CHAPTER 6

Power supplies
Chapter summary
This chapter deals with the unsung hero of most electronic systems,
the power supply. Nearly all electronic circuits require a source of well-
regulated d.c. at voltages of typically between 5 V and 30 V. In some cases
this supply can be derived directly from batteries (e.g. 6 V, 9 V, 12 V) but
in many others it is desirable to make use of a standard a.c. mains outlet.
This chapter explains how rectifier and smoothing circuits operate and
how power supply output voltages can be closely regulated. The chapter
concludes with a brief description of some practical power supply circuits.

Electronic Circuits: Fundamentals and applications. 978-0-367-42199-1. © Mike Tooley.

Published by Taylor & Francis. All rights reserved.
 6 Power supplies

The block diagram of a d.c. power supply is

shown in Fig. 6.1. Since the mains input is at a
relatively high voltage, a step-down transformer of
appropriate turns ratio is used to convert this to a
low voltage. The a.c. output from the transformer
secondary is then rectified using conventional
silicon rectifier diodes (see Chapter 5) to produce
an unsmoothed (sometimes referred to as
pulsating d.c.) output. This is then smoothed and
filtered before being applied to a circuit which will
regulate (or stabilize) the output voltage so that
it remains relatively constant in spite of variations
in both load current and incoming mains voltage. Figure 6.3 A simple d.c. power supply
Fig. 6.2 shows how some of the electronic
components that we have already met can be
used in the realization of the block diagram in
Fig. 6.1. The iron-cored step-down transformer Rectifiers
feeds a rectifier arrangement (often based on
a bridge circuit). The output of the rectifier is Semiconductor diodes (see Chapter 5) are
then applied to a high-value reservoir capacitor. commonly used to convert alternating current
This capacitor stores a considerable amount of (a.c.) to direct current (d.c), in which case they
charge and is being constantly topped-up by the are referred to as rectifiers. The simplest form
rectifier arrangement. The capacitor also helps to of rectifier circuit makes use of a single diode
smooth out the voltage pulses produced by the and, since it operates on only either positive or
rectifier. Finally, a stabilizing circuit (often based negative half-cycles of the supply, it is known as a
on a series transistor regulator and a zener half-wave rectifier.
diode voltage reference) provides a constant Fig. 6.4 shows a simple half-wave rectifier circuit.
output voltage. We shall now examine each stage Mains voltage (220 to 240 V) is applied to the
of this arrangement in turn, building up to some primary of a step-down transformer (T1). The
complete power supply circuits at the end of the secondary of T1 steps down the 240 V r.m.s.
chapter. to 12 V r.m.s. (the turns ratio of T1 will thus

Figure 6.1 Block diagram of a d.c. power supply

Figure 6.2 Block diagram of a d.c. power supply showing principal components
118
 6 Power supplies

voltage normally associated with silicon diodes.

However, during the negative half-cycle the peak
a.c. voltage will be dropped across D1 when it is
reverse biased. This is an important consideration
when selecting a diode for a particular application.
Assuming that the secondary of T1 provides
12 V r.m.s., the peak voltage output from the
Figure 6.4 A simple half-wave rectifier circuit transformer’s secondary winding will be given by:
Vpk = 1.414 × Vr.m.s. = 1.414 × 12 V = 16.97 V
be 240/12 or 20:1). Diode D1 will only allow The peak voltage applied to D1 will thus be
the current to flow in the direction shown (i.e. approximately 17 V. The negative half-cycles are
from cathode to anode). D1 will be forward blocked by D1 and thus only the positive half-
biased during each positive half-cycle (relative to cycles appear across RL. Note, however, that the
common) and will effectively behave like a closed actual peak voltage across RL will be the 17 V
switch. When the circuit current tries to flow in positive peak being supplied from the secondary
the opposite direction, the voltage bias across the on T1, minus the 0.7 V forward threshold voltage
diode will be reversed, causing the diode to act dropped by D1. In other words, positive half-cycle
like an open switch (see Figs 6.5(a) and 6.5(b), pulses having a peak amplitude of 16.3 V will
respectively). appear across RL.
The switching action of D1 results in a pulsating
output voltage which is developed across the load Example 6.1
resistor (RL). Since the mains supply is at 50 Hz, A mains transformer having a turns ratio of 44:1
the pulses of voltage developed across RL will is connected to a 220 V r.m.s. mains supply. If
also be at 50 Hz even if only half the a.c. cycle is the secondary output is applied to a half-wave
present. During the positive half-cycle, the diode rectifier, determine the peak voltage that will
will drop the 0.6 V to 0.7 V forward threshold appear across a load.

Solution
The r.m.s. secondary voltage will be given by:
VS = VP / 44 = 220 / 44 = 5 V

The peak voltage developed after rectification will

be given by:
VPK = 1.414 × 5 V = 7.07 V

Assuming that the diode is a silicon device with

a forward voltage drop of 0.6 V, the actual peak
voltage dropped across the load will be:
VL = 7.07 V − 0.6 V = 6.47 V

Reservoir and smoothing circuits

Figure 6.5 (a) Half-wave rectifier circuit with Fig. 6.6 shows a considerable improvement to the
D1 conducting (positive-going half-cycles of circuit of Fig. 6.4. The capacitor, C 1, has been
secondary voltage); (b) half-wave rectifier with added to ensure that the output voltage remains
D1 not conducting (negative-going half-cycles of at, or near, the peak voltage even when the diode
secondary voltage) is not conducting. When the primary voltage is

119
 6 Power supplies

Figure 6.6 A simple half-wave rectifier circuit

with reservoir capacitor

Figure 6.7 A simple half-wave rectifier circuit

first applied to T1, the first positive half-cycle with reservoir capacitor
output from the secondary will charge C1 to the
peak value seen across RL. Hence C1 charges
to 16.3 V at the peak of the positive half-cycle. Fig. 6.7 shows the secondary voltage waveform
Because C1 and RL are in parallel, the voltage together with the voltage developed across RL
across RL will be the same as that across C1. with and without C1 present. This gives rise to a
small variation in the d.c. output voltage (known
The time required for C1 to charge to the
as ripple). Since ripple is undesirable we must
maximum (peak) level is determined by the
take additional precautions to reduce it. One
charging circuit time constant (the series
obvious method of reducing the amplitude of the
resistance multiplied by the capacitance value). In
ripple is that of simply increasing the discharge
this circuit, the series resistance comprises the
time constant. This can be achieved either by
secondary winding resistance together with the
increasing the value of C1 or by increasing the
forward resistance of the diode and the (minimal)
resistance value of RL. In practice, however, the
resistance of the wiring and connections. Hence
latter is not really an option because RL is the
C1 charges very rapidly as soon as D1 starts
effective resistance of the circuit being supplied
to conduct.
and we don’t usually have the ability to change
The time required for C1 to discharge is, in it! Increasing the value of C1 is a more practical
contrast, very much greater. The discharge time alternative and very large capacitor values (often
constant is determined by the capacitance value in excess of 4,700 μF) are typical.
and the load resistance, RL. In practice, RL is very
Fig. 6.8 shows a further refinement of the simple
much larger than the resistance of the secondary
power supply circuit. This circuit employs two
circuit and hence C1 takes an appreciable time
additional components, R1 and C1, which act
to discharge. During this time, D1 will be reverse
as a filter to remove the ripple. The value of
biased and will thus be held in its non-conducting
C1 is chosen so that the component exhibits a
state. As a consequence, the only discharge path
negligible reactance at the ripple frequency (50 Hz
for C1 is through RL.
for a half-wave rectifier or 100 Hz for a full-wave
C1 is referred to as a reservoir capacitor.
It stores charge during the positive half-
cycles of secondary voltage and releases it
during the negative half-cycles. The circuit of
Fig. 6.6 is thus able to maintain a reasonably
constant output voltage across RL. Even so,
C1 will discharge by a small amount during the
negative half-cycle periods from the transformer
secondary.
Figure 6.8 Half-wave rectifier circuit with R–C
smoothing filter
120

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 120 11/10/2019 11:48

 6 Power supplies

rectifier – see later). In effect, R1 and C1 act like a

potential divider. The amount of ripple is reduced
by an approximate factor equal to:
XC
R + X C2
2

Example 6.2 Figure 6.9 Half-wave rectifier circuit with L–C

The R–C smoothing filter in a 50 Hz mains smoothing filter
operated half-wave rectifier circuit consists of
R1 = 100 Ω and C 2 = 1,000 μF. If 1 V of ripple
Example 6.3
appears at the input of the circuit, determine the
amount of ripple appearing at the output. The L–C smoothing filter in a 50 Hz mains
operated half-wave rectifier circuit consists of
Solution L1 = 10 H and C2 = 1,000 μF. If 1 V of ripple
First we must determine the reactance of the appears at the input of the circuit, determine the
capacitor, C1, at the ripple frequency (50 Hz): amount of ripple appearing at the output.
1 1
XC = = Solution
2π fC 6.28 × 50 × 1,000 × 10−6
Once again, the reactance of the capacitor, C1, is
1,000
= = 3.18 Ω 3.18 Ω (see Example 6.2). The reactance of L1 at
314
50 Hz can be calculated from:
The amount of ripple at the output of the circuit
XL = 2πfL = 2 × 3.14 × 50 × 10 = 3,140 Ω
(i.e. appearing across C1) will be given by:
The amount of ripple at the output of the circuit
XC 3.18
Vripple = 1× = 1× (i.e. appearing across C1) will be approximately
R 2 + X C2 1002 + 3.182 given by:
From which: XC 3.18
V = 1× = 1× ≈ 0.001 V
V = 0.032 V = 32 mV XC + XL 3140 + 3.18
Hence the ripple produced by this arrangement
(with 1 V of 50 Hz a.c. superimposed on the
Improved ripple filters rectified input) will be a mere 1 mV. It is worth
A further improvement can be achieved comparing this value with that obtained from the
by using an inductor, L1, instead of a resistor previous example!
in the smoothing circuit. This circuit also Finally, it is important to note that the amount of
offers the advantage that the minimum ripple present at the output of a power supply will
d.c. voltage is dropped across the inductor increase when the supply is loaded.
(in the circuit of Fig. 6.7, the d.c. output voltage
is reduced by an amount equal to the voltage
Full-wave rectifiers
drop across R1).
Fig. 6.9 shows the circuit of a half-wave power Unfortunately, the half-wave rectifier circuit is
supply with an L–C smoothing circuit. At the relatively inefficient as conduction takes place
ripple frequency, L1 exhibits a high value of only on alternate half-cycles. A better rectifier
inductive reactance while C1 exhibits a low value arrangement would make use of both positive and
of capacitive reactance. The combined effect is negative half-cycles. These full-wave rectifier
that of an attenuator which greatly reduces the circuits offer a considerable improvement over
amplitude of the ripple while having a negligible their half-wave counterparts. They are not only
effect on the direct voltage. more efficient but are significantly less demanding
in terms of the reservoir and smoothing
121
 6 Power supplies

components. There are two basic forms of full-

wave rectifier; the bi-phase type and the bridge
rectifier type.

Bi-phase rectifier circuits

Fig. 6.10 shows a simple bi-phase rectifier circuit.
Mains voltage (240 V) is applied to the primary of
the step-down transformer (T1) which has two
identical secondary windings, each providing 12 V
r.m.s. (the turns ratio of T1 will thus be 240/12 or
20:1 for each secondary winding).
On positive half-cycles, point A will be positive
with respect to point B. Similarly, point B will be
positive with respect to point C. In this condition
D1 will allow conduction (its anode will be positive
with respect to its cathode) while D2 will not
allow conduction (its anode will be negative with
respect to its cathode). Thus D1 alone conducts
on positive half-cycles.
On negative half-cycles, point C will be positive
with respect to point B. Similarly, point B will be
positive with respect to point A. In this condition
D2 will allow conduction (its anode will be positive
with respect to its cathode) while D1 will not
allow conduction (its anode will be negative with Figure 6.11 (a) Bi-phase rectifier with
respect to its cathode). Thus D2 alone conducts D1 conducting and D2 non-conducting
on negative half-cycles. (b) bi-phase rectifier with D2 conducting and D1
non-conducting
Fig. 6.11 shows the bi-phase rectifier circuit with
the diodes replaced by switches. In Fig. 6.11(a)
D1 is shown conducting on a positive half-cycle
while in Fig. 6.11(b) D2 is shown conducting. The Furthermore, this current is derived alternately
result is that current is routed through the load from the two secondary windings.
in the same direction on successive half-cycles. As with the half-wave rectifier, the switching
action of the two diodes results in a pulsating
output voltage being developed across the load
resistor (RL). However, unlike the half-wave circuit
the pulses of voltage developed across RL will
occur at a frequency of 100 Hz (not 50 Hz). This
doubling of the ripple frequency allows us to
use smaller values of reservoir and smoothing
capacitor to obtain the same degree of ripple
reduction (recall that the reactance of a capacitor
is reduced as frequency increases).
As before, the peak voltage produced by each
of the secondary windings will be approximately
17 V and the peak voltage across RL will be 16.3 V
Figure 6.10 Bi-phase rectifier circuit
122

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 122 11/10/2019 11:48

 6 Power supplies

(i.e. 17 V less the 0.7 V forward threshold voltage

dropped by the diodes).
Fig. 6.12 shows how a reservoir capacitor (C1)
can be added to ensure that the output voltage
remains at, or near, the peak voltage even when
the diodes are not conducting. This component
operates in exactly the same way as for the
half-wave circuit, i.e. it charges to approximately
16.3 V at the peak of the positive half-cycle and
holds the voltage at this level when the diodes
are in their non-conducting states. The time
required for C1 to charge to the maximum (peak) Figure 6.13 Waveforms for the bi-phase rectifier
level is determined by the charging circuit time
constant (the series resistance multiplied by Bridge rectifier circuits
the capacitance value). In this circuit, the series An alternative to the use of the bi-phase circuit
resistance comprises the secondary winding is that of using a four-diode bridge rectifier (see
resistance together with the forward resistance Fig. 6.14) in which opposite pairs of diodes
of the diode and the (minimal) resistance of conduct on alternate half-cycles. This arrangement
the wiring and connections. Hence C1 charges avoids the need to have two separate secondary
very rapidly as soon as either D1 or D2 starts to windings.
conduct. The time required for C1 to discharge is,
A full-wave bridge rectifier arrangement is shown
in contrast, very much greater. The discharge time
in Fig. 6.15. Mains voltage (240 V) is applied to
contrast is determined by the capacitance value
the primary of a step-down transformer (T1).
and the load resistance, RL. In practice, RL is very
much larger than the resistance of the secondary The secondary winding provides 12 V r.m.s.
circuit and hence C1 takes an appreciable time (approximately 17 V peak) and has a turns ratio
to discharge. During this time, D1 and D2 will be of 20:1, as before. On positive half-cycles, point
reverse biased and held in a non-conducting state. A will be positive with respect to point B. In this
As a consequence, the only discharge path for C1 condition D1 and D2 will allow conduction while
is through RL. Fig. 6.13 shows voltage waveforms D3 and D4 will not allow conduction. Conversely,
for the bi-phase rectifier, with and without C1 on negative half-cycles, point B will be positive
present. Note that the ripple frequency (100 Hz) with respect to point A. In this condition D3 and
is twice that of the half-wave circuit shown D4 will allow conduction while D1 and D2 will not
previously in Fig. 6.7. allow conduction.
Fig. 6.16 shows the bridge rectifier circuit
with the diodes replaced by four switches. In
Fig. 6.16(a) D1 and D2 are conducting on a
positive half-cycle while in Fig. 6.16(b) D3 and

Figure 6.12 Bi-phase rectifier with reservoir

capacitor Figure 6.14 Four diodes connected as a bridge
123
 6 Power supplies

Figure 6.15 Full-wave bridge rectifier circuit Figure 6.17 Bridge rectifier with reservoir
capacitor

Figure 6.18 Waveforms for the bridge rectifier

holds the voltage at this level when the diodes are

in their non-conducting states. This component
operates in exactly the same way as for the
Figure 6.16 (a) Bridge rectifier with D1 and D2 bi-phase circuit and the secondary and rectified
conducting, D3 and D4 non-conducting (b) bridge output waveforms are shown in Fig. 6.18. Once
rectifier with D1 and D2 non-conducting, D3 and again note that the ripple frequency is twice that
D4 conducting of the incoming a.c. supply.
Finally, R–C and L–C ripple filters can be added
to bi-phase and bridge rectifier circuits in exactly
D4 are conducting. Once again, the result is that
the same way as those shown for the half-wave
current is routed through the load in the same
rectifier arrangement (see Figs 6.8 and 6.9).
direction on successive half-cycles. As with the
bi-phase rectifier, the switching action of the two
diodes results in a pulsating output voltage being Voltage regulators
developed across the load resistor (RL). Once
A simple voltage regulator is shown in Fig. 6.19.
again, the peak output voltage is approximately
RS is included to limit the zener current to a safe
16.3 V (i.e. 17 V less the 0.7 V forward threshold
value when the load is disconnected. When a
voltage).
load (RL) is connected, the zener current (IZ) will
Fig. 6.17 shows how a reservoir capacitor (C1) fall as current is diverted into the load resistance
can be added to maintain the output voltage when (it is usual to allow a minimum current of 2 mA to
the diodes are not conducting. This component 5 mA in order to ensure that the diode regulates).
operates in exactly the same way as for the The output voltage (VZ) will remain at the zener
bi-phase circuit, i.e. it charges to approximately voltage until regulation fails at the point at
16.3 V at the peak of the positive half-cycle and which the potential divider formed by RS and RL
124
 6 Power supplies

⎛V ⎞
RS max. = RL × ⎜⎜ IN − 1⎟⎟
⎝VIN ⎠
thus:
⎛9 ⎞
RS max. = 400 × ⎜ − 1⎟ = 400 × (1.8 − 1) = 320 Ω
⎝5 ⎠
Now we need to determine the minimum value
Figure 6.19 A simple shunt zener voltage
for the series resistor, RS:
regulator
VINVZ − VZ 2
produces a lower output voltage that is less than RS min. =
PZ max.
VZ. The ratio of RS to RL is thus important. At the
thus:
point at which the circuit just begins to fail to
regulate: ( 9 × 5 ) − 52 45 − 25
RS min. = = = 40 Ω
RL 0.5 0.5
VZ = VIN × Hence a suitable value for RS would be 150 Ω
RL + RS
(roughly mid-way between the two extremes).
where VIN is the unregulated input voltage. Thus
the maximum value for RS can be calculated from:
⎛V ⎞
Output resistance and voltage
RS max. = RL × ⎜⎜ IN − 1⎟⎟ regulation
⎝VIN ⎠
In a perfect power supply, the output voltage
The power dissipated in the zener diode will be
would remain constant regardless of the current
given by PZ = IZ × VZ, hence the minimum value for
taken by the load. In practice, however, the
RS can be determined from the off-load condition
output voltage falls as the load current increases.
when:
To account for this fact, we say that the power
RS min. =
VIN − VZ
=
VIN − VZ (V −V ) ×VZ
= IN Z supply has internal resistance (ideally this should
IZ ⎛ P max. ⎞ PZ max. be zero). This internal resistance appears at the
⎜⎜ Z ⎟⎟
⎝ VZ ⎠ output of the supply and is defined as the change
Thus: in output voltage divided by the corresponding
change in output current. Hence:
VINVZ − VZ 2
RS min. = change in output voltage DVout
PZ max. Rout = =
change in output current DIout
where PZ max. is the maximum rated power
where DIout represents a small change in
dissipation for the zener diode.
output (load) current and DVout represents a
Example 6.4 corresponding small change in output voltage.
A 5 V zener diode has a maximum rated power The regulation of a power supply is given by the
dissipation of 500 mW. If the diode is to be used relationship:
in a simple regulator circuit to supply a regulated change in output voltage
Regulation = × 100%
5 V to a load having a resistance of 400 Ω, change in line (input) voltage
determine a suitable value of series resistor for Ideally, the value of regulation should be very
operation in conjunction with a supply of 9 V. small. Simple shunt zener diode regulators
of the type shown in Fig. 6.19 are capable of
Solution
producing values of regulation of 5% to 10%.
We shall use an arrangement similar to that More sophisticated circuits based on discrete
shown in Fig. 6.19. First we should determine the components produce values of between 1% and
maximum value for the series resistor, RS: 5% and integrated circuit regulators often provide
values of 1% or less.

125

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 125 11/10/2019 11:48

 6 Power supplies

Example 6.5
The following data were obtained during a test
carried out on a d.c. power supply:
(i) Load test
Output voltage (no-load) = 12 V
Output voltage (2 A load current) = 11.5 V Figure 6.20 Simple d.c. power supply with shunt
zener regulated output
(ii) Regulation test
Output voltage (mains input, 220 V) = 12 V
Output voltage (mains input, 200 V) = 11.9 V
Determine (a) the equivalent output resistance
of the power supply and (b) the regulation of the
power supply.

Solution Figure 6.21 Improved regulated d.c. power

The output resistance can be determined from supply with series-pass transistor
the load test data:
change in output voltage 12 − 11.5
Rout = = = 0.25 Ω
change in output current 2− 0
The regulation can be determined from the
regulation test data:
change in output voltage
Regulation = × 100%
change in line (input) voltage
Figure 6.22 Variable d.c. power supply
thus
12 − 1.9 0.1
Regulation = × 100% = × 100% = 0.5%
220 − 200 20

Practical power supply circuits

Fig. 6.20 shows a simple power supply circuit
capable of delivering an output current of up to
250 mA. The circuit uses a full-wave bridge Figure 6.23 Power supply with three-terminal IC
rectifier arrangement (D1 to D4) and a simple voltage regulator
C–R filter. The output voltage is regulated by the
shunt-connected 12 V zener diode. a small heatsink to conduct away any heat
Fig. 6.21 shows an improved power supply in produced). Fig. 6.22 shows a variable power
which a transistor is used to provide current gain supply. The base voltage to the series-pass
and minimize the power dissipated in the zener transistor is derived from a potentiometer
diode (TR1 is sometimes referred to as a series- connected across the zener diode, D5. Hence
pass transistor). The zener diode, D5, is rated at the base voltage is variable from 0 V to 13 V. The
13 V and the output voltage will be approximately transistor requires a substantial heatsink (note
0.7 V less than this (i.e. 13 V minus the base- that TR1’s dissipation increases as the output
emitter voltage drop associated with TR1). Hence voltage is reduced).
the output voltage is about 12.3 V. The circuit is Finally, Fig. 6.23 shows a d.c. power supply based
capable of delivering an output current of up to on a fixed-voltage three-terminal integrated
500 mA (note that TR1 should be fitted with
126

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 126 11/10/2019 11:48

 6 Power supplies

Figure 6.25 A voltage doubler

Figure 6.24 This four-diode bridge rectifier

arrangement is part of a high-voltage d.c. supply.
Each BY253 diode is rated for a reverse repetitive
maximum voltage (VRRM) of 600 V, and a maximum
forward current (IF max.) of 3 A

circuit voltage regulator. These devices are

available in standard voltage and current ratings
(e.g. 5 V, 12 V, 15 V at 1 A, 2 A and 5 A) and they
provide excellent performance in terms of output Figure 6.26 A voltage tripler
resistance, ripple rejection and voltage regulation.
In addition, such devices usually incorporate
arrangement shown in Fig. 6.26. Here C1 charges
overcurrent protection and can withstand a direct
to the positive peak secondary voltage, while
short-circuit placed across their output terminals.
C2 and C3 charge to twice the positive peak
This is an essential feature in many practical
secondary voltage. The result is that the output
applications!
voltage is the sum of the voltages across C1
and C3 which is three times the voltage that
Voltage multipliers would be produced by a single diode. The ladder
arrangement shown in Fig. 6.25 can be easily
By adding a second diode and capacitor, we
extended to provide even higher voltages but
can increase the output of the simple half-wave
the efficiency of the circuit becomes increasingly
rectifier that we met earlier. A voltage doubler
impaired and high-order voltage multipliers of
using this technique is shown in Fig. 6.25. In this
this type are only suitable for providing relatively
arrangement C1 will charge to the positive peak
small currents.
secondary voltage while C2 will charge to the
negative peak secondary voltage. Since the output
is taken from C1 and C2 connected in series the
Switched mode power supplies
resulting output voltage is twice that produced by
one diode alone. Power supplies can be divided into two principal
The voltage doubler can be extended to categories, linear and non-linear types. Linear
produce higher voltages using the cascade power supplies make use of conventional
127

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 127 11/10/2019 11:48

 CHAPTER 7
Amplifiers
Chapter summary
This chapter introduces the basic concepts of amplifiers and amplification.
It describes the most common types of amplifier and outlines the basic
classes of operation used in both linear and non-linear amplifiers. The
chapter also describes methods for predicting the performance of an
amplifier based on equivalent circuits and on the use of semiconductor
characteristics and load lines. Once again, we conclude with a selection of
practical circuits that can be built and tested.

Electronic Circuits: Fundamentals and applications. 978-0-367-42199-1. © Mike Tooley.

Published by Taylor & Francis. All rights reserved.
 7 Amplifiers

Types of amplifier
Many different types of amplifier are found
in electronic circuits. Before we explain the
operation of transistor amplifiers in detail, we shall
briefly describe the main types of amplifier.

a.c. coupled amplifiers

In a.c. coupled amplifiers, stages are coupled
together in such a way that d.c. levels are isolated
and only the a.c. components of a signal are
transferred from stage to stage.

d.c. coupled amplifiers

In d.c. (or direct) coupled amplifiers, stages are
coupled together in such a way that stages are not
isolated to d.c. potentials. Both a.c. and d.c. signal
components are transferred from stage to stage.
Figure 7.1 Part of a high-gain, wideband d.c.
Large-signal amplifiers
coupled amplifier using discrete components
Large-signal amplifiers are designed to cater for
appreciable voltage and/or current levels (typically
from 1 V to 100 V or more). Low-noise amplifiers
Low-noise amplifiers are designed so that they
Small-signal amplifiers
contribute negligible noise (signal disturbance) to
Small-signal amplifiers are designed to cater for the signal being amplified. These amplifiers are
low-level signals (normally less than 1 V and often usually designed for use with very small signal
much smaller). Small-signal amplifiers have to be levels (usually less than 10 mV or so).
specially designed to combat the effects of noise.

Audio frequency amplifiers Gain

Audio frequency amplifiers operate in the band of One of the most important parameters of an
frequencies that is normally associated with audio amplifier is the amount of amplification or gain
signals (e.g. 20 Hz to 20 kHz). that it provides. Gain is simply the ratio of output
voltage to input voltage, output current to input
Wideband amplifiers
current, or output power to input power (see
Wideband amplifiers are capable of amplifying a Fig. 7.2). These three ratios give, respectively,
very wide range of frequencies, typically from a the voltage gain, current gain and power gain.
few tens of hertz to several megahertz. Thus:
Radio frequency amplifiers Vout
Voltage gain, Av =
Radio frequency amplifiers operate in the band of Vin
frequencies that is normally associated with radio
Iout
signals (e.g. from 100 kHz to over 1 GHz). Note Current gain, Ai =
Iin
that it is desirable for amplifiers of this type to
be frequency selective and thus their frequency Pout
response may be restricted to a relatively narrow Power gain, Ap =
Pin
band of frequencies (see Fig. 7.9 on page 144).

140
 7 Amplifiers

be affected by a number of factors including

the amount of bias applied (see later) and the
amplitude of the input signal.
It is also worth noting that a linear amplifier will
become non-linear when the applied input signal
exceeds a threshold value. Beyond this value the
Figure 7.2 Block diagram for an amplifier amplifier is said to be overdriven and the output
showing input and output voltages and currents will become increasingly distorted if the input
signal is further increased.
Note that, since power is the product of current Amplifiers are usually designed to be operated
and voltage (P = I V ), we can infer that: with a particular value of bias supplied to the
active devices (i.e. transistors). For linear
Pout Iout ×Vout Iout Vout
Ap = = = × = Ai × Av operation, the active device(s) must be operated
Pin Iin ×Vin Iin Vin in the linear part of their transfer characteristic
(Vout plotted against Vin). In Fig. 7.3 the input and
Example 7.1 output signals for an amplifier are operating in
An amplifier produces an output voltage of 2 V for linear mode. This form of operation is known
an input of 50 mV. If the input and output currents as Class A and the bias point is adjusted to
in this condition are, respectively, 4 mA and the mid-point of the linear part of the transfer
200 mA, determine: characteristic. Furthermore, current will flow in
(a) the voltage gain; the active devices used in a Class A amplifier
(b) the current gain; during a complete cycle of the signal waveform.
(c) the power gain. At no time does the current fall to zero.
Fig. 7.4 shows the effect of moving the bias point
Solution down the transfer characteristic and, at the same
(a) The voltage gain is calculated from: time, increasing the amplitude of the input signal.
From this, you should notice that the extreme
Vout 2V
Av = = = 40 negative portion of the output signal has become
Vin 50 mV distorted. This effect arises from the non-linearity
(b) The current gain is calculated from: of the transfer characteristic that occurs near the
Iout 200 mA origin (i.e. the zero point). Despite the obvious
Ai = = = 50
Iin 4 mA
(c) The power gain is calculated from:
Iout ×Vout 200 mA × 2 V 0.4 W
Ap = = = = 2,000
Iin ×Vin 4 mA × 50 mV 200 μW
Note that the same result is obtained from:
Ap = Ai × Av = 50 × 40 = 2,000

Class of operation
An important requirement of most amplifiers is
that the output signal should be a faithful copy
of the input signal, albeit somewhat larger in
amplitude. Other types of amplifier are non-linear,
in which case their input and output waveforms
will not necessarily be similar. In practice, the Figure 7.3 Class A (linear) operation
degree of linearity provided by an amplifier can
141
 7 Amplifiers

Figure 7.6 Class B operation (no bias applied)

The output signal will only comprise a series of

Figure 7.4 Effect of reducing bias and increasing
positive-going half-cycles and the active device(s)
input signal amplitude (the output waveform is no
will only be conducting during half-cycles of the
longer a faithful reproduction of the input)
waveform (i.e. they will only be operating 50% of
the time).
non-linearity in the output waveform, the active
This mode of operation is known as Class B and
device(s) will conduct current during a complete
is commonly used in high-efficiency push–pull
cycle of the signal waveform.
power amplifiers where the two active devices in
Now consider the case of reducing the bias the output stage operate on alternate half-cycles
even further while further increasing the of the waveform.
amplitude of the input signal (see Fig. 7.5). Here
Finally, there is one more class of operation to
the bias point has been set at the projected
consider. The input and output waveforms for
cut-off point. The negative-going portion of the
Class C operation are shown in Fig. 7.7. Here,
output signal becomes cut-off (or clipped) and the
the bias point is set at beyond the cut-off (zero)
active device(s) will cease to conduct for this part
point and a very large input signal is applied. The
of the cycle. This mode of operation is known as
output waveform will then comprise a series of
Class AB.
quite sharp positive-going pulses. These pulses
Now let’s consider what will happen if no bias of current or voltage can be applied to a tuned
at all is applied to the amplifier (see Fig. 7.6). circuit load in order to recreate a sinusoidal signal.
In effect, the pulses will excite the tuned circuit

Figure 7.5 Class AB operation (bias set at Figure 7.7 Class C operation (bias is set beyond
projected cut-off) cut-off)
142

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 142 11/10/2019 11:48

 7 Amplifiers

Table 7.1 Classes of operation

Class of Bias point Conduction angle Efficiency Application

operation (typical) (typical)
A Mid-point 360° 5% to 20% Linear audio amplifiers
AB Projected cut-off 210° 20% to 40% Push–pull audio amplifiers
B At cut-off 180° 40% to 70% Push–pull audio amplifiers
C Beyond cut-off 120° 70% to 90% Radio frequency power amplifiers

and its inherent flywheel action will produce

a sinusoidal output waveform. This mode of
operation is only used in RF power amplifiers that
must operate at very high levels of efficiency.
Table 7.1 summarizes the classes of operation
used in amplifiers.

Input and output resistance

Figure 7.8 Input and output resistances ‘seen’
Input resistance is the ratio of input voltage to
looking into the input and output terminals,
input current and it is expressed in ohms. The
respectively
input of an amplifier is normally purely resistive
(i.e. any reactive component is negligible) in the
middle of its working frequency range (i.e. the this chapter. Finally, it’s important to note that,
mid-band). In some cases, the reactance of the although these resistances are meaningful in
input may become appreciable (e.g. if a large terms of the signals present, they cannot be
value of stray capacitance appears in parallel with measured using a conventional meter!
the input resistance). In such cases we would
refer to input impedance rather than input
resistance. Frequency response
Output resistance is the ratio of open-circuit The frequency response characteristics for
output voltage to short-circuit output current and various types of amplifier are shown in Fig. 7.9.
is measured in ohms. Note that this resistance Note that, for response curves of this type,
is internal to the amplifier and should not be frequency is almost invariably plotted on a
confused with the resistance of a load connected logarithmic scale.
externally. The frequency response of an amplifier is usually
As with input resistance, the output of an specified in terms of the upper and lower cut-off
amplifier is normally purely resistive and we frequencies of the amplifier. These frequencies
can safely ignore any reactive component. If this are those at which the output power has dropped
is not the case, we would once again need to to 50% (otherwise known as the −3 dB points) or
refer to output impedance rather than output where the voltage gain has dropped to 70.7% of
resistance. its mid-band value.
Fig. 7.8 shows how the input and output Figs 7.10 and 7.11, respectively, show how the
resistances are ‘seen’ looking into the input bandwidth can be expressed in terms of either
and output terminals, respectively. We shall be power or voltage (the cut-off frequencies, f1 and
returning to this equivalent circuit a little later in f2, and bandwidth are identical).

143
 7 Amplifiers

Figure 7.9 Frequency response and bandwidth (output power plotted against frequency)

Solution
The mid-band voltage gain corresponds with the
flat part of the frequency response characteristic.
At that point the voltage gain reaches a maximum
of 35 (see Fig. 7.12).
The voltage gain at the two cut-off frequencies
can be calculated from:
Av cut-off = 0.707 × Av max = 0.707 × 35 = 24.7
This value of gain intercepts the frequency
response graph at f1 = 57 Hz and f2 = 590 kHz (see
Figure 7.10 Frequency response and bandwidth Fig. 7.12).
(output power plotted against frequency)

Bandwidth
The bandwidth of an amplifier is usually taken
as the difference between the upper and
lower cut-off frequencies (i.e. f2 − f1 in Figs
7.10 and 7.11). The bandwidth of an amplifier
must be sufficient to accommodate the range
of frequencies present within the signals
that it is to be presented with. Many signals
contain harmonic components (i.e. signals at
2f, 3f, 4f, etc. where f is the frequency of the
fundamental signal). To reproduce a square
wave, for example, requires an amplifier with a
Figure 7.11 Frequency response and bandwidth very wide bandwidth (note that a square wave
(output voltage plotted against frequency) comprises an infinite series of harmonics). Clearly
it is not possible to perfectly reproduce such a
wave, but it does explain why it can be desirable
Example 7.2 for an amplifier’s bandwidth to greatly exceed
Determine the mid-band voltage gain and upper the highest signal frequency that it is required
and lower cut-off frequencies for the amplifier to handle!
whose frequency response is shown in Fig. 7.12.

144
 7 Amplifiers

Figure 7.12 See Example 7.2

Phase shift feedback determines the overall (or closed-loop)

gain. Because this form of feedback has the effect
Phase shift is the phase angle between the input of reducing the overall gain of the circuit, this form
and output signal voltages measured in degrees. of feedback is known as negative feedback. An
The measurement is usually carried out in the alternative form of feedback, where the output
mid-band where, for most amplifiers, the phase is fed back in such a way as to reinforce the
shift remains relatively constant. Note also that input (rather than to subtract from it) is known as
conventional single-stage transistor amplifiers positive feedback. This form of feedback is used
provide phase shifts of either 180° or 360°. in oscillator circuits (see Chapter 9).
Fig. 7.13 shows the block diagram of an amplifier
Negative feedback stage with negative feedback applied. In this
circuit, the proportion of the output voltage fed
Many practical amplifiers use negative feedback
back to the input is given by β and the overall
in order to precisely control the gain, reduce
voltage gain will be given by:
distortion and improve bandwidth. The gain can be
reduced to a manageable value by feeding back Vout
Overall gain, G =
a small proportion of the output. The amount of Vin

Figure 7.13 Amplifier with negative feedback applied

145
 7 Amplifiers

Now Vin' = Vin − βVout (by applying Kirchhoff’s The overall voltage gain with negative feedback
Voltage Law) (note that the amplifier’s input will then be:
voltage has been reduced by applying negative Av 60 60
feedback) thus: G= = = = 7.14
1+ βAv 1+ (0.1× 60) 7
Vin =Vin' + βVout The increase in overall voltage gain, expressed as
and a percentage, will thus be:
Vout = Av × Vin (note that Av is the internal gain 8.57 − 8.33
× 100% = 2.88%
of the amplifier) 8.33
Hence: Note that this example illustrates one of the
Av × V 'in Av × V 'in important benefits of negative feedback in
Overall gain, G = = stabilizing the overall gain of an amplifier stage.
V 'in + βVout V 'in + β ( Av × V 'in )
Thus: Example 7.5
Av An integrated circuit that produces an open-
G=
1+ βAv loop gain of 100 is to be used as the basis of an
Hence, the overall gain with negative feedback amplifier stage having a precise voltage gain of
applied will be less than the gain without 20. Determine the amount of feedback required.
feedback. Furthermore, if Av is very large (as
Solution
is the case with an operational amplifier – see Av
Chapter 8) the overall gain with negative feedback Re-arranging the formula, G =
1+ βAv
applied will be given by: to make β the subject gives:
G = 1/β (when Av is very large) 1 1
β= −
Note, also, that the loop gain of a feedback G Av
amplifier is defined as the product of β and Av. Thus:
Example 7.3 1 1
β= − = 0.05 − 0.01 = 0.04
An amplifier with negative feedback applied has 20 100
an open-loop voltage gain of 50, and one-tenth
of its output is fed back to the input (i.e. β = 0.1). Transistor amplifiers
Determine the overall voltage gain with negative
feedback applied. Regardless of what type of transistor is employed,
three basic circuit configurations are used. These
Solution three circuit configurations depend upon which
With negative feedback applied the overall voltage one of the three transistor connections is made
gain will be given by: common to both the input and the output. In the
case of bipolar transistors, the configurations
Av 50 50
G= = = = 8.33 are known as common emitter, common
1+ βAv 1+ (0.1× 50) 6 collector (or emitter follower) and common
base. Where field effect transistors are used,
Example 7.4
the corresponding configurations are common
If, in Example 7.3, the amplifier’s open-loop source, common drain (or source follower) and
voltage gain increases by 20%, determine the common gate.
percentage increase in overall voltage gain.
The three basic circuit configurations (Figs 7.14
Solution to 7.19) exhibit quite different performance
The new value of voltage gain will be given by:
Av = Av + 0.2Av = 1.2 × 50 = 60

146
 7 Amplifiers

Figure 7.38 A practical common-emitter amplifier Figure 7.41 A practical emitter-follower stage
stage

Figure 7.39 An improved version of the common- Figure 7.42 An improved emitter-follower stage
emitter amplifier stage
Two practical emitter-follower circuits are shown
in Figs 7.41 and 7.42. These circuits offer a
Fig. 7.40 shows a practical common-emitter
voltage gain of unity (1) but are ideal for matching
amplifier with bias stabilization. This stage
a high-resistance source to a low-resistance load.
provides a gain of 150 to well over 200
It is important to note that the input resistance
(depending upon the current gain, hfe, of the
varies with the load connected to the output of
individual transistor used). The circuit will operate
the circuit (it is typically in the range 50 kΩ to
with supply voltages of between 6 V and 18 V.
150 kΩ). The input resistance can be calculated
by multiplying hfe by the effective resistance of R2
in parallel with the load connected to the output
terminals.
Fig. 7.42 is an improved version of Fig. 7.41
in which the base current is derived from the
potential divider formed by R1 and R2. Note,
however, that the input resistance is reduced
since R1 and R2 effectively appear in parallel with
the input. The input resistance of the stage is thus
typically in the region of 40 kΩ to 70 kΩ.

Multi-stage amplifiers
Figure 7.40 Operating point and quiescent values In order to provide sufficiently large values of
shown on the load line for a bipolar transistor gain, it is frequently necessary to use a number
operating in common-emitter mode of interconnected stages within an amplifier. The
156

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 156 11/10/2019 11:48

 7 Amplifiers

overall gain of an amplifier with several stages

(i.e. a multi-stage amplifier) is simply the product
of the individual voltage gains. Hence:
AV = AV1 × AV2 × AV3, etc.
Note, however, that the bandwidth of a multi-
stage amplifier will be less than the bandwidth of
each individual stage. In other words, an increase
in gain can only be achieved at the expense of a
reduction in bandwidth.
Signals can be coupled between the individual Figure 7.44 A typical two-stage high-gain R–C
stages of a multi-stage amplifier using one of a coupled common-emitter amplifier
number of different methods shown in Fig. 7.43.
The most commonly used method is that of
R–C coupling as shown in In Fig. 7.43(a). In this
coupling method, the stages are coupled together
using capacitors having a low reactance at the
signal frequency and resistors (which also provide
a means of connecting the supply). Fig. 7.44
shows a practical example of this coupling
method.
A similar coupling method, known as L–C
coupling, is shown in Fig. 7.43(b). In this method,
the inductors have a high reactance at the signal
frequency. This type of coupling is generally only
used in RF and high-frequency amplifiers.
Two further methods, transformer coupling and
direct coupling, are shown in Figs 7.43(c) and
7.43(d), respectively. The latter method is used
where d.c. levels present on signals must be
preserved.

Power amplifiers
The term ‘power amplifier’ can be applied to any
amplifer that is designed to deliver an appreciable
level of power. There are several important
considerations for amplifiers of this type, including
the ability to deliver current (as well as voltage)
to a load, and also the need to operate with
a reasonable degree of efficiency (recall that
conventional Class A amplifiers are inefficient).
In order to deliver sufficient current to the load,
power amplifiers must have a very low value of
output impedance. Thus the final stage (or output
stage) is usually based on a device operating
in emitter-follower configuration. In order to
Figure 7.43 Different methods used for operate at a reasonable level of efficiency,
interstage coupling the output stage must operate in Class AB or
157

9780367421991_TOOLEY_Electronic_Circuit_5e.indd 157 11/10/2019 11:48