Feedback Amplifiers

Dr. Sandeep J. Rajput EC,GECG

 Definition

▪ Feedback : Feedback is a process in which some

part of output (current or voltage) is returned
back to input.
▪ Feedback amplifier : The amplifier which
operates on the principle of feedback is known
as feedback amplifier.
▪ Negative Feedback
▪ Positive Feedback
Positive Feedback Negative Feedback
Positive feedback is also called as Negative Feedback is also called as
regenerative feedback. degenerative feedback

The system gain of the positive The system gain of the negative
feedback is high feedback is low.

The Stability of positive feedback is The Stability of negative feedback is

less Comparatively more

The phase shift of positive feedback is The phase shift of negative feedback
00 or 3600 is 1800

Feedback is taken from Non-inverting Feedback is taken from Inverting

terminal of an op-amp terminal of an op-amp

Transfer Function of positive Transfer Function of negative

feedback is TF = G/1-GH feedback is TF = G/1+GH

Sensitivity of positive feedback is Sensitivity of negative feedback is

Low High

The positive feedback can be used The negative feedback can be used in
in Oscillators Amplifiers
 Introduction
▪ It’s impossible to think of electronic circuits
without some forms of feedback.
▪ Negative feedback
▪ Desensitize the gain
▪ Reduce nonlinear distortion
▪ Reduce the effect of noise
▪ Control the input and output impedance
▪ Extend the bandwidth of the amplifier
The basic idea of negative feedback is to trade off
gain for other desirable properties.
The General Feedback Structure
This is a signal-flow diagram, and the quantities x represent
either voltage or current signals.
 The General Feedback Equation
▪ Closed loop and open loop
▪ Closed loop gain
xo A
Af  =
xs 1 + A
▪ Feedback factor β
▪ Loop gain Aβ
▪ Amount of feedback (1+ Aβ)
Some Properties of Negative Feedback

▪ Gain desensitivity
dA f 1 dA
=
A f 1 + A A
▪ Bandwidth extension
▪ Noise reduction
▪ Reduction in nonlinear distortion
 Classification of basic amplifier

▪ Voltage amplifier

▪ Current amplifier

▪ Trans-resistance amplifier

▪ Trans-conductance amplifier
Thevenin’s equivalent
Circuit of voltage amplifier
Norton’s equivalent circuit
of current amplifier
Equivalent circuit of
Transconductance amplifier
Equivalent circuit of Trans-
resistance amplifier
 The Four Basic Feedback Topologies
▪ Voltage amplifier - series-shunt feedback

Voltage mixing and voltage sampling

▪ Current amplifier - shunt-series feedback

Current mixing and current sampling

▪ Transconductance amplifier - series-series feedback

Voltage mixing and current sampling

▪ Trans-resistance amplifier - shunt-shunt feedback

Current mixing and voltage sampling

Feedback Topologies
 Gain of Feedback Amplifier

▪ The gain of the amplifier is represented by “A” and is the ratio of the
output voltage Vo to the input voltage Vi and is known as the
feedback fraction. The voltage feedback is given by, Vf = βVo from
the output of the amplifier.
▪ For negative feedback :
Vi = Vs − Vf = Vs − βVo
▪ For positive feedback :
Vi = Vs + Vf = Vs + βVo
 Gain of Feedback Amplifier
▪ Let’s consider the case of negative feedback, the output voltage Vo
should be equal to Vs−βVo times gain of the amplifier “A”.
AVs − AβVo = Vo
AVs = Vo(1+Aβ)
Vo/Vs = A/1+Aβ
▪ Where Vs is applied signal voltage. The overall gain with feedback
from the amplifier is,
Af = Vo/Vs
▪ Where Vo is the output voltage and Vs is the input signal voltage. The
gain feedback for a positive feedback amplifier,
Af = A/1−Aβ
▪ And for a negative feedback amplifier,
Af=A/1+Aβ
General Characteristics of
Negative Feedback
Amplifiers
 Stabilization of Gain
Sensitivity of transfer Amplification: The fractional change in
amplification with feedback divided by the fractional change without
feedback is called the sensitivity of the transfer gain.

Mathematically, the sensitivity of the dA f / A f

transfer gain can be written as follows: S = dA / A

A
We know that, A f =
1 + A
d (1 + A) dA
dA f A dA − (1 + A) dA A − 1 − A 1
So, = = =−
dA (1 + A) 2 (1 + A) 2 (1 + A) 2

dA f 1 A 1 Af
=− =− =−
dA (1 + A) 2 1 + A A(1 + A) A(1 + A)
 Stabilization of Gain
dA f / A 1
S= =
dA / A 1 + A

1
Hence, the sensitivity is S =
1 + A

For example, if S=0.1 then, dA f dA

= 0.1
A A

That means the percentage change in gain with feedback is one-tenth

the percentage variation in amplification if no feedback is present.
 Stabilization of Gain
Desensitivity of transfer Amplification: The reciprocal of the
sensitivity is called the desensitivity D, or

D = 1 + A (13 − 7)

From Eq. Af = A/(1+bA), it is seen that the transfer gain is divided by

the desensitivity after feedback is added.
Thus A f = A / D (13 − 8)

A A 1
If A  1, Then A f =  = (13 − 9)
1 + A A 

The gain may be made to depend entirely on the feedback

network.
 Frequency Distortion
▪ It is seen from Eq. Af=1/ that if the feedback network
is purely resistive, the gain with feedback is not
depends on frequency even though the basic amplifier
gain is frequency dependent.

▪ So, the frequency distortion arising because of varying

gain, A, with frequency is considerably reduced in a
negative feedback amplifier.
 Nonlinear Distortion
▪ Suppose that a large amplitude signal is applied to a stage
of an amplifier so that the operation of the device extends
slightly beyond its range of linear operation, and as a
consequence the output signal is slightly distorted.

▪ Negative feedback now introduced, and the input signal

is increased by the same amount by which the gain is
reduced, so that the output signal amplitude remains the
same.

▪ It is clear from Eq. Af=1/ that the negative feedback

reduces the dependence of the overall closed-loop gain
(Af) on the open-loop gain (A) of the amplifier.
 Nonlinear Distortion

The open loop gain drops

from 1000 (A1) to 100 (A2).
Negative feedback gain
=0.01 then,

1000
Af 1 = = 90.9
1 + 1000  0.01
100
Af 2 = = 50
1 + 100  0.01
That means the closed-loop gain drops from 90.9 to 50.
This means it gets linearized.
 Reduction of Noise
Negative feedback can be employed to reduce the noise in
an amplifier (i.e. to increase the signal-to-noise ratio, SNR).
S Vs
The signal-to-noise (SNR) for the amplifier A1 is =
N Vn

We may precede the original amplifier A1 by the clean

amplifier A2 and apply negative feedback.
 Vs A1A2 Vn A1
Vo = +
1+ A1A2 1+ A1

When negative feedback is applied, the noise introduced in

the amplifier is attenuated by the factor 1/(1+AB). If
(1+AB)≫1, the noise will be significantly reduced.

It is seen from above equation that the signal-to-noise

ratio is increased by the use of negative feedback.
Increase Bandwidth

Gain x Bandwidth = Constant

 Input Resistance
If the feedback signal is returned to the input in series with the
applied voltage, it increases the input resistance.

Since the feedback voltage Vf oppose Vs, the input current Ii is

less than it would be if Vf were absent.

Hence the input resistance Rif = Vs/Ii is greater than the input
resistance without feedback Ri.

For this type of feedback topology Rif = Ri(1+A) = RiD.

If the feedback is returned to the input in shunt with the

applied current, it decreases the input resistance.
 Input Resistance
Since, Is = Ii+If, then the current Ii (for a fixed value of Is) is
decreased from what it would be if there were no feedback
current.

Hence the input resistance Rif = Vi/Is = RiIi/Is is decreased

because of this type of feedback.

For this type of feedback topology Rif = Ri/(1+A) = Ri/D.

 Output Resistance
Negative feedback which samples the output voltage, regardless of
how this output signal is returned to the input, tends to decrease the
output resistance (Rof << Ro).
Negative feedback which samples the output current, regardless of
how this output signal is returned to the input, tends to increase the
output resistance (Rof >> Ro).
The output resistance for the feedback amplifier can be defined as
the resistance with feedback Rof looking into the output terminals
with RL disconnected.
The output resistance is determined by applying a voltage V,
resulting in a current I, with Vs (Vs =0 where input source is
voltage) shorted out or Is (Is =0 where input source is current)
opened out. So, Rof=V/I.
 Input Resistance of Voltage-Series Feedback
In the topology the amplifier is replaced by its Thevenin’s model.
The input impedance with feedback is Rif=Vs/Ii. Also,
Vs = Ii Ri + V f = Ii Ri + Vo (13 − 11)
AvVi RL Av I i Ri RL
and Vo = = = AV I i Ri (13 − 12)
Ro + RL Ro + RL
V A R
where, AV = o = v L (13 − 13)
Vi Ro + RL
 Input Resistance of Voltage-Series Feedback

Vs = Ii Ri + AV Ii Ri = Ii Ri (1 + AV )

Vs
Rif = = Ri (1 + AV ) = Ri D (13 − 14)
Ii

Here, D =1 + AV

Whereas, Av represents the open-circuit voltage gain without feedback,

AV is the voltage gain without feedback taking the load RL into account.
Therefore,
Av RL
Av = lim AV = lim (13 − 15)
RL →  RL → Ro + RL
 Output Resistance of Voltage-Series Feedback
Replacing the output voltage Vo by V of Figure and setting Vs=0, we
obtained
V − AvVi
I= and Vi = −V f = − V
Ro
From this equations, we
V + AvV V (1 + Av )
obtained as follows: I= = (13 − 28)
Ro Ro
V Ro
Hence, Rof = = (13 − 29)
I 1 + Av
Note that Ro is divided by
the desensitivity factor
1+Av, which contains the
open-circuit voltage gain Av
(not AV).
 Output Resistance of Voltage-Series Feedback
The output resistance with feedback which includes RL as part of the
amplifier is given by Rof in parallel with RL, or

Ro
RL
' R R
of L 1 + Av
Rof = Rof RL = =
Rof + RL Ro
+ RL
1 + Av
Ro RL Ro RL /( Ro + RL )
= =
Ro + RL + Av RL 1 + Av RL /( Ro + RL )

R R /( R + R ) R '
' =
Rof o L o L = o (13 − 30)
1 + Av RL /( Ro + RL ) 1 + AV

Where Ro’=Ro||RL is the output resistance without feedback but with RL

considered as part of the amplifier.
 Input Resistance of Current-Series Feedback
The amplifier is replaced by its Thevenin’s input model and Norton’s
output model.
The input impedance with feedback is Rif = Vs/Ii.
Also Vs = Ii Ri + V f = Ii Ri + I o (13 − 11.1)
GmVi Ro Gm I i Ri Ro
and, I o = = = GM I i Ri (13 − 12.1)
Ro + RL Ro + RL
Io Gm Ro
where, GM = = (13 − 18)
Vi Ro + RL
 Input Resistance of Current-Series Feedback
Vs = I i Ri + GM I i Ri = I i Ri (1 + GM )

Vs
Rif = = Ri (1 + GM ) = Ri D (13 − 16)
Ii

Here, D = 1 + GM

Whereas Gm represents the short-circuit transconductance without

feedback, Eq. (13-18) indicates that GM is the transconductance
without feedback taking the load RL into account.
Therefore,
G m Ro
G m = lim G M = lim (13 − 17)
RL →0 R L → 0 Ro + R L
 Output Resistance of Current-Series Feedback
Replacing the output voltage Vo and Io by V and -I, respectively, and
setting Vs=0, we obtained
V
= GmVi + I and Vi = −V f = I
Ro

From this equations, we obtained as follows: V = G I + I

m
Ro
V
Rof = = Ro (1 + Gm ) (13.38.1)
I

Note that Ro is
multiplied by the
desensitivity factor
1+Gm, which contains
the transconductance
Gm (not GM).
 Output Resistance of Current-Series Feedback
The output resistance with feedback which includes RL as part of the
amplifier is given by Rof in parallel with RL, or

' =R Rof RL
Rof of RL =
Rof + RL

' Ro (1 + Gm ) RL
Rof =
Ro (1 + Gm ) + RL

' Ro RL (1 + Gm )
Rof =
Ro + RL + Gm Ro
' = Ro RL (1 + Gm ) /( Ro + RL )
Rof
1 + Gm Ro /( Ro + RL )

' ' 1 + Gm

Rof = Ro (13.38.2)
1 + GM
 Input Resistance of Current-Shunt Feedback
The amplifier replaced by its Norton’s model.
The input impedance with feedback is Rif = Vi/Is.
Also I s = I i + I f = I i + I o (13 − 19)
Ai I i Ro
and Io = = AI I i (13 − 20)
Ro + RL
Io Ai Ro
where, AI = = (13 − 21)
I i Ro + RL
 Input Resistance of Current-Shunt Feedback
I s = Ii (1 + + AI ) (13 − 22)
Using Rif=Vi/Is and Ri=Vi/Ii, we obtain
Vi Vi Vi 1
Rif = = =
I s I i (1 + + AI ) I i 1 + + AI

Ri R
Rif = = i (13 − 23)
1 + + AI D

Here, D = 1 + + AI
Whereas Ai represents the short-circuit current gain without feedback
and AI is the current gain without feedback taking the load RL into
account.
Therefore, Ai Ro
Ai = lim AI = lim (13 − 24)
RL → 0 RL →0 Ro + RL
 Output Resistance of Current-Shunt Feedback
Replacing the output voltage Vo and Io by V and -I, respectively, of
and setting Is=0, we obtained
V
= Ai I i + I and Ii = − I f = − I o = I
Ro
V
From this equations, we obtained as follows: = Ai I + I
Ro
V
Rof = = Ro (1 + Ai  ) (13 − 35)
I

Note that Ro is multiplied

by the desensitivity
factor 1+Ai, which
contains the short-circuit
current gain Ai (not AI).
www.assignmentpoint.com
 Output Resistance of Current-Shunt Feedback
The output resistance with feedback which includes RL as part of the
amplifier is given by Rof in parallel with RL, or
' =R Rof RL
Rof of RL =
Rof + RL

' Ro (1 + Ai  ) RL
Rof =
Ro (1 + Ai  ) + RL

' Ro RL (1 + Ai  )
Rof =
Ro + RL + Ai Ro

' Ro RL (1 + Ai  ) /( Ro + RL ) ' ' 1 + Ai

Rof = Rof = Ro (13 − 37)
1 + Ai Ro /( Ro + RL ) 1 + AI

For RL=, AI=0 and , so that Eq. reduces to

' = R (1 + A ) that means R ' = R
Rof o i of of
 Input Resistance of Voltage-Shunt Feedback
The amplifier is replaced by its Norton’s input model and Thevenin’s
output model.
From the Fig. the input impedance with feedback is Rif=Vi/Is.
Also, I s = Ii + I f = Ii + Vo (13.19.1)

R I R
and Vo = m i L = RM I i (13.20.1)
Ro + RL
Rm RL
where, RM = (13 − 26)
Ro + RL
 Input Resistance of Voltage-Shunt Feedback
I s = I i + I f = I i + RM I i = I i (1 + RM )

From Fig., Rif=Vi/Is and Ri=Vi/Ii. we obtain,

Vi Vi Vi 1
Rif = = =
I s I i (1 + RM ) I i 1 + RM
Ri R
Rif = = i (13 − 25)
1 + RM D

Here, D = 1 + RM
Whereas Rm represents the open-circuit transresistance without
feedback, Eq. indicates that RM is the transresistance without feedback
taking the load RL into account.
Therefore,
Rm RL
Rm = lim RM = lim (13 − 27)
RL →  RL → Ro + RL
 Output Resistance of Voltage-Shunt Feedback
Replacing the output voltage Vo and Io by V and -I, respectively, of Fig.
and setting Is=0, we obtained
V − Rm I i
I= and I i = − I f = − Vo = − V
Ro
V + VRm
From this equations, we obtained as follows: I =
Ro
V Ro
Rof = = (13 − 32.1)
I 1 + Rm

Note that Ro is divided by

the desensitivity factor
1+Rm, which contains
the transresistance Rm
(not RM).
 Output Resistance of Voltage-Shunt Feedback
The output resistance with feedback which includes RL as part of the
amplifier is given by Rof in parallel with RL, or
Rof R L Ro
' =R RL
Rof of RL = ' 1 + Rm
Rof + R L Rof =
Ro
+ RL
1 + Rm
' == Ro RL Ro R L /( Ro + R L )
Rof ' =
Ro + RL + Rm RL Rof
1 + Rm R L /( Ro + R L )

R '
' =
Rof o (13 − 32.2)
1 + RM
where R’O is the output resistance without feedback but with RL
considered as part of the amplifier.
Note that R’O is now divided by the desensitivity factor D=1+RM which
contains the voltage gain RM that takes RL into account.
Comparison of feedback topologies
 Method of Analysis of a Feedback Amplifier
It is desirable to separate the feedback amplifier into two blocks, the
basic amplifier A and the feedback network , because with a knowledge
of A and , we can calculate the important characteristics of the feedback
system, namely, Af, Rif, and Rof.
The basic amplifier configuration without feedback but taking the
loading of the  network into account is obtained by applying the
following rules:
To find the input circuit:
1. Set Vo=0 for voltage sampling. In other words, short the output node.
2. Set Io=0 for current sampling. In other words, open the output loop.
To find output circuit:
1. Set Vi=0 for shunt comparison. In other words, short the input node.
2. Set Ii=0 for series comparison. In other words, open the input loop.
 Method of Analysis of a Feedback Amplifier
The complete analysis of a feedback amplifier is obtained by
carrying out the following steps:
1. Identify the topology
(a) Is the feedback signal Xf a voltage or a current?
In other words, is Xf applied in series or in shunt with the
external excitation?
(b) Is the sampled signal Xo a voltage or current?
In other wards, is the sampled signal taken at the output node
or from the output loop?
2. Draw the basic amplifier circuit without feedback,
following the rules listed above.
 Method of Analysis of a Feedback Amplifier
3. Use the Thevenin’s source if Xf is a voltage and a Norton’s
source if Xf is a current.
4. Replace each active device by the proper model (for
example, hybrid- model for a transistor at high frequency or
the h-parameter model at low frequency).
5. Indicate Xf and Xo on the circuit obtained by carrying out
steps 2, 3, and 4. Evaluate = Xf / Xo.
6. Evaluate A by applying KVL and KCL to the equivalent
circuit obtained after step 4.
7. From A and , find D, Af, Rif, Rof, and Rof ’.
 Amplifier Characteristics

Feedback Topology Input Resistance Output Resistance

Increases Decreases
Voltage Series
Rif = Ri*(1+A*β) Rof=Ro/(1+A*β)

Increases Increases
Current Series
Rif = Ri*(1+A*β) Rof = Ro*(1+A*β)

Decreases Increases
Current Shunt
Rif = Ri/(1+A*β) Rof = Ro*(1+A*β)

Decreases Decreases
Voltage Shunt
Rif = Ri*(1+A*β) Rof=Ro/(1+A*β)