You are on page 1of 7

192 IEEE TRANSACTIONS ON EDUCATION, VOL. E-17, NO.

4, NOVEMBER 1974

these particular positions. Stress design work and 2) The individual teacher can do much to rectify this
design oriented consulting output in lieu of conven- imbalance simply by modifying the conduct of
tional publications when making pay, promotion existing courses. Questions of synthesis and design
and tenure decisions concerning these individuals. can be introduced in nearly every course at every
level. What is mainly needed is a shift in view-
CONCLUSIONS point, and the allocation of sufficient time and
effort toward fostering creativity in design.
1) The beneficial trend of the past two decades toward 3) School administrators can play a decisive role in
up-grading analytical aspects of engineering ed- restoring a healthier balance in engineering ed-
ucation has unfortunately been accompanied by a ucation by adopting policies of faculty recruitment,
concomitant deterioration in the quality of in- pay, promotion, and tenure that foster and reward
struction offered in engineering synthesis and creative design skills to a degree comparable to
design. that now lavished on research and publications.

A Simplified Metho of Feedbacd Amplifier Analysis


SOLOMON ROSENSTARK, MEMBER, IEEE

Abstract-An exact asymptotic method is presented for performing difficulties by placing a phantom voltage (or current)
gain calculations on feedback amplifiers. The method is algorithmic source at the break, but the method is approximate in-
and utilizes only Ohm's law, voltage and current division and source
conversion and does not require the breaking of the feedback loop. asmuch as forward transmission through the feedback net-
For impedance calculations Blackman's formula is used. A set of work is ignored.
quick-reference tables is presented for the most common feedback Another approach is to represent the amplifier and feed-
amplifier configurations. back network in terms of their respective two port ma-
trices [3]. The student is required to choose a Y, Z, H,
or G matrix representation depending on the categoriza-
1. INTRODUCTION tion of the amplifier on the basis of the input-output feed-
A COMMON approach used in the teaching of feed- back connection (shunt-shunt, series-series etc.). This
,Ax back amplifiers to undergraduates consists of pre- method is complicated and so approximations to this
senting the fundamental principles on a block diagram method are often used.
basis. This is very suitable for demonstrating the general Finally, there is the problem of finding input and output
effects of feedback, such as improvement in gain stability, impedances. The usual method is to multiply or divide the
distortion, and changes in bandwidth. But the block dia- open loop impedance by the return difference depending
gram method is of limited usefulness in practical feedback on the amplifier categorization. This makes no provision
amplifier circuits, since the feedback network causes sig- for finding impedance for cases not falling into the four
nificant loading on the basic amplifier and so it is impos- basic classifications, for example amplifiers with un-
sible to separate the feedback amplifier into two distinct balanced bridge feedback.
blocks. A number of different methods have been used to This paper presents the asymptotic formula for gain
circumvent the problem. calculations in sections 2 and 3. A simple derivation is
The traditional method [1] requires the breaking of the given in appendix A. The asymptotic gain method has
feedback loop at some point and carefully terminating the following advantages:
with the proper impedance at the break. This often gives a. It is exact.
rise to conceptual problems which are difficult to resolve. b. It is algorithmic. No ingenuity is required to apply
For example, how can this procedure be applied to a very the method.
simple feedback amplifier such as the emitter follower? c. It is simple. Ohm's law, voltage and current division,
A more recent method [2] overcomes some of the above and source conversion suffice to find all feedback quan-
tities.
d. It is general. The subject of breaking the loop never
Manuscript received January 9 1974. really comes up.
The author is with the Department of Electrical Engineering,
Newark College of Engineering, Newark, N. J. 07102. For impedance calculation Blackman's impedance rela-

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
ROSENSTARK: FEEDBACK AMPLIFIER ANALYSIS 193
tion is reviewed in section 4, and a simple derivation is
given in appendix B. This formula is very general and can -o2
be used in all situations including unbalanced bridge feed-
back. VI
VI 2'
In appendix C a set of quick-reference tables is pre- jr2
sented for the most common feedback amplifier con-
figurations. Figure 1. Series-shunt feedback pair.
2. ASYMPTOTIC GAIN FORMULA
Rather than use the customary A and : approach we
shall here analyze amplifiers by using the asymptotic gain VI(
formula (which is derived in appendix A)

Gf = K T + G (1)
1+ T 1 +T Figure 2. Simplified equivalent circuit for the series-shunt feedback
pair.
where
Gf _ Feedback Amplifier Gain (2)
(The term s'eries-shunt is used for identification and not
Go = Gf |T=0 _ Direct Transmission Term for classification.) (3)
K = Gf IT-.+_ Asymptotic Gain We shall arbitrarily select the controlled source of the
(4)
second transistor to calculate all the desired quantities.
T Return Ratio. (5)
Using the method of section 2.1 we draw the equivalent
circuit of figure 3 for calculating the return ratio T. We
All the quantities which enter equation 1 must be cal- proceed to calculate ib2 by inspection, and then the nega-
culated with respect to one and only one controlled source tive of ib2 is T. We find in a very straightforward manner
within the feedback amplifier. We shall refer to the con-
trolled source quantity Xb related to the controlling quan- T-= h R2 Re aiRl
+2
tity xa by the parameter k as follows: JR2 + Rf + (Re hibl) Re + hibi + hie2 (7)
|| R
Xb = kxa . (6) To find the asymptotic gain K we return to figure 2 and
impose the condition hfe2 -- c. This causes ib2 -* 0 and
2.1 Calculation of the Return Ratio T in turn ia --* 0, hence ibl O 0 (since hfel remains finite).
The return ratio T is determined by replacing the de- Accordingly Ve = V1 and then the relevant part of the
pendent source kxa by an independent source of value k, circuit is shown in figure 4. We see by inspection that
setting all independent sources to zero, and computing the
value of the variable Xa in the resulting system; the return V1= V2 j'
ratio T is equal to-Xa. Re + Rf

2.2 Calculation of the Asymptotic Gain K hence th-e ratio V2/Vi corresponding to K is
To find K we let the return ratio T -* oo. This is the
same as letting k -* c. In appendix A it is shown that the
+ Rf
Re (8) K=Re
controlling quantity Xa goes to zero. The amplifier gain
calculated with this condition imposed is K. It may be and this is approximately equal to the final gain of the
remarked at this point that inspection of equation 1 shows feedback amplifier if T >> 1. A gain specification can be
that for (loop gain) T >> 1, K will approximately equal the used to determine the ratio RfJRe.
final gain Gf of the feedback amplifier. Although Go is usually not of interest as mentioned in
section 2.3 (this is particularly true for this example where
2.3 Calculation of the Direct Transmission Term Go KT >» Go) we shall illustrate the method of calculation
We simply set T to zero by setting k equal to zero. For for situations where it might be of interest. In figure 2
many amplifiers Go, << KT and so contributes very little we set hf,2 0 and refer the resultant circuit to the emitter
=
to the gain Gf. In those situations it can be ignored and of the first transistor. Voltages are unchanged by this
need not be calculated. transformation and we get the diagram shown in figure 5.
The ratio V2/V1 which equals Go is determined by in-
3. APPLICATION OF THE ASYMPTOTIC METHOD spection, with the result
The use of the asymptotic gain formula will now be R2 Re
illustrated through some examples. R2+ Rf + (Re 1 hibl) hibl+ Re
(9)
Example 1: Consider the series-shunt feedback pair in
figure 1 and its simplified equivalent circuit in figure 2. We can at this point calculate the quantity KT/Go by

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
194 IEEE TRANSACTIONS ON EDUCATION, NOVEMBER 1974

ie, lb2 ib

-" h ( )s1 liII.eR1 2

VI R,
Figure 3. Circuit for calculating the return ratio T. (a) (b)
Figure 6. The emitter follower (a) and equivalent circuit (b).

G.f 1.
When equations 12, 13 and 14 arc combined in equation 1,
Figure 4. Circuit for calculating the asymptotic gain K. weobtain (to no one's surprise)

(1
hie + (1 +
+ hfe)Re
hfe)Re'
(14)
=hjb,Rf
hiF
4. FINDING IMPEDANCES
V4I R.
It has been 30 years since R. B. Blackman presented
his method for finding impedances in feedback amplifiers.
Figure 5. Circuit for calculating the direct transmission term G,L Although the method is very simple, it is largely ignored
in favor of techniques which require that the amplifier
using the results of equations 7, 8 and 9, to obtain be classified into one of four recognized configurations
before proceeding. The classification is used to determine
KT R1 Re±+Rf whether 1 + T should multiply or divide the open-loop
KT= aihf2 R
(10)
Go ~~R1 + hie2 R impedance. Blackman's impedance relation allows de-
and this can be used to ascertain the relative contribution termination of impedances in an unequivocal manner,
of Go to Gf. without the need to break the feedback loop and without
The problem is solved since the approximate final am- the need to categorize the amplifier.
plifier gain Gf is known and the amount of feedback T Blackman's method of finding impedance in a feedback
is also known. If a very accurate answer is desired then amplifier is embodied in the very simple relation
substitution into equation 1 can be carried out. We see + T8c (15)
Zab Zabo
that at no time was it necessary to classify the circuit as
=

1 + Toe
to the type of feedback being used, and also the question
of breaking the loop never came up. Furthermore, the where
methods of analysis required were merely voltage division, Zab Impedance at terminals a b with feedback -

current division and source conversion. amplifier normal. (16)


Example 2: It is an accepted fact that the emitter fol- Zabo Impedance at terminals a b with controlled -

lower of figure 6a possesses feedback, but it is a difficult source Xb = kxa set to zero. (17)
circuit to analyze by the A, approach, since there is no T,, -Return ratio for source Xb computed with ter-
logical way of removing feedback by breaking the loop. minals a b short-circuited.
- (18)
This circuit can be readily analyzed by ordinary methods, T,, Return ratio for source Xb computed with ter-
but a feedback technique that is general should be able to minals a b open-circuited.
- (19)
stand the test of being applied to degenerate circuits. We
In many cases either T,, or T,, will be zero, and the
will therefore proceed to test the asymptotic formula on
non-zero return ratio will correspond to the return ratio
the emitter follower equivalent circuit shown in figure 6b.
Applying the rules of sections 2.1, 2.2, and 2.3 we obtain computed when obtaining the amplifier gain, hence the
directly only new quantity that is required is Zab°
Example 3: For the series-shunt feedback pair of figures
1 and 2 find the input impedance Znl, and the output
hie (11) impedance Z22'. We set hfe2 0 and then -find by inspec-
=

K = 1 (12)
tion that
Zil,° = hie, + (1 + hfel) [Re II (Rf + R2)] (20)
hie Re (13)
.e+ Z22,0 R2 11 [Rf + Re || hibl].
G
Re = (21)
Equation 12 states the familiar result for the emitter (Note: The output impedance is by convention found
follower with the input source set to zero.)

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
ROSENSTARK: FEEDBACK AMPLIFIER ANALYSIS 195

We now turn our attention to finding the various return _ 0


ratios. +
VXa
Xb-
+ x +
When terminals 1 - 1' are open-circuited then ibl = 0, Xk
hence
Tll oc = 0. (22) Figure Al. Circuit for deriving the asymptotic gain formula.
When terminals 1 - 1' are short-circuited the circuit
is the same as in figure 3, hence Xb = kxa. (A-3)
Tll, = T(ofeq.7). (23) Solving for V2/V1 we obtain after some manipulation
By similar observations we find Gf = V2/VI = (A -BC/D) (-kD) + A
1 -kD (A-4)
T22, = T(ofeq.7) (24)
and We now need an interpretation of terms for the above
equation.
T22tsc = 0. (25) If the source V1 is set to zero and Xb is replaced by k,
We thus find by substitution into equation 15 that then we find for xa from equation A-2
Zni, = ZjjO1[I + T(of eq. 7)] (26) xa-= kD.
and But the above conditions correspond to those found in
section 2.1, hence La found above is the negative of the
Z22 t 0
return ratio. Accordingly
Z22 -1 ± T(ofo eq. 7)
=I (27)
- kD = T -Return Ratio. (A-5)
and the solution was obtained without prior knowledge
as to whether 1 + T belongs in the numerator or de- From equation A-4
nominator. BC
For bridge feedback Toc and T,6 are both non-zero, and A- D = Gf k-o = GfIT-oo
the impedance calculations cannot be performed by tradi- D
tional methods. In that case Blackman's method has a These conditions correspond to those found in section 2.2,
clear advantage. hence
5. CONCLUSION BC
A- = K- Asymptotic Gain. (A-6)
Use of the asymptotic gain method and Blackman's
impedance relation has led to greater student confidence
in being able to evaluate amplifier parameters irrespective Again from equation A-4
of the feedback connection. Students are particularly A = Gf k=o = Gf |T=0-
gratified to find that the results obtained by these methods
are in complete agreement with those obtained by mesh or The above conditions correspond to those found in section
nodal analysis. 2.3, hence
In addition the method used is directly applicable to A = G,, -Direct Transmission Term.
operational amplifiers, so that subject need not be covered (A-7)
separately. Using equations A-5, A-6, and A-7 in equation A-4 we
obtain the asymptotic gain formula
APPENDIX A
DERIVATION OF THE ASYMPTOTIC Gf =K T + Go (A-8)
1+T 1+T
GAIN FORMULA
We shall now establish the condition that is imposed
We draw the feedback amplifier as shown in figure Al on the feedback amplifier when k -> oo which is equiv-
and display the controlled source Xb which is contained alent to T X-* as can be seen from equation A-5. Elim-
inside the amplifier. inating Xb from equations A-2 and A-3 we obtain
We shall consider V1 and Xb as sources and V2 and xa as
outputs. Accordingly we write CVV1.
c
Xa
La1 - kCD
(A-9) =

V2 AV1 + BXb
= (A-1)
Xa= CV1 + DXb. (A-2) From this we conclude that for finite V1
In addition we have for the controlled source lim Xa = lim Xa = 0. (A-10)
k-.oo T-oo

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
196 IEEE TRANSACTIONS ON EDUCATION, NOVEMBER 1974

APPENDIX B
A SIMPLE DERIVATION OF
BLACKMAN'S RELATION
As in appendix A we draw the circuit in figure Bl and Figure Bi. Circuit for deriving Blackman's impedance formula.
treat I and Xb as sources, and V and xa as outputs.
V =AI+ BXb (B-1)
Xa =CI + Dxb. (B-2)
Also El
Xb = kxa. (B-3)
Solving for V/I, we obtain the impedance at terminals
a - b after some rearranging

Z= A 1 - k(AD - BC)/A (B-4)


1 - kD
We now need an interpretation of terms for the above
equation. Z, Z, H,
; ZL
z,L ZZ +ZL
Z, + H,
=

From equation B-1 we have


v
A = X6 Z= +Z2+ZL
TZ+H K -Z2
=
I x b=°
The above corresponds to the definition in equation 17, KT. Hf Z
hence Go H,
A = Zabo. (B-5) Figure Cl. Shunt-shunt amplifier.

If V = 0 and Xb = k, then equations B-1 and B-2 give


the result for x.:
Xa= k (AD - BC) /A.
But the above conditions correspond to those found in
equation 18, hence
-k (AD - BC)/A = T8c. (B-6)
If I = 0 and Xb = k, then from equation B-2
Xa= kD.
The above conditions correspond to those found in equa-
tion 19, hence Z, +HI ;
Z2 =2 Z42 Ze2+ re2+ Z+2
= -

-kD =
Toc. (B-7)
Substituting equations B-5, B-6, and B-7 into equation T =Z_ Hf Z
H1+ZI Z4+Z42(1+Zk2Ze
K =L2(ZL+IZ2)
B-4 we obtain
KT=Hf Z (I + Z2)
Zab =
Zab 1 +

+
T8C

T0c
(B-8)
Figure C2. Shunt-series BJT amplifier.
which is Blackman's impedance relation.
To use the tables, a portion of the amplifier has to be
APPENDIX C replaced by an unilateral equivalent circuit. For example,
in the series-shunt feedback case of figure C3, the ampli-
A TABLE OF SOME COMMON CIRCUITS fier denoted by the triangle and the collector resistor of
In this section a table of common feedback amplifier the first transistor must be replaced by a hybrid model
configurations is presented for quick reference (Figures as shown in the second diagram. The quantities T, K, and
C1-C8). Although the tables are self-explanatory, some Go were found with respect to the only controlled source
comments are in order. depicted in the second diagram.

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
ROSENSTARK: FEEDBACK AMPLIFIER ANALYSIS 197

Z/el = z ei rb ; ZLL =ZzZL


+
re-, R +ZL = Ze + rel + rbi+;
zIelZei+rej rbfZ . Ze2=Ze2+r~2+
ze2 S + re2+ rb2
+Z
+8

T= a, Hf Z'L K = +1 ILHfZ
Z2 + Z/ +Z/IZ(Ze Ze T =
41+ Ze2 + Ze/
KTZe
K = 1
2 Ze
e2 I
KT
a, Hf(I +-2) G= -aaHz
KT
Hf Ze
Figure C3. Series-shunt BJT amplifier. Figure C4. Series-series BJT amplifier.

El ~~~~~z

El?) Z2 s2 IT
zs
zl r - -

--I ,J
II
+Z E2
E2 + I--' L 2

zizi Z2
Z/2S2 s' rd2-
z/ 1 Zizi z Z ZL ZIZ + Zj +
I L2
1
-1-zi L- Z + ZL Zs2=s
+ L +2
P2 A Z'
;0
=

T ZL AZI K =- Z IRZ,+2+42 AZI+ Z'- Z2 (


Z + ZL Z, 2+ZL
KT =A A KT=A- 2 (1++Z
GO 1 +2 Zs
Go Z

Figure C5. Shunt-shunt amplifier. Figure C6. Shunt-series FET amplifier.

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.
198 IEEE TRANSACTIONS ON EDUCATION, NOVEMBER 1974

E2
i

rd2+Z2

rdi +Zi
Zs + rdi + Zi
KL=__ T = L2 Z-. AZ K=- I 2ZL
1-fL2 Zl 4 I+ Lj Zs
zs5
Z+ZL Z+z2/L+Z2 t + Z2 + t,1 45
~ Zs KT=_A Zi. P2
KT AZi(1 ZL? G. Zs I+ A2
Figure C8. Series-series FET amplifier.
Figure C7. Series-shunt FET amplifier.
and to Dr. Jeannette Rosenstark for her valuable criti-
The triangular amplifier is integrated or discrete, so cism of the manuscript.
the analysis is sufficiently general to cover a multitude of
situations. Since the triangular amplifier may contain REFEREN CES
either BJT's or FET's, the analysis of amplifiers with
mixed active elements is also possible. 1. H. S. Stewart, "Engineering Electronics", Allyin & Bacon, 1969,
Chapter 12.
2. C. Belove and D. L. Schilling, "Feedback Made Easy for the
ACKNOWLEDGMENT Undergraduate", IEEE Trans. on Education, Vol. 12, June 1969,
pp. 97-103.
The author wishes to express his gratitude to Prof. 3. P. E. Gray and C. L. Searle, "Electronic Principles, Physics,
Joseph Frank of Newark College of Engineering, for the Models and Circuits", John Wiley, 1967, Ch. 18.
4. R. B. Blackman, "Effect of Feedback on Impedance", BSTJ
many enlightening discussions on the subject of feedback, Vol. 22, October 1943, pp. 269-277.

Short Notes
MECAP-An Analysis Program for Microwave sort of analysis program should be used in an undergraduate micro-
Engineering Courses wave engineering course.

JOHN C. FIELD, MEMBER, IEEE, AND DAVID L. HERRICK, INTRODUCTION


STUDENT MEMBER, IEEE
Most undergraduate microwave engineering courses include
Abstract-There exists a need, in undergraduate microwave en- impedance matching methods, e.g., stub tuning and quarterwave
gineering courses, for a simple analysis program similar to ECAP, transformers. However, it is quite tedious to calculate the network's
but which is applicable to distributed circuits. Such a program has frequency response by hand even when using Smith chart methods.
been developed at the University of Maine, Orono to analyze a In order to perform these and other calculations many industrial
network of cascaded two-ports with the possibility of one feedback firms have developed analysis programs specifically for microwave
path. This configuration, while not the most general, will encompass networks [1-3]. These programs are usually of two general types.
The first type analyzes cascaded two port networks and may or
nearly all networks encountered in an undergraduate course. It is may not allow feedback. The second type is more general in that
also fast and very easy to use. Some examples of its use are given. it will handle multiport networks but it is less efficient when applied
Student response was very favorable and it is concluded that some to cascaded networks.
An analysis program would then be very useful in a microwave
engineering course. It would allow the student to both verify his
Manuscript received April 3, 1974. designs and determine their frequency responses. In addition it
The authors are with the Department of Electrical Engineering,
University of Maine, Orono, Me. 04473. would be following the industrial approach. To satisfy this need

Authorized licensed use limited to: Politecnico di Torino. Downloaded on March 23, 2009 at 10:49 from IEEE Xplore. Restrictions apply.