0.5) = +102.4 V / V
AVo(dB) = 20log(− 24.6 ) = 27.7dB AVo (dB ) = 40.2dB
1 1
ωZP1 = ωZP2 =0 ωZP3 = = = 42 rad / s
ωZP1 = ωZP2 =0 ωZP3 =
1
=
1
= 252 rad / s RBCB 2K(12µF )
RECE 0.33K (12µF )
1 1
Low Frequency ωPL1 =
[
1
=
]
1
RS + RB (rx + rπ ) CC1 5.7K (2µF )
= 88 rad / s
ωPL1 =
{RB [rx + rπ + (1+ gmrπ )(RE RS )]}CB =
12µF(1K )
= 83 rad / s

Poles and Zeros 1 1 ωPL2 =

1
=
1
= 5.0x104 rad / s
ωPL2 = =
(RL + RC )CC2 10.2K (3µF )
= 33 rad / s ⎧⎪ ⎡ r + r ⎤⎫⎪ 2µF(0.010K )
CC1⎨Rs + RE ⎢ x π ⎥⎬
ωPL3 =
1
=
1
= 5,342 rad / s ⎪⎩ ⎣1+ gmrπ ⎦⎪⎭
⎡ ⎛ rx + rπ + Rs RB ⎞⎤ 0.016K (12µF ) 1 1
⎢RE ⎜ ⎟⎥CE ωPL3 = = = 33 rad / s
⎢
⎣
⎜
⎝ β +1 ⎟⎥
⎠⎦ (RL + RC )CC2 10.2K(3µF)
g m 206 mA / V
ωZH 1 = ∞, ω ZH 2 = = = 1.6 x1011 rad / s ω ZH 1 = ∞, ω ZH 2 = ∞
Cµ 1.3 pF
High Frequency 1 1 ω PH 1 =
1
=
1
= 2.5 x1010 rad / s
Poles and Zeroes
ω PH 1 =
[ =
]
rπ (rx + RB RS ) Cπ 0.59 K (17 pF )
= 1.0 x108 rad / s
[ ]
re RE Rs Cπ 2.4Ω(17 pF )
1 1 1
ω PH 2 = ω PH 2 = = = 7.1x108 rad / s
⎧⎪ ⎡ ⎛
⎨(RC RL )⎢1 + ⎜⎜ g m +
1
+
1 ⎞
( ⎤ ⎫⎪
⎟⎟ rπ (rx + RB RS ) ⎥ ⎬Cµ ) (RC RL )Cµ 1.05K (1.3 pF )
⎪⎩ ⎣ ⎝ RC RL ⎠ ⎦ ⎪⎭
1
= = 5.9 x106 rad / s Note: CB amplifier has much better
(
130 K 1.3 pF )
high frequency performance!
38
 Comparison of CB to CE Amplifier (with same Rs = 5 Ω)
CE (with RS = 5 Ω) CB (with RS = 5Ω)
⎡ r ⎤⎡ [(rx + rπ ) RB ] ⎤
= [− gm (RL RC )]⎢ π ⎥⎢
) [ [ ] ]
Vo Vo Vπ Vi ⎛ − rπ ⎞
AVo = =
Vs Vπ Vi Vs
⎥
⎣ rx + rπ ⎦⎣⎢ Rs + [(rx + rπ ) RB ]⎦⎥
V V V
AVo = o π e = − g m RL RC ( RE re
⎜
Rs + RE re ⎜⎝ rx + rπ
⎟
⎟
Midband Gain AVo = (− 218)(0.94)(0.93) = −191 V / V
Vπ Ve Vs ⎠
AVo = (− 218)(− 0.94 )(0.5) = +102.4 V / V
AVo (dB) = 20log(−191) = 45.6dB
AVo (dB ) = 40.2dB
1 1
1 1 ωZP1 = ωZP2 =0 ωZP3 = = = 42 rad / s
ωZP1 = ωZP2 =0 ωZP3 = =
RECE 0.33K (12µF )
= 252 rad / s RBCB 2K(12µF )
1 1 1 1
Low Frequency ωPL1 =
[RS + RB (rx + rπ )]CC1 0.7K(2µF ) = 714 rad / s
= ωPL1 =
{RB [rx + rπ + (1+ gmrπ )(RE RS )]}CB =
12µF(1K )
= 83 rad / s

Poles and Zeros ωPL2 =

1
=
1
= 33 rad / s ωPL2 =
1
=
1
= 5.0x104 rad / s
(RL + RC )CC 2 10.2K (3µF ) ⎧⎪ ⎡ r + r ⎤⎫⎪ 2µF(0.010K )
1 1 CC1⎨Rs + RE ⎢ x π ⎥⎬
ωPL3 = = = 1.7 x104 rad / s ⎪⎩ ⎣1+ gmrπ ⎦⎪⎭
⎡ ⎛ rx + rπ + Rs RB ⎞⎤ 0.005K (12µF )
⎢RE ⎜⎜ ⎟⎟⎥CE 1 1
⎢⎣ β +1 ⎠⎥⎦ ωPL3 = = = 33 rad / s
⎝ (RL + RC )CC2 10.2K(3µF)
g m 206 mA / V
ωZH 1 = ∞, ω ZH 2 = = = 1.6 x1011 rad / s ω ZH 1 = ∞, ω ZH 2 = ∞
Cµ 1.3 pF
High Frequency 1 1 ω PH 1 =
1
=
1
= 2.5 x1010 rad / s
Poles and Zeroes
ω PH 1 =
[ =
]
rπ (rx + RB RS ) Cπ 0.065K (17 pF )
= 9.0 x108 rad / s
[ ]
re RE Rs Cπ 2.4Ω(17 pF )
1 1 1
ω PH 2 = ω PH 2 = = = 7.1x108 rad / s
⎧⎪ ⎡ ⎛
⎨(RC RL )⎢1 + ⎜⎜ g m +
1
+
1 ⎞
( ⎤ ⎫⎪
⎟⎟ rπ (rx + RB RS ) ⎥ ⎬Cµ ) (RC RL )Cµ 1.05K (1.3 pF )
⎪⎩ ⎣ ⎝ RC RL ⎠ ⎦ ⎪⎭

=
1
= 5.0 x107 rad / s
Note: CB amplifier has much better
(
15.4 K 1.3 pF ) high frequency performance!
39
 Conclusions
* Voltage gain
z Can get good voltage gain from both CE and CB amplifiers.
z Low frequency performance similar for both amplifiers.
z CB amplifier gives better high frequency performance !
CE amplifier has dominant pole at 5.0x107 rad/s.
CB amplifier has dominant pole at 7.1x108 rad/s.
* Bandwidth approximately 14 X larger!
* Miller Effect multiplication of Cµ by the gain is avoided in
CB configuration.

* Current gain
z For CE amplifier, current gain is high AI = Ic / Ib
z For CB amplifier, current gain is low AI = Ic / Ie (close to one)!
z Frequency dependence of current gain similar to voltage gain.
* Input and output impedances are different for the two amplifiers!
z CB amplifier has especially low input resistance.

40
 Emitter-Follower (EF) Amplifier
* DC biasing
z Calculate IC, IB, VCE
z Determine related small signal
equivalent circuit parameters
Transconductance gm
Input resistance rπ
* Midband gain analysis
* Low frequency analysis
z Gray-Searle (Short Circuit)
Technique
Determine pole frequencies ωPL1,
ωPL2, ... ωPLn
z Determine zero frequencies
High and Low Frequency AC Equivalent Circuit ωZL1, ωZL2, ... ωZLn
* High frequency analysis
z Gray-Searle (Open Circuit)
Technique
Determine pole frequencies
ωPH1, ωPH2, ... ωPHn
z Determine zero frequencies
ωZH1, ωZH2, ... ωZHn

41
EF Amplifier - DC Analysis (Nearly the Same as CE Amplifier)
* GIVEN: Transistor parameters:
z Current gain β = 200
z Base resistance rx = 65 Ω
z Base-emitter voltage VBE,active = 0.7 V
z Resistors: R1=10K, R2=2.5K, RC=1.2K, RE=0.33K

* Form Thevenin equivalent for base; given VCC = 12.5V

z RTh = RB = R1||R2 = 10K||2.5K = 2K
z VTh = VBB = VCC R2 / [R1+R2] = 2.5V
z KVL base loop
IB = [VTh-VBE,active] / [RTh+(β +1)RE]
IB = 26 µA
* DC collector current IC = β IB
IC = 200(26 µ A) = 5.27 mA
* Transconductance gm = IC / VT ; VT = kBT/q = 26 mV
gm = 5.27 mA/26 mV = 206 mA/V
* Input resistance
rπ = β / gm = 200/[206 mA/V]= 0.97 K
* Check on transistor region of operation
R1 = 10K z KVL collector loop
R2 = 2.5K z VCE = VCC - (β +1) IB RE = 10.8 V (was 4.4 V for
RC = 0 K CE amplifier) (okay since not close to zero volts).
RE = 0.33K Note: Only difference here from CE case is VCE is larger
since RC was left out here in EF amplifier.
42
 EF Amplifier - Midband Gain Analysis
DC analysis is nearly the same!
Iπ RL = 9 K
IB , IC and gm are all the same. + +
Only VCE is different since RC=0. RC = 0 K
Ri
gm =
I C 5.27 mA
= = 206 mA / V RE = 0.33K
VT 26 mV Vb
R1 = 10 K
β 200 VO
rπ = =
g m 206 mA / V
= 0.97 K Vi R2 = 2.5 K
RS = 0.005K = 5Ω
_ _

Vo Vo Vπ Vi Vb
AVo = =
Vs Vπ Vi Vb Vs Equivalent input resistance Ri
⎡V ⎤
(R RE )⎢ π + g mVπ ⎥ V V + Vo
Ri = i = π
⎛ V
= rπ ⎜⎜1 + o
⎞
⎟ = 0.97 K (1 + 66 ) = 65 K
⎦ = (R R )1 + g m rπ = (9 K 0.33 K ) (201) = 66 ⎟
L
Vo
= ⎣ rπ Iπ ⎛ Vπ ⎞ ⎝ Vπ ⎠
⎜ ⎟
Vπ Vπ
L E
rπ 0.97 K ⎜r ⎟
⎝ π ⎠
Vπ Vπ 1 1
= = = = 0.015
Vi Vπ + Vo 1 + Vo 1 + 66
Vπ
AVo = (66 )(0.015)(0.999 )(0.998) = 0.987
Vi Ri 65 K
= =
Vb rx + Ri 0.065 K + 65 K
= 0.999 AVo (dB ) = 20 log(0.987 ) = −0.1dB
Vb RB (rx + Ri ) 2 K (0.065 K + 65 K ) 1 .9 K NOTE: Voltage gain is only ~1!
= = = = 0.998 This is a characteristic of the EF amplifier!
Vs RS + RB (rx + Ri ) 0.005 K + 2 K (0.065 K + 65 K ) 1.905 K
Cannot get voltage gain >1 for this amplifier!

43
 Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
* Draw low frequency AC circuit
z Substitute AC equivalent circuit for transistor
(hybrid-pi for bipolar transistor)
z Include coupling capacitors CC1, CC2
z Ignore (remove) all transistor capacitances Cπ , Cµ
* Turn off signal source, i.e. set Vs= 0
z Keep source resistance RS in circuit (do not remove)
* Consider the circuit one capacitor Cx at a time
z Replace all other capacitors with short circuits
z Solve remaining circuit for equivalent resistance Rx
seen by the selected capacitor
1
z Calculate pole frequency using ω Px =
Rx C x

z Repeat process for each capacitor finding equivalent

resistance seen and corresponding pole frequency
* Calculate the final low 3 dB frequency using
ω LP = ∑ω Px = ω P1 + ω P 2 ...ω Pn = ∑ R 1C
x x

44
 Emitter Follower - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
Input coupling capacitor CC1 = 2 µF

RxC1 = x = Rs + RB (rx + Ri )
V V
Ri = i
Ix Iπ
( ) [ (
Vi = Iπ rπ + (Iπ + gmVπ ) RE RL = Iπ rπ + (1 + gmrπ ) RE RL )]
(
Ri = i = rπ + (1 + gmrπ ) RE RL
V
Iπ
)

( )
= 0.97K + (201) 0.33K 9K = 65K

RxC1 = Rs + RB (rx + Ri ) IX
Iπ
= 0.005K + 2K (0.065K + 65K ) = 1.95K
Ri

CC1RxC1 = 2µF (1.95K ) = 3.9x10−3 sec

ω PL1 =
1
=
1
= 256 rad / s Vi
CC1RxC1 3.9x10−3 sec

45
Emitter Follower - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique

* Output coupling capacitor CC2 = 3 µF

V
RxC 2 = RL + RE re re = e
Ie
I e = − Iπ − g mVπ = − Iπ (1 + g m rπ )
[
Ve = − Iπ rπ + rx + RB RS ]
− Iπ [rπ + rx + RB RS ] rπ + rx + RB RS
re = =
− Iπ (1 + g m rπ ) 1 + g m rπ
0.97 K + 0.065K + 2 K 0.005K
= = 0.005K
201
RxC 2 = RL + RE re = 9 K + 0.33K 0.005K = 9.005K

1 1
ω PL 2 = = = 37 rad / s
Iπ RxC 2CC 2 9.005K (3µF )

Ve * Low 3 dB frequency
ω PL = ω PL1 + ω PL 2
re Ie
IX
= 256 + 37 = 293 rad / s

So dominant low frequency pole is due to CC1 !

46
 Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
* Draw high frequency AC equivalent circuit
z Substitute AC equivalent circuit for transistor
(hybrid-pi model for transistor with Cπ, Cµ)
z Consider coupling and emitter bypass
capacitors CC1 and CC2 as shorts
z Turn off signal source, i.e. set Vs = 0
z Keep source resistance RS in circuit
z Neglect transistor’s output resistance ro
* Consider the circuit one capacitor Cx at a time
z Replace all other transistor capacitors with
open circuits
z Solve remaining circuit for equivalent
resistance Rx seen by the selected capacitor
z Calculate pole frequency using
1
ω PHx =
Rx C x
z Repeat process for each capacitor

* Calculate the final high frequency pole using

−1 −1

∑
⎡ 1 ⎤ ⎡ 1 1 1 ⎤ 1
ω PH = ⎢ ⎥ =⎢ + ... + ⎥ =
⎣ ω Px ⎦ ⎣ ω PH 1 ω PH 2 ω PHn ⎦
∑R C x x

47
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
Vo
Z eq = = R E R L R S'
Ie
Vo
where R S' = = equivalent resistance due
− g mV π
to dependent current source
Vπ V
KCL at E + sCπVπ + gmVπ − o = 0
Ie rπ RE RL

⎧1 ⎫
= (RE RL )⎨ + sCπ + gm ⎬ = (RE RL ){yπ + gm }
Vo
* Redrawn High Frequency Equivalent Vπ ⎩ rπ ⎭
Circuit 1 1
where we define yπ = = +
1 1
= + sCπ
zπ rπ ZCπ rπ
zπ =1/yπ 1
= − (RE RL ){yπ + gm }
Vo
so RS' = −
E gmVπ gm
Ie
Since Zeq = RE RL RS' we can find
Zeq
1 1 1 1 ⎛ 1 ⎞⎡ g m ⎤
= + ' = − ⎜⎜ ⎟⎢
⎟ ⎥
Zeq RE RL RS RE RL ⎝ E L ⎠⎣ yπ + gm ⎦
R R
⎛ 1 ⎞⎡ yπ ⎤
= ⎜⎜ ⎟⎢
⎟ ⎥
⎝ E L ⎠⎣ yπ + gm ⎦
R R
⎡ y + gm ⎤
Zeq = (RE RL )⎢ π ⎥
⎣ yπ ⎦ 48
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
ZB’
Total impedance between B' and ground is Z b'

zπ =1/yπ Z b' = Z eq +
1
= (RE RL )⎢ π
⎡ y + gm ⎤ 1
= (RE RL ) +
(1 + g m (RE RL ))
⎥+
yπ ⎣ yπ ⎦ yπ yπ
1
= (RE RL ) +
⎛ yπ ⎞
⎜ ⎟
⎜ 1 + g (R R ) ⎟
Replace this ⎝ m E L ⎠
with this. 1 1 + srπ Cπ
Substituting for yπ = + sCπ =
rπ rπ
1
Z b' = (RE RL ) +
⎛ ⎡1 + srπ Cπ ⎤ ⎞
Modified Equivalent Circuit ⎜ ⎢ ⎥ ⎟
⎜ ⎣ rπ ⎦ ⎟
⎜ 1 + g (R R ) ⎟
⎜ m E L ⎟
⎜ ⎟
ZB’ ⎝ ⎠
1
= (RE RL ) +
1 Cπ
+s
rπ [1 + g m (RE RL )] [1 + g m (RE RL )]
so Z b' is
Z b' = (RE RL ) + (rπ [1 + g m (RE RL )]) Z C ' Looks like a resistor in
parallel with a capacitor.
1 1
where Z C ' = =
sC ' s Cπ
1 + g m (RE RL )
49
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

RxCπ

* Pole frequency for Cπ =17 pF

[
RxCπ ' = [rπ (1 + g m RE RL )] rx + RS RB + RE RL ]
[ ][
= 0.97 K (1 + (206 mA / V )(0.33K 9 K )) 0.065K + 0.005K 2 K + 0.33K 9 K ]
= [64.6 K ] [0.39 K ] = 0.386 K

1 1 1 1 1
ω PH 1 = = = = =
RxCπ 'Cπ ' ⎡ Cπ ⎤ ⎡ 17 pF ⎤ 0.386 K (0.255 pF ) 9.86 x10 −11 s
RxCπ ' ⎢ ⎥ 0.386 K ⎢ ⎥
⎣1 + g m RE RL ⎦ ⎣1 + (206mA / V )(0.33K 9 K )⎦
ω PH 1 = 1.0 x1010 rad / s
50
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique

* Pole frequency for Cµ =1.3 pF

RxCµ = ([rπ (1 + g m RE RL )] + RE RL )[r + R R ]

x S B

= ([0.97 K (1 + (206 mA / V )(0.33K 9 K ))]+ (0.33K 9 K )) [0.065 K + 0.005 K 2 K ]

= (64.6 K + 0.32 K ) 0.07 K = 0.07 K

1 1 1
ω PH 2 = = =
RxCµ C µ 0.07 K (1.3 pF ) 9.1x10 −11 s
ω PH 2 = 1.1x1010 rad / s
51
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
* Alternative Analysis for Pole Due to Cπ
Ix-Iπ Ix Note : Vx = Vπ and RxCπ =
Vx Vπ rπ Iπ
= =
Ix Ix Ix
Vx KVL around base loop gives
Iπ − (I e + g mVπ )(RE RL ) − Vπ + (I x − Iπ )(rx + RS RB ) = 0
E But I e = Iπ − I x and Vπ = Iπ rπ so
Ie − (Iπ − I x + g m rπ Iπ )(RE RL ) − Vπ + (I x − Iπ )(rx + RS RB ) = 0
Ie+gmVπ Rearranging we get
Iπ [(1 + g m rπ )(RE RL ) + (rx + RS RB )] = I x (RE RL + rx + RS RB )
Iπ RE RL + rx + RS RB
=
I x (1 + g m rπ )(RE RL ) + (rx + RS RB )
so

Vx rπ Iπ RE RL + rx + RS RB
RxCπ = = = rπ
Ix Ix (1 + g m rπ )(RE RL ) + (rx + RS RB )
0.33K 9 K + 0.065K + 0.005K 2 K
We get the same result here for the high = 0.97 K = 0.006K
frequency pole associated with Cπ as we (201)(0.33K 9K ) + 0.065K + 0.005K 2K
did using the equivalent circuit So the pole frequency is
transformation. ωPH 1 =
1
=
1
= 1.0 x1010 rad / s
RxCπ Cπ 0.006K (17 pF )

52
 Emitter Follower - Analysis of High Frequency Poles
Gray-Searle (Open Circuit) Technique
Ix-Iπ * Alternative Analysis for Pole Due to Cµ
⎡ R R ⎤
Vx = Vπ + (Iπ + gmVπ )(RE RL ) = Vπ ⎢1 + (1 + gmrπ ) E L ⎥ so
⎣ rπ ⎦
Ix ⎡ rπ ⎤
Vx Vπ = Vx ⎢ ⎥
⎢⎣ rπ + (1 + gmrπ )(RE RL )⎥⎦
We can also write
Iπ E
Vx = (I x − Iπ )[rx + Rs RB ] = I x [rx + Rs RB ] − [rx + Rs RB ]
Vπ
rπ
Substituting for Vπ we get
Iπ+gmVπ
⎡ ⎤
Vx = I x [rx + Rs RB ] −
[rx + Rs RB ]Vx ⎢
1 rπ
⎥
rπ ⎢⎣ rπ + (1 + gmrπ )(RE RL )⎥⎦
Rearranging we get
Vx rx + Rs RB 1
RxCµ = = =
Ix rx + Rs RB 1 1
1+ +
rπ + (1 + gmrπ )(RE RL ) rx + Rs RB rπ + (1 + gmrπ )(RE RL )
= [rx + Rs RB ] [rπ + (1 + gmrπ )(RE RL )]
We get the same result here for the high
frequency pole associated with Cµ as = [0.065K + 0.005K 2K ] [0.97K + (201)(0.33K 9K )]
we did using the equivalent circuit = [0.07K ][65.0K ] = 0.07K
transformation. So the pole frequency for Cµ is
1 1
ωPH 2 = = = 1.1x1010 rad / s
RxCµ Cµ 0.07K (1.3 pF )

53
 Comparison of EF to CE Amplifier (For RS = 5Ω )
CE EF
⎡ r ⎤⎡ [(rx + rπ ) RB ] ⎤
= [− gm (RL RC )]⎢ π ⎥⎢ ⎥ AVo = Vo Vπ Vi Vb = (RL RE ) ⎛⎜ Ri ⎞⎟ RB (rx + Ri )
Vo Vo Vπ Vi
AVo = =
Vs Vπ Vi Vs r +
⎣ x π ⎦⎣⎢ s
r R + [(r + rπ ) R ]⎥
B ⎦ Vπ Vi Vb Vs [rπ + [RE RL ]] ⎜⎝ rx + Ri ⎟⎠ RS + RB (rx + Ri )
Midband Gain x

AVo = (− 218)(0.94)(0.93) = −191 V / V AVo = (66 )(0.015)(0.999 )(0.998) = 0.987 V / V

AVo (dB) = 20log(−191) = 45.6dB AVo (dB ) = −0.1dB

1 1
ωZP1 = ωZP2 =0 ωZP3 = = = 252 rad / s
RECE 0.33K (12µF ) ωZP1 = ωZP2 = 0
1 1 1 1
ωPL1 =
Low Frequency [RS + RB (rx + rπ )]CC1 0.7K(2µF ) = 714 rad / s
= ωPL1 =
[ =
]
CC1 Rs + RB (rx + Ri ) 2µF(1.95K )
= 256 rad / s

1 1
Poles and Zeros ωPL2 = = = 33 rad / s ωPL2 =
1
=
1
= 37 rad / s
(RL + RC )CC 2 10.2K (3µF ) [RL + (RE re )]CC2 3µF(9K )
1 1
ωPL3 = = = 1.7x104 rad / s
⎡ ⎛ rx + rπ + Rs RB ⎞⎤ 0.005K (12µF ) Better low frequency response !
⎢RE ⎜ ⎟⎥CE
⎜ β + 1 ⎟
⎢⎣ ⎝ ⎠⎥⎦

g m 206 mA / V ⎞ 1
⎛ 1 + g m rπ 201
ω ZH 1 =∞ ωZH 2 = = = 1.6 x1011 rad / s ωZH1 =∞ ωZH 2 = ⎜⎜ ⎟⎟ = = 1.2 x1010 rad / s
Cµ 1.3 pF ⎝ rπ ⎠ π
C 0.97 K (17 pF )
High Frequency 1 1 1 1
ω PH 1 = = = 9.0 x108 rad / s ωPH1 = = = 1.0 x1010 rad / s
Poles and Zeroes [ ]
rπ (rx + RB RS ) Cπ 0.065K (17 pF ) ⎡
RxCπ ' ⎢
Cπ ⎤
⎥
0.386K (0 .26 pF )
1 ⎣1 + g m RE RL ⎦
ω PH 2 =
⎧⎪ ⎡ ⎛ ⎤ ⎫⎪ 1
⎨(RC RL )⎢1 + ⎜⎜ g m +
1
+
1 ⎞
( )
⎟⎟ rπ (rx + RB RS ) ⎥ ⎬Cµ ωPH 2 =
[rπ (1 + gm RE RL ) + RE RL ] [rx + RS RB ]Cµ
⎪⎩ ⎣ ⎝ RC RL ⎠ ⎦ ⎪⎭
1 1
= = 5.0 x107 rad / s = = 1.1x1010 rad / s
15.4 K (1.3 pF ) (
0.07K 1.3 pF ) Much better high frequency response !
54
 Conclusions
* Voltage gain
z Can get good voltage gain from CE but NOT from EF amplifier (AV ¡ 1).
z Low frequency performance better for EF amplifier.
z EF amplifier gives much better high frequency performance!
CE amplifier has dominant pole at 5.0x107 rad/s.
EF amplifier has dominant pole at 1.0x1010 rad/s.
* Bandwidth approximately 200 X larger!
* Miller Effect multiplication of Cµ by the gain is avoided in EF.

* Current gain
z For CE amplifier, current gain is high β = Ic/Ib
z For EF amplifier, current gain is also high Ie/Ib = β +1 !
z Frequency dependence of current gain similar to voltage gain.
* Input and output impedances are different for the two amplifiers!

55
 Cascade Amplifier
EF CE

β1 = β 2 = 100
rx1 = rx 2 ≈ 0
Cπ 1 = Cπ 2 = 13.9 pF
C µ1 = C µ 2 = 2 pF

* Emitter Follower + Common Emitter (EF+CE)

* Voltage gain from CE stage, gain of one for EF.
* Low output resistance from EF provides a low source resistance for CE amplifier so
good matching of output of EF to input of CE amplifier
* High frequency response (3dB frequency) for Cascade Amplifier is improved over CE
amplifier.

56
 Cascade Amplifier - DC analysis
R2 100 K
VTh1 = VCC = 10V = 5V
R1 + R2 200 K
IB1 RTh1 = R1 R2 = 100 K 100 K = 50 K
IB2 KVL Base Q1
IE1 β1 = β 2 = 100 VTh1 − V BE 1 = I B1 RTh1 + {[ β 1 + 1] I B1 − I B 2 }R E 1
Neglecting I B 2 as a first approximat ion

IRE1 5V − 0.7V
I B1 ≈ = 8 .9 µA
50 K + (100 + 1)4.3 K
Then
I E 1 = (β 1 + 1)I B1 = (101)8 .9 µA = 899 µA
Small Signal Parameters
I C1 β1 I B1 100(8.9 µA) mA
g m1 = = = = 34.8 Now calculate V B 2 and I B 2
VT VT 0.0256V V
β1 100 V B 2 ≈ I E 1 R E 1 = 899 µA( 4.3 K ) = 3 .87V
rπ 1 = = = 2.9 K
g m1 34.8mA / V V B 2 = V BE 2 + (β 2 + 1)I B 2 R E 2
3.87V − 0.7V
I C 2 β 2 I B 2 100(8.7 µA)
I B2 = = 8 .7 µ A
gm2 = = = = 34.0
mA (101)3.6 K
VT VT 0.0256V V
I B2 << I E 1 so approximat e analysis is okay .
β2 100
rπ 2 = = = 2.9 K
gm2 34.0mA / V
57
 Cascade Amplifier - Midband Gain Analysis
Note: rx1 = rx2 = 0 so equivalent circuit is simplified.
Iπ1
+
+ Vπ1
_
Vi +
Vπ2
_ _
Ri
Vo Vo Vπ 2 Vπ 1 Vi
AVo = = I π 1rπ 1 + (1 + g m1rπ 1 )I π 1 [(RE1 rπ 2 )]
Vs Vπ 2 Vπ 1 Vi Vs Ri =
Vo [− g m 2Vπ 2 ](RL RC )
Iπ 1
= = −68
Vπ 2 Vπ 2 = rπ 1 + (1 + g m1rπ 1 )(RE1 rπ 2 )
Vπ 2 (I π 1 + g m1Vπ 1 )RE1 rπ 2 ⎛ 1 ⎞ mA
= = ⎜⎜ + g m1 ⎟⎟ RE1 rπ 2 = 35.1 (4.3K 2.9 K ) = 60.8 = 2.9 K + (101)(4.3K 2.9 K ) = 178 K
Vπ 1 Vπ 1 ⎝ rπ 1 ⎠ V
Vπ 1 I π 1rπ 1 2.9 K
= = = 0.016
Vi I π 1rπ 1 + (1 + g m1rπ 1 )I π 1 ( RE1 rπ 2 ) 2.9 K + (101)(4.3K 2.9 K )
Vi ( R1 R2 ) Ri (100 K 100 K ) 178 K AVo = (− 68)(60.8)(0.016 )(0.91) = −60.2 V / V
= = = 0.91
Vs RS + RB ( R1 R2 ) 4 K + (100 K 100 K ) 178 K AVo (dB ) = 20 log( − 60.2 ) = 35.6dB

Note: Voltage gain is nearly equal to

that of the CE stage, e.g. – 68 !
58
Cascade Amplifier - Low Frequency Poles and Zeroes

(s + ωZ 1L )(s + ωZ 2 L )(s + ωZ 3L ) * Use Gray-Searle (Short Circuit)

FL ( s ) = Technique to find the poles.
(s + ω P1L )(s + ω P 2 L )(s + ω P3 L )
z Three low frequency poles
⎛ ω Z 1L ⎞⎛ ω Z 2 L ⎞⎛ ω Z 3 L ⎞
⎜1 + ⎟⎜1 + ⎟⎜1 + ⎟ z Equivalent resistance may
= ⎝ s ⎠⎝ s ⎠⎝ s ⎠
depend on rπ for both
⎛ ω P1L ⎞⎛ ω P 2 L ⎞⎛ ω P 3 L ⎞
⎜1 + ⎟⎜1 + ⎟⎜1 + ⎟ transistors.
⎝ s ⎠⎝ s ⎠⎝ s ⎠
* Find three low frequency zeroes.

59
 Cascade Amplifier - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
Input coupling capacitor CC1 = 1 µF

Vx Vi
RxC1 = = Rs + RB Ri Ri = rπ1 Vπ1
Ix Iπ
Vi = Iπ1rπ1 + (Iπ1 + gm1Vπ1 )(RE1 rπ 2 ) = Iπ1[rπ1 + (1+ gm1rπ1 )(RE1 rπ 2 )]

= rπ1 + (1+ gm1rπ1 )(RE1 rπ 2 )

Vi
Ri = +
Iπ rπ2
RE1 Vπ2
= 2.9K + (101)(4.3K 2.9K ) = 178K
_
RxC1 = Rs + RB Ri IX
= 4K + 50K 178K = 43.0K Iπ1
Ri
CC1RxC1 = 1µF(43.0K ) = 4.3x10−2 sec
1 1
ωPL1 = = = 23 rad / s Vi
CC1RxC1 4.3x10−2 sec
RE1 rπ2

60
Cascade Amplifier - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
rX2 CC2
Vo

r π2 Vπ2 gm2Vπ2 RC RL

RE2 CE

* Output coupling capacitor CC2 = 1 µF

VX
Vo
RC RL

RC 2 = RL + RC = 4 K + 4 K = 8 K
1 1
ω PL 2 = = = 125 rad / s
RC 2CC 2 8K (1µF )
61
 Cascade Amplifier - Analysis of Low Frequency Poles
Gray-Searle (Short Circuit) Technique
Emitter bypass capacitor CE = 47 µF
Iπ1
V
R Ex = x = R E 2 re 2
Ix
VE 2 − I π 2 ( rπ 2 + R E 1 re 1 ) rπ 2 + R E 1 re 1 r π1 Vπ1 gm1Vπ1
re 2 = = = VE1
I e2 − I π 2 (1 + g m 2 rπ 2 ) 1 + g m 2 rπ 2
V E1 − I π 1 ( rπ 1 + R S ' ) r + RS ' Ie1
re 1 = = = π1 RS ' = RS R1 R2 Iπ2
I e1 − I π 1 (1 + g m 1 rπ 1 ) 1 + g m 1 rπ 1
= 3.7 K re1 rπ2 Vπ2 gm2Vπ2
2 .9 K + 3 .7 K RE1 VE2
= = 0 . 065 K
101
Ie2 Ix
2 . 9 K + 4 . 3 K 0 . 065 K re2 RE2
re 2 = = 0 . 029 K VX
101
IE2
R Ex = R E 2 re 2 = 3 . 6 K 0 . 029 K = 0 . 029 K

1 1
ω PL 3 = = = 734 rad / s Low 3 dB Frequency
R Ex C E 0 . 029 K (47 µ F )

ω PL = ω PL1 + ω PL 2 + ω PL 3 = 23 + 125 + 734 = 882 rad / s

The pole for CE is the largest and therefore the

most important in determining the low 3 dB frequency.
62
 Comparison of Cascade to CE Amplifier
CE* Cascade (EF+CE)
Vo Vo Vπ 2 Vo Vπ 2 Vπ 1 Vi 2 X improvement
AVo = = AVo =
Vs Vπ 2 Vs Vπ 2 Vπ 1 Vi VS in voltage gain !
Midband Gain
AVo = (− 81.2)(0.37) = −30 V / V AVo = (− 68)(60.8)(0.016 )(0.91) = −60.2 V / V
AVo (dB) = 20log(− 30) = 29.5dB AVo (dB ) = 20 log( − 60.2 ) = 35.6dB
1 1 1 1
ωZP1 = ωZP2 = 0 ωZP3 = = = 5.9 rad / s ωZP1 = ωZP2 = 0 ωZP3 = = = 5.9 rad / s
RE 2CE 3.6K (47µF ) RE2CE 3.6K(47µF )
1 1 1 1
ωPL1 = = = 157 rad / s
Low Frequency [RS + RB rπ 2 ]CC1 4.4K (1µF ) ωPL1 =
{ } =
RS + RB [rπ1 + (1+ gm1rπ1 )(RE2 rπ 2 )] CC1 1µF(43K )
= 23 rad / s

Poles and Zeros ωPL2 =

1
=
1
= 125 rad / s 1 1
(RL + RC )CC 2 8K (1µF ) ωPL2 = = = 125 rad / s
(RL + RC )CC2 8K(1µF)
1 1 1 1
ωPL3 = = = 591 rad / s ωPL3 =
⎡ ⎛ rπ 2 + Rs RB ⎞⎤ 0.036K (47µF ) =
(RE2 re2 )CE 47µF(0.03K )
= 734 rad / s
⎢RE 2 ⎜ ⎟
⎜ β +1 ⎟⎥CE
⎢⎣ ⎝ ⎠⎥⎦

g m 2 40.6 mA / V ωZH 1 = ∞, ωZH 2 = ∞,

ωZH 1 = ∞, ω ZH 2 = = = 2.0 x1010 rad / s
Cµ 2 2 pF
ωZH 3 = 2.5 x109 rad / s, ω ZH 4 = 1.7 x1010 rad / s
High Frequency ω PH 1 =
1
=
1
= 4.8 x10 7 rad / s 1
Poles and Zeroes [r π2 ]
RB RS Cπ 2 1.5K (13.9 pF ) ω PH 1 =
0.09 K (13.9 pF )
= 8.0 x108 rad / s,
1 25 X improvement
ω PH 2 = 1
⎧⎪ ⎤ ⎫⎪ ω PH 2 = = 1.4 x10 rad / s, in bandwidth !
8
⎡ ⎛ 1 ⎞
( ) 3.6 K (2 pF )
⎨(RC RL )⎢1 + ⎜⎜ g m 2 + ⎟⎟ rπ 2 (RB RS ) ⎥ ⎬Cµ 2
1
+
⎪⎩ ⎣ ⎝ RC RL ⎠ ⎦ ⎪⎭ 1
ω PH 3 = = 1.0 x108 rad / s,
1 0.063K (152 pF )
= = 4.0 x106 rad / s
(
125K 2 pF ) 1
ω PH 4 = = 2.5 x108 rad / s
2 K (2 pF ) 63
* CE stage with same transistor, biasing resistors, source resistance and load as cascade.
 Comparison of Cascade to CE Amplifier
* Why the better voltage gain for the cascade?
z Emitter follower gives no voltage gain!
z Cascade has better matching with source than CE.
Cascade amplifier has an input resistance that is
higher due to EF first stage.
Ri1 Ri1 = rπ 1 + (β + 1)RE1 rπ 2
Ri2 = 2.9 K + (101)(4.3K 2.9 K ) = 178K
Versus Ri2 = rπ2 = 2.5 K for CE
So less loss in voltage divider term (Vi / Vs ) with
the source resistance.
* 0.91 for cascade vs 0.37 for CE.
* Why better bandwidth?
Pole for Capacitor CT = 152 pF z Low output resistance re1 of EF stage gives smaller
effective source resistance for CE stage and higher
frequency for dominant pole due to CT (including
Cµ2) V − I (r + R ') r + R '
re1 = e1
= π1 π1 S
= π1 S
I e1 − I π 1 (1 + g m1rπ 1 ) 1 + g m1rπ 1
re1
2.9 K + 3.7 K
= = 0.065 K
101
RX = re1 RE1 rπ 2 =0.065 K 4.3K 2.9 K = 0.063K
CT = Cπ 2 + C µ 2 (1 + g m 2 RL ') = 13.9 pF + 2 pF {1 + (34mA / V )2 K } = 152 pF 1
ω PH 3 = = 1.0 x108 rad / s for the cascade
0.063K (152 pF )
RL ' = RL RC = 4 K 4 K = 2 K
versus 4.0 x106 rad / s for the CE amplifier.
64
 Another Useful Amplifier – Cascode (CE+CB) Amplifier

* Common Emitter + Common Base

(CE + CB) configuration
* Voltage gain from both stages
* Low input resistance from second CB stage
provides first stage CE with low load
resistance so Miller Effect multiplication of
Cµ1 is much smaller.
* High frequency response dramatically
improved (3 dB frequency increased).
z Bandwidth is much improved (~130 X).

For CE amplifier the high frequency performanc e is limited by

C in 1 = C π 1 + C µ 1 [1 + g m 1 (R L R C )] = 17 pF + 1 . 3 pF (1 + 206 mA / V (9 K 1 . 2 K )) = 17 pF + 1 . 3 pF (1 + 218 ) = 302 pF
1
ω PHin = −8
= 5 . 6 x10 6 rad / s Large Miller Effect
1 . 8 x10 sec
For cascode amplifier , R L for first stage CE amplifier becomes R i of CB amplifier where R i is given by
rx + rπ Small Miller Effect
Ri = = 5Ω
1 + g m rπ
so C in 1 = C π 1 + C µ 1 [1 + g m (R L R C )] = 17 pF + 1 . 3 pF (1 + 206 mA / V (0 . 005 K 1 . 2 K )) = 17 pF + 1 . 3 pF (1 + 1) = 19 . 6 pF
1 Bandwidth is improved by a factor of 130X
ω PHin = = 7 . 3 x10 8 rad / s
(1 . 2 K 9 K )(2 . 6 pF ) over that for the CE amplifier !
65
Other Examples of Multistage Amplifiers
CE CE EF EF

Darlington Pair

66
 Other Examples of Multistage Amplifiers
Push – Pull Amplifier Amplifier with Npn and Pnp Transistors

Amplifier with FETs and

Bipolar Transistors

67
 Differential Amplifier
* Similar to CE amplifier, but two CE’s
operated in parallel
* Signal applied between two equivalent inputs
_ instead of between one input and ground
+V * Common emitter resistor or current source
o
used
* Current shared or switched between two
transistors (they compete)
* Analyze using equivalent half-circuit
z 1/2 of signal at input
z 1/2 of signal at output
z 1/2 of source resistance
* Gain and frequency response similar to CE
amplifier for high frequencies
* Advantage:
z Rejects common noise pickup on input
z No coupling capacitors so can operate
down to zero frequency.
68
 Differential Amplifier Analysis
Midband Gain
Vo ⎛ Vo ⎞⎞⎛ ⎛ ⎞
2 Vo ⎜ 2 ⎟ ⎛ − g mVπ RC ⎞⎜
⎟⎜ Vπ rπ ⎟
AVo =
Vs
= =⎜ ⎟ = ⎜⎜
⎟⎜ V ⎟⎜
⎟ ⎟
Vo
Vs ⎜ Vπ ⎟ ⎝
⎟⎜ s Vπ ⎠⎜ rπ + rx + RS ⎟
2 ⎝ ⎠⎠⎝ 2 ⎝ 2 ⎠
Vo /2 ⎛ 0.97 K ⎞
= (− 206mA / V )(9 K )⎜ ⎟ = −509
⎝ 0.97 K + 0.065K + 2.5 K ⎠
AVo (dB) = 20 log 509 = 54.1dB

Low Frequency Poles and Zeros

* Direct coupled so no coupling capacitors
and no emitter bypass capacitor
* No low frequency poles and zeros
* Flat (frequency independent) gain
down to zero frequency

Vo /2
High Frequency Poles and Zeros
Dominant pole using Miller’s Thoerem
1
ω PH =
[rπ (rx + Rs / 2)][Cπ + (1 + g m RC )Cµ ]
1 1
= =
[ ]
0.97 K (0.065 K + 5K / 2) [17 pF + 201(1.3 pF )] (0.70 K )(278 pF )
1
= = 5.1x106 rad / s
−7
1.95 x10 sec
High frequency performance is very similar to CE amplifier.
69
 Summary
* In this chapter we have shown how to analyze the high and low frequency
dependence of the gain for an amplifier.
z Analyzed the effects of the coupling capacitors on the low frequency response
Found the expressions for the corresponding poles and zeros.
Demonstrated Bode plots of magnitude and phase.
z Analyzed the effects of the capacitances within the transistor on the high
frequency response.
Found the expressions for the corresponding poles and zeros.
Demonstrated Bode plots of the magnitude and phase.

* Analyzed the high and low frequency performance of the three bipolar transistor
amplifiers: common emitter, common base and emitter follower.
z Found the expressions for the corresponding poles and zeros.
z Demonstrated Bode plots of the magnitude and phase.

* Demonstrated how to find the expressions for the gain and the high and low
frequency poles and zeros for multistage amplifiers.

70