Unit 4

Frequency Response and Feedback Amplifier

Feedback is a powerful technique that finds wide application in analog circuits. For example,
negative feedback allows high-precision signal processing, and positive feedback makes it
possible to build oscillators.
Important benefits that result from feedback are
1. Gain Desensitization
2. Terminal Impedance Modification
3. Bandwidth Modification
4. Nonlinearity Reduction
Four feedback topologies are
1. Voltage Amp.
2. Transimpedance Amp
3. Transconductance Amp
4. Current Amp
Feedback circuit analysis can be done using following techniques
1. Two-Port Technique
2. Bode’s Technique
3. Blackman’s Theorem
5.3 General Considerations
Figure 5.7 shows a negative-feedback system, where H(s) is called the feedforward network
and G(s) is called the feedback network. Since the output of G(s) is equal to G(s)Y (s), the
input to H(s), called the feedback error, is given by [X(s) − G(s)Y (s)].
That is
Y (s) = H(s)[X(s) − G(s)Y (s)] (1)
Thus, Y (s) /X(s) = H(s) /[1 + G(s)H(s)] (2)

Figure 5.7 General Feedback System

H(s) the “open-loop” transfer function
Y (s)/X(s) the “closed-loop” transfer function.
For amplifier circuit analysis, H(s) represents an Amplifier and G(s) is a frequency-
independent quantity. A fraction of the output signal is sensed and compared with the input,
generating an error term.
In a well-designed negative-feedback system, the error term is minimized, thereby making
the output of G(s) an accurate “copy” of the input and hence the output of the system a
faithful or scaled replica of the input as shown in Figure 5.8 .
We can also approximate that the input of H(s) is a “virtual ground” because the signal
amplitude at this point is small. In subsequent developments, we replace G(s) by a
frequency-independent quantity β called “feedback factor.”

Figure 5.8 Similarity between output of feedback network and input signal.

1
The elements exist in every feedback system are
(1) The feedforward amplifier
(2) A means of sensing the output
(3) The feedback network
(4) A means of generating the feedback error, i.e., a subtractor or an adder.

5.4 Properties of Feedback Circuits

Important benefits that result from feedback are
1. Gain Desensitization
2. Terminal Impedance Modification
3. Bandwidth Modification
4. Nonlinearity Reduction

5.4.1 Gain Desensitization

Consider the common-source stage shown in Fig. 5.9, where the voltage gain is equal to
gm1rO1. A critical drawback of this circuit is the poor definition of the gain: both gm1 and
rO1 vary with process and temperature.

Figure 5.9 Common-Sources IC Amplifier

The common source amplifier with feedback had shown Fig. 5.10. At relatively low
frequencies such that C2 draws a negligible small-signal current from the output node.
Hence Vout/VX = −gm1rO1 because the entire drain current flows through rO1.

Figure 5.10 Common-Sources IC Amplifier with feedback (from output to input through C2)

(Vout − VX )C2S = (VX − Vin)C1S (1)

If gm1rO1 is sufficiently large, the 1/(gm1rO1) terms in the denominator can be neglected,

(3)

2
  Compared to gm1rO1, this gain can be controlled with much higher accuracy because
it is given by the ratio of two capacitors.
 If C1 and C2 are made of the same material, then process and temperature variations
do not change C1/C2.
 The above example reveals that negative feedback provides gain “desensitization,
1. The closed-loop gain is less sensitive to device parameters than the open-loop gain
is.
2. The negative feedback “stabilizes” the gain and hence “improves the stability.”
 The frequency stability is typically worsens as a result of negative feedback.

For a more general case in Fig. 5.11, gain desensitization can be quantified by writing

(4)
Where β is called “feedback factor, A is called “open-loop gain” , Y/X is “closed-loop gain”
and the quantity βA is called “loop gain,”

Figure 5.11 Simple Feedback System

Assumed that β A >>1, we note that the closed-loop gain is determined, to the first order
by the feedback factor, β.

(5)
For higher βA, Y/X is less sensitive for the variations in A. To maximize βA, as β increases,
the closed loop gain, Y/X ≈ 1/β, decreases. Hence trade-off between precision and the closed-
loop gain i.e., we begin with a high-gain amplifier and apply feedback to obtain a low, but
less sensitive closed-loop gain.
The output of the feedback network is equal to βY. From equation (4), substituting for X we
get
βY = β [X ・A/(1 + β A)] (6)
As βA becomes much greater than unity, βY also approaching X.
Steps to determine loop gain βA (Figure 5.12)
1. Set the main input to (ac) zero,
2. Break the loop at some point,
3. Inject a test signal in the right direction,
4. Follow the signal around the loop,
5. Obtain the value that returns to the break point.
6. The negative of the transfer function thus derived is the loop gain.

3
 Figure 5.12 Computation of loop gain
we have VF = Vt β(−1)A (7)
VF/Vt = −βA. (8)

Example 5.4 Determine the loop gain of the circuit shown in Figure 5.13

Figure 5.13
Solution

Conclusion
1. Large variations in A affect Y/X negligibly if βA is large. Such variations can arise
from different sources: process, temperature, frequency, and loading.
2. A drops at high frequencies, Y/X varies to a lesser extent, and the bandwidth is
increased.
3. A decreases because the amplifier drives a heavy load, Y/X is not affected much.

5.4.2 Terminal Impedance Modification

Objective is to compute the input resistance and output resistance at relatively low
frequencies with and without feedback.
Input Resistance
Consider a common-gate circuit Figure 5.14 with feedback with current source M2.

4
 Figure 5.14 Common-gate circuit with feedback
where a capacitive voltage divider senses the output voltage of a common-gate stage,
applying the result to the gate of M2.
Neglecting channel length modulation and the current drawn by C1, the resistance Rin,open is
computed by breaking the feedback loop as shown in Figure 5.15.

Figure 5.15 Open-loop circuit

(1)
For the closed-loop circuit, as depicted in Figure 5.16 the input is denoted as Rin,closed ,
computed as Rin,closed = VX /IX

Figure 5.16 Closed-loop circuit - calculation of input resistance

5
The small-signal drain current of M2 is

The current Ix is sum of drain current of M1 and M2

We therefore conclude that this type of feedback reduces the input resistance by a factor of
[1 + gm2RDC1/(C1 + C2)]. The gm2RDC1/(C1 + C2) is loop gain(βA) of the circuit.

Output Resistance
Consider CS stage with feedback as shown in Figure 5.17. M1, RS, and RD constitute a
common-source stage and C1,C2, and M2 sense the output voltage.

Figure 5.17 CS stage with feedback

The drain current of M2 fed to source of M1 is [C1/(C1 + C2)]Voutgm2. To compute the output
resistance at relatively low frequencies, we set the input to zero as in Figure 5.18.

Figure 5.18 Closed-loop circuit - calculation of output resistance

6
Hence from above equation implies that this type of feedback decreases the output resistance.
The denominator is indeed equal to one plus the loop gain(βA).

5.4.3 Bandwidth Modification

Objective is to compute the bandwidth at relatively low frequencies with and without
feedback.
Consider the feedforward amplifier has a one-pole transfer function A(s), where A0 denotes
the low-frequency gain and ω0 is the 3-dB bandwidth

For the closed looped system Figure 5.19, transfer function is given as

Figure 5.19 Closed Loop System

The numerator is simply the closed-loop gain at low frequencies and the denominator reveals
a pole at (1 + βA0)ω0.
Thus, the 3-dB bandwidth has increased by a factor of (1 + βA0), albeit at the cost of a
proportional reduction in the gain as depicted in Figure 5.20

7
 Figure 5.20 Bandwidth modification as a result of feedback
The increase in the bandwidth fundamentally originates from the gain desensitization
property of feedback. The open loop gain A varies with frequency (as shown in figure 5.20)
rather than process or temperature, but negative feedback still suppresses the effect of this
variation.
If A is large enough, the closed-loop gain remains approximately equal to 1/β even if A
experiences substantial variations. At high frequencies, A drops to such low levels that βA
becomes comparable with unity and the closed-loop gain falls below 1/β.
The “gain-bandwidth product” of a one-pole system is equal to A0ω0 and does not change
much with feedback.

5.4.4 Nonlinearity Reduction

Objective is to understand reduction of nonlinearity in analog circuits with feedback as
shown in Figure 5.21. A nonlinear characteristic is one that departs from a straight line,
i.e., one whose slope varies.

Figure 5.21 Input-output characteristic of a nonlinear amplifier

(a) before and (b) after applying feedback
The gain of an open-loop amplifier varies from A1 to A2 . With feedback, that is a closed-
loop feedback system incorporating such an amplifier exhibits less gain variation and hence a
higher linearity.
To quantify this effect, we note that the open-loop gain ratio between regions 1 and 2 in Fig.
5. 21 is equal to

Assuming A2 = A1 − ΔA, we can write

8
Let us place this amplifier in a negative-feedback loop. For the closed-loop gain ratio, we
have

The gain ratio is much closer to 1 with feedback if the loopgain (1 + βA2) is large.

9
5.5 Types of Amplifiers
The amplifiers can be categorized based on input and output impedance of amplifier which
are determined whether input sense current/voltage and produce current/voltage. Four
different amplifiers based on impedance are

1. Voltage Amplifier
2. Transimpedance Amplifier
3. Transconductance Amplifier
4. Current Amplifier
Figure 5.22 show types of amplifiers along with their idealized models i.e Zin/Zout is zero or
infinity.

Figure 5.22 Types of amplifiers along with their idealized models

Figure 5.23 show types of amplifier with nonideal model. A nonideal voltage amplifier
draws current from its input and exhibits a finite output impedance. A nonideal
transimpedance amplifier may have finite input and output impedances. These amplifiers may
also have internal feedback from their output to their input, e.g., due to CGD, but we neglect
that for now.

10
 Figure 5.23 Equivalent models for nonideal amplifiers

Figure 5.24 illustrates simple implementations of each amplifier. In Fig. 5.24 (a), a common-
source stage senses and producesVoltages, and in Fig. 5.24 (b), a common-gate circuit serves
as a transimpedance amplifier, converting the source current to a voltage at the drain. In Fig.
5.24 (c), a common-source transistor operates as a transconductance amplifier (also called a
V/I converter), generating an output current in response to an input voltage, and in Fig. 5.24
(d), a common-gate device senses and produces currents.

Figure 5.24 Four types of amplifiers

5.6 Feedback Topologies
As depicted in Figure 5.25, X and Y can be a current or a voltage quantity. The main
amplifier is called the “feedforward” or simply the “forward” amplifier, around which we
apply feedback to improve the performance. Objective is to study four “canonical”
topologies resulting from placing each of the four amplifier types in a negative-feedback
loop.
1. Voltage-Voltage Feedback : Forward Amplifier is Voltage Amplifier
2. Current-Voltage Feedback : Forward Amplifier is Transconductance Amplifier
3. Voltage-Current Feedback : Forward Amplifier is Transimpedence Amplifier
4. Current-Current Feedback : Forward Amplifier is Current Amplifier

Figure 5.25 Canonical feedback system

11
5.6.1 Voltage-Voltage Feedback
This topology senses the output voltage and returns the feedback signal as a voltage. This
configuration is also called “series-shunt” feedback, where the first term refers to the input
connection and the second to the output connection. The feedback network is connected in
parallel with the output and in series with the input port as shown in Figure 5.26.
An ideal voltage –voltage feedback network exhibits infinite input impedance and zero
output impedance because it senses a voltage and generates a voltage.
Therefore,
VF = βVout,
Ve = Vin − VF ,
Vout = A0(Vin − βVout)

Figure 5.26 Voltage-Voltage Feedback

The overall gain has dropped by (1+βA0).

Voltage-voltage feedback modifies the input and output impedances
1. Output Impedance
Rout represents the output impedance of the feedforward amplifier and output resistance with
feedback is determined by setting the input to zero and applying a voltage at the output as
shown in Figure 5.27

Figure 5.27 Calculation of output resistance of a Voltage-Voltage Feedback

VF = βVX
Ve = −βVX
VM = −β A0VX
IX = [VX − (−βA0VX )]/Rout

Thus, the output impedance is lowered by (1 + βA0).

12
2. Input Impedance
The input impedance of the feedforward amplifier sustains the entire input voltage in Figure
5.28(a), but only a fraction of Vin in Fig. 5.28(b). As a result, the current drawn by Rin in the
feedback topology is less than that in the open-loop system, suggesting that returning a
voltage quantity to the input increases the input impedance.

Figure 5.28 Effect of voltage-voltage feedback on input resistance

Input resistance with feedback is determined by setting applying a voltage at the the input and
disconnecting RL as shown in Figure 5.29

Figure 5.29 Calculation of input impedance of a voltage-voltage feedback circuit.

Ve = IX Rin
VF = βA0 IX Rin
Ve = VX − VF = VX − βA0 IX Rin
IX Rin = VX − βA0 IX Rin

The input impedance therefore increases by the ubiquitous factor (1+βA0), bringing the
circuit closer to an ideal voltage amplifier.

5.6.2 Current-Voltage Feedback

Consider the general current-voltage feedback system illustrated in Figure 5.30 also called
“series-series” feedback. Since the feedback network senses the output current and returns a
voltage, its feedback factor, β, has the dimension of resistance and is denoted by RF . It is
important to note that a Gm stage must be loaded by a finite impedance ZL, to ensure that it
can deliver its output current.

13
 Figure 5.30 Current-voltage feedback.
If ZL = ∞, then an ideal Gm stage would sustain an infinite output voltage.
VF = RF Iout,
Ve = Vin − RF Iout
Iout = Gm(Vin − RF Iout)

Hence the gain is decreased by a factor (1+GmRF) and GmRF is the loop gain. An ideal
feedback network in this case exhibits zero input and output impedances.

Input and Output Impedance

1. Output Resistance
Consider the current-voltage feedback topology shown in Figure 5.31, where Rout represents
the finite output impedance of the feedforward amplifier. Note that Rout is placed in parallel
with the output because the ideal transimpedance amplifier is modeled by a voltage
dependent current source.

Figure 5.31 Calculation of output resistance of a current-voltage feedback amplifier.

VF = RFIX ; The feedback network produces a voltage VF proportional to IX

The current generated by Gm equals −RFIXGm.
−RF IXGm = IX – (VX /Rout)

The output impedance therefore increases by a factor of (1 + GmRF) .

14
2. Input Resistance
The current-voltage feedback increases the input impedance by a factor equal to one plus the
loop gain.
As illustrated in Figure 5.32, we have
IX RinGm = Iout.
Ve = VX − Gm RF IX Rin

Figure 5.32 Calculation of input resistance of a

current-voltage feedback

5.6.3 Voltage-Current Feedback amplifier.

In this type of feedback, the output voltage is sensed and a proportional current is returned to
the summing point at the input. This topology is also called “shunt-shunt” feedback. the
feedforward path incorporates a transimpedance amplifier with gain R0 and the feedback
factor has a dimension of conductance.
A voltage-current feedback topology is shown in Figure 5.33. Sensing a voltage and
producing a current, the feedback network is characterized by a transconductance gmF, ideally
exhibiting infinite input and output impedances.

Figure 5.33 Voltage-current feedback

IF = gmFVout
Ie = Iin – IF
Vout = R0Ie = R0(Iin − gmFVout)

This type of feedback lowers the transimpedance by a factor equal to one plus the loop gain.

Input and Output Impedance

1. Input Resistance
As shown in Figure 5.34, the input resistance of R0 appears in series with its input port.

IF = IX – (VX /Rin)

15
(VX /Rin)R0gmF = IF

That voltage current feedback decreases

the input impedances by a factor equal
to one plus the loop gain.

2. Output Resistance
The current source is reduced to zero and a voltage sourse Vx connected to output node as
shown in Figure. 5.35

Figure 5.35 Calculation of output resistance of a

voltage-current feedback amplifier.

IF = VX gmF,
Ie = −IF ,
VM = −R0 gmFVX .
Neglecting the input current of the feedback network, we write
IX = (VX − VM)/Rout
IX = (VX + gmF R0VX )/Rout.

That voltage current feedback decreases the output impedances by a factor equal to one plus
the loop gain.

16