EECS 105 Fall 2003, Lecture 20

Lecture 20:
Frequency Response: Miller Effect

Prof. Niknejad

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Lecture Outline

Department of EECS

Frequency response of the CE as

voltage amp
The Miller approximation
Frequency Response of Voltage
Buffer

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Last Time: CE Amp with Current Input

Can be MOS

Calculate the short circuit current gain of device

(BJT or MOS)
Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

CS Short-Circuit Current Gain

MOS Case

MOS
BJT

0 dB

T
Ai ( j ) =

g m (1 jCgd / g m )
j (C gs + Cgd )

gm

j (C gs + Cgd )

Note: Zero occurs when all of gm current flows into Cgd:

g m vgs = vgs jCgd

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Common-Emitter Voltage Amplifier

Small-signal model:
omit Ccs to avoid
complicated analysis
Can solve problem directly
by phasor analysis or using
2-port models of transistor
OK if circuit is small
(1-2 nodes)

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

CE Voltage Amp Small-Signal Model

Two Nodes! Easy

Same circuit works for CS with

Department of EECS

r
University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Frequency Response
KCL at input and output nodes; analysis is made
complicated due to Z branch
see H&S pp. 639-640.
Vout
=
Vin

r
g m
r + RS

[ro || roc || RL ](1 j / z )

(1 + j / )(1 + j / )
p1

p2

Low-frequency gain:

Vout
=
Vin

r
r + RS

gm

Two Poles

[ ro || roc || RL ] (1 j 0 )

(1 + j 0 )(1 + j 0 )

gm
Zero: z > T =
C +C
Department of EECS

Zero

gm
r

r
+ RS

[ ro || roc || RL ]

Note: Zero occurs due to feed-forward

current cancellation as before

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Exact Poles
1
p1
)C }+ Rout
C
(RS || r ){C + (1 + g m Rout
Usually >> 1

p2

/ (RS || r )
Rout

)C }+ Rout
C
(RS || r ){C + (1 + g m Rout

These poles are calculated after doing some

algebraic manipulations on the circuit. Its hard
to get any intuition from the above expressions.

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Miller Approximation

Results of complete analysis: not exact and little insight

Look at how Z affects the transfer function: find Zin

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Input Impedance Zin(j

)
I t = (Vt Vout ) / Z
At output node:

g mVt Rout

Vout = ( g mVt I t ) Rout

Why?

I t = (Vt AvC Vt ) / Z

Z in = Vt / I t =
Department of EECS

Z
1 AvC
University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Miller Capacitance CM
Effective input capacitance:
1
1
=
Z in =
jC M 1 AvC

jC

1
=
j (1 A )C
vC

Cx
+
AV,Cx

+
Vin

Department of EECS

AV,Cx

+
Vout

(1-Av,Cx)Cx

+
Vout

(1-1/Av,Cx)Cx

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Some Examples
Common emitter/source amplifier:

AvC =

Negative, large number (-100)

CM = (1 AV ,C )C
100C
Miller Multiplied Cap has Detrimental Impact on bandwidth

Common collector/drain amplifier:

AvC =

Slightly less than 1

CM = (1 AV ,C )C
0
Bootstrapped cap has negligible impact on bandwidth!
Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

CE Amplifier using Miller Approx.

Use Miller to transform C
RS || r

Analysis is straightforward now single pole!

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Comparison with Exact Analysis

Miller result (calculate RC time constant of input pole):

1
p1

) C }
= ( RS || r ) {C + (1 + g m Rout

Exact result:

)C }+ Rout
C
p1 = (RS || r ){C + (1 + g m Rout
1

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Method of Open Circuit Time Constants

This is a technique to find the dominant pole of a

circuit (only valid if there really is a dominant
pole!)
For each capacitor in the circuit you calculate an
equivalent resistor seen by capacitor and form the
time constant i=RiCi
The dominant pole then is the sum of these time
constants in the circuit

p ,dom =
Department of EECS

1 + 2 +

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Equivalent Resistance Seen by Capacitor

For each small capacitor in the circuit:

Open-circuit all other small capacitors

Short circuit all big capacitors
Turn off all independent sources
Replace cap under question with current or voltage
source
Find equivalent input impedance seen by cap
Form RC time constant

This procedure is best illustrated with an

example

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Example Calculation

Consider the input capacitance C1 = C + CM

Open all other small caps (get rid of output cap)
Turn off all independent sources
Insert a current source in place of cap and find
impedance seen by source
RM = r || RS

) C }
1 = ( RS || r ) {C + (1 + g m Rout

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Common-Collector Amplifier
Procedure:
1. Small-signal twoport model
2. Add device (and
other) capacitors

Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Two-Port CC Model with Capacitors

Gain ~ 1

Find Miller capacitor for C -- note that the base-emitter

capacitor is between the input and output
Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Voltage Gain AvC Across C

AvC Rout /( Rout + RL ) 1

g m RL >> 1

Rout

1
=
gm

Note: this voltage gain is neither the two-port gain nor the
loaded voltage gain

Cin = C + C M = C + (1 AvC )C
1
Cin = C +
C
1 + g m RL
Cin C
Department of EECS

University of California, Berkeley

EECS 105 Fall 2003, Lecture 20

Prof. A. Niknejad

Bandwidth of CC Amplifier
Input low-pass filters 3 dB frequency:
p

= (RS || Rin ) C +
1 + g m RL

Substitute favorable values of RS, RL:

RS 1 / g m
p

RL >> 1 / g m

(1 / g m ) C +
C / g m
1 + BIG

Model not valid at

these high frequencies

p g m / C > T
Department of EECS

University of California, Berkeley