Electronic Fault Diagnosis

Some other ELBS low-priced editions
Benson and ELECfRIC-CIRCUIT THEORY e(h'ard Amn/d

H<lrrison
Francis ELEcm.ICAL INSTALLATION WORK LO.lgI1WIl
Gosling A FIRST COURSE IN APPLIED ELECTRONICS Macmillan

Gregory AN INTRODUCTION TO ELECTRICAL Maclllillan
'INSTRUMENTATION
Havill and ELEMENTS OF ELECTRONICS FOR PHYSICAL Macmillan
Walton SCIENTISTS
Hughes ELECfRICAL TECHNOLOGY Longmall
Lewin LOGICAL DESIGN OF SWITCHING CIRCUITS' . Neison
Lewin THEORY AND DESIGN OF DlGrrAL Nei'iOl1
COMPUTER SYSTEMS
:.". Meadows TECHNICIAN ELECTRONICS 2 Cassell
McKenzie Smith BASIC ELECTRICAL ENGINEERING Longman
and Hosie . SCIENCE
Neidle ELECTRICAL INSTALLAnON TECHNOLOGY NelvlIes-Bwterwort
Paddock and ELECTRICAL INSTALLATION TECHNOLOGY Hodeler & Stought,
Galvin AND PRACfICE
Ramshaw f'OWER ELECTRONICS Chapman & Hall
Say ALTERNATING CURRENT MACHINES Pitmall
Say INTRODUCTION TO THE UNIFIED THEORY Pitman·
OF ELECTROMAGNETIC MA CHINES
Say and Taylor DmECT CURRENT MACHINES Pirman
Shepherd, Morton HIGHER ELECrRICAL ENGINEERING Pirman
and Spencc
~...
,
Williams AN INTRODUCTION TO ELECTRICAL Macmillan
CIRCUIT THEORY
Young and AUtOMOBILE ELECTRICAL AND ELECTRONIC Nell'nes-Bwterworl
Griffiths EQUIPMENT
Electronic
Fault
Diagnosis
George Loveday CEng, MIERE
lecturer in Electronic Engineering
I-Iavering Technical College.
ll·
,
THE ENGLISH LANGUAGE BOOK SOCIETY

AND
PITMAN
PITMAN BOOKS LIMITED
39 Parker Street, London WelB 5PB
© G. C. Loveday ]977
First published in Great Britain 1977

Reprinted 1979
Reprinted 1981
ELBS edition first published 1981
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmiHed, in any form or by any
means, electronic, mechanical, photocopying, recording and/or
otherwise without the prior written permission of the publishers.
This book may not be lent, resold, hired out or otherwise disposed of
by way of trade in any form of binding or cover other than thai in
which it is published, without the prior consent of the publishers.
Text set in 10112 pt IBM Press Roman,

printed and bound in Great Britain
at The Pitman Press, Bath
ISBN 0 273 01779 9

.
Contents
1 Basics of Fault Diagnosis 2

4.10 Exercise: audio power amplifier:·:SO::.-:
••;:... J'.
1.1 Circuits and test readings 2
J.2 Components and common faults 2

5 Oscillator and Time Base Cin:uiu"::.. 52'
13 Operating principles of common active
5.1 Principles of oscillators 52 .}:>."~:" ':
components 3
5.2 Measurement offrequency 54. i
1.4 . Measuring instruments and testing 5.3 Frequency stability 54 .' ... ';
methods 7 5.4 Harmonic distortion 5S <:'t;'
1.5 Fault rmding on electronic instruments and
5.5 Square and pulse wavero~nr:S
systems 12
5.6 Sawtooth and ramp circuits ;.S6
5.7 Negative resistance oscillatorS; ~,
~ Single Stage Transistor Amplifier 14
5.8 Fault findingonoscillatoIS .s8!::N:., ,,'
2.1 Basic principles
14 5.9 Exercise: Wien bridge osdJJ.aCot~ 59 0', ,',
2.2 Resistor faults
15 5.10 Exercise: blocking oscillator "wCoOth;~ ,
2.3 Capacitorfaults
17 generator 60 '::":;';;;~'f' .
2.4 Transistor faults

18 5.11 Exercise: free running sawtooth ~tor 62
5.12 Exercise: slow speed lamp gtn...'>Jalor ~,64

3 Power Supply Circuits 22
5.13 Exercise: gated UJT pulse generalor:~: 65
3.J.,v Basic principles of d.c. power supplies 22

~. t·;;~~·.·~~(:;::,~:
3':2 The linear stabilized power unit 22

6 Pulse and Waveform Sbaping Cf:tUits" 66
3.3 Switching mode power supplies 24

6.1 Introduction 66
3.4 Power supply protection circuits 25
6.2 Unear passive circuits-the iakgrator and.
3.5 Testing power supply circuits 26

differentiator 66 '.
3.6 Fault finding techniques and typical

6.3 Diode wavefonn sbapers 68
: I
fault conditions 27
6.4 Active pulse shaping ciIcuits
3.7 Exercise: power unit with a simple linear
6.5 Schmitt trigger circuit 71
regulator 28 .
6.6 The mOflostable 72
3.8 Exercise: stabilited power supply with
6.7 Fault finding in pulse and Wa~fonn
current limit circuit 30

sbaping circuits 13 ..:.
3.9 Exercise: switching mode power supply 32

6.8 Exercise: wavefofm shapingdmdt~, 74
6.9 Exercise: monoslable circuit .. 75.;' .

4 Amplifier Circuits 34
6.10 Exercise: Schmitt trigger dn::uit . 76
4.1 Types and classes of amplifier 34

6.11 Exercise: logic interface circuit 76·
4.2 Negative feedback 37
6.12 Exercise: FET input Schmin tritter
4.3 Testing amplifiers: basic measurements 39

circuit 78
4.4 Transient testing of amplifiers 40

.~ ~
4.5 Distortion measurements 41
(j) Thyristor and Triac Cin:uits . SO" ,"
4.6 Faults in amplifiers 43
7.1 Principle of operation of ~ thyristor 80
4.7 Exercise: two·stage pre·amplifier 44

7.2 Applications of the thyristor 81
4.8 Exercise: pre·amplifier with FET input 46

7.3 Basic operating prinCiple of the trUe. 83
4.9 Exercise: d.c. amplifier 48

7.4 Applications of the triac 84
7.5 Fault conditions and fault finding in 8.2 Types of integrated circuit 93
thyristor and triac circuits 84 83 Servicing instruments containing ICs 95
7.6 Exercise: alarm unit 86 3A Exercise: heater control unit using a 741
7.7 Exercise: lamp dimming circuit 87 op-amp 97
7.8 Exercise: sequential control unit 88 8.5 Exercise: frequency standard cicuit
7.9 Exercise: lamp flasher unit 90 using TIL logic 99
7.10 Exercise: motor speed control circuit 91 8.6 Exercise: boiler control unit using TTL
logic 100
8 Circuits using Linear and Digital
Integrated Circuits 93
8.1 Introduction to integrated circuits 93 Answers to Exercises 102
,
Preface
The ability to rapidly diagnose the causes of !aults fault finding methods.
in electronic equipment and circuits is one of the The majority of the circuits have been built,
important skills that can be acquired by the eJe~tronic tested, and th.en measurements made under fault
. technician or mechanic. This book is intended to conditions, and it is intended that the student
serve as an introduction to the subject. Naturally, should construct or breadboard the circuits as
fault diagnosis skill is not aclrleved easily. since it practical project work. For this reason, readily
coinbines a good understanding of component and available components, wherever possible, have been
circuit operation together with knowledge on testing c~oseh.
methods and on how components fail. The exercises Primarily the book is intended for students
throughout the book are deSigned to assist the student studying City and Guilds 224 Electronic & T. V.
in improving his fault diagnosis technique. The text Mechanics, aty and Guilds 272 Electronic Tech
concentrates nui.inIy on component faults occurring nicians, and the TEC Certificate and Dipioma COUiSes
in particular types of circuit rather than on the fault However, it is hoped that the text and exercises, will
rmding techniques used for localising faults in com prove interesting and helpful to a much wider reader
plete electronic instruments or systems. There is, ship.
however, a section that deals briefly with system
~'
,
1i
II
tl.
~
l!
1 Basics of Fault Diagnos.is
r-----~------~~--~~----o+24V
1.1. Circuits and Test Readings
An electronic circuit is a collection of components
connected together to perform a particular electronic
function. Each component has its part to play in'the
operation of. the circuit. If any con,wonent should
fail, then the operation will be drastically chan~d.
For an example, consider the simple relay amplifier
circuit of Fig, 1.1. If RI were to'go open circuit,
there would be no forward bias current for Trl' The
collector voltage of Trl would rise, Tr2 would then 0----------------_---0 0 V
conduct, and the relay coil would be permanently Fig. 1.1 Relay amplifier
energized. was faulty. Component testing is desc:ibed later in
A faulty component produces a certain set of
this chapter. '
symptoms, which can be used to indicate the com
for more complex circuits, especially those which
'ponent and its type of fault. Such symptoms are,
use direct coupling, the effects of one component
for example, the voltage levels at ~arious points in
fault can be extensive. However the fault symptom
the circuit. ~, .
invariably inilicates i'lhich component is at fault, and
The voltages, * ~easured ;.Vith! a standard multi the exercises in the follOwing chapters are intended
range meter at the test points of Fig. 1.1, when the to give the reader experience in diagnosing faulty
... circuit is working correctly, and wi~h no input components from a given set of symptoms.
applied, are Skilful fault diagnosis requires both theoretical
knowledge and practical experience. Before attemp
-
Test !point
.
Volt ape
I +0;7
1 1+0·1
2 I ting the diagnosis of faulty components the technicial
will need to understand the purpose of the circuit
. However, with Rl open circuit the readings would
and its operation. ll1is clearly presupposes that he .
change to
also understands the principle of opeotion of the
Test point
Voltage
110 I I 2
+0,7
3
+0·15
various electronic components used. A review of the
common components follows in the next section.
1.2 Components and Common Faults
These readings indicate that Trl is non-conducting
and, since the base vol tage of Tr 1 is at iero volts, Before considering the individual types of component
.that possibly Rl is open circuit and cannot supply let us look at the ways in which a component can
base current to Tri. It is worth noting here that a faiL A component can be said to have failed if any
base-erriitter short circuit on Trl would also cause the one of its constants is out of its specified limit.
same symptclms. A resistance check would be . For example, if a 5k6 ohm ± 5% resistor actually
necessary to discover which of the two components has a resistance value of 6 kS1, or if the leakage
current of a 64 JiF 12 V electrolytic capacitor is
*Throughout the book, the voltage readings in these test
150 JiA when it is specified as a maximum of 10 JiA,
point tables are given in VQlts. .
then both components have failed.
2
3 BASICS OF FAULT DIAGNOSIS
Both these examples can be described as partial It is perhaps easy to understand failures caused by
failures since they do not necessarily lead to a com defects and overloads. but why should a c(lInponent
plete l~ of performance, rather to a small change. fail in normal use? Basically the component is ageing
Partial failures are especially important when the beCause of the stresses that are acting continuously
component is used in a critical circuit position. upon it. These stresses are of two kinds, operating
llle faults we are more concerned with are called and environmental. The operating stress is due to
c.1tastropllic failures, when the failure of the com the design conditions, and fhe life of a component
ponent is both sudden and complete. for example, can be prolonged by operating it well within it.s
a resistor goes very high in value, or becomes open rated maximum value of current, voltage, and power.
circuited, or a diode develops a short circuit 111is is called derating. Environmental stresses are
between anode and cathode. Such failures Jead to a those caused by the surrounding conditions. High
complete loss of performance and are usually temperature, high hUmidity, mecha!1ical shock and
accompanied by a drastit change in d.c. bias levels. vibration, high or low pressure. and corrosive chemi.
As a general rule certain types of component fail cals or dust in the air, are the major adverse conditions.
in a particular way. When resistors. especially the All oftbese stresses affect the component and cause
film type. fail they often go open circuit. since a some deviation from the specification, and finally
small break in the resistaqce spiral is a much more the component will faU. For example, consider a
likely event than a short circuit across the whole component SUbjected to continual cycles of heating
resistor_ Electrolytic capacitors on the other hand . and cooling; this may cause the materials from which
are inore prone to fail short circuit. Here we are the component is made to become brittle, and any
discussing the way in which components fail; this mechanical shock may then cause the component to
should not be confused with the rate at which they fail open cirCUit.
fail. nle reliability o( present-day components is The effects of adverse environmental conditions
extremely high. In other words. the failure rate is can usually be minimized by careful design, and this
low. Resistors, in particular, are very reUable. is increasingly important when an electronic instru
ment fomls an integral part of some b~dustrjal manu
Table 1.1 indicates the more probable types of facturing process where higll temperatures, vibration
failure for various types of electronic component~ and other hazards are present. .
Another cause of component fallure is higll
~. volbge pulses or "spikes", generated from switched
,
inductive loads, being transmitted along the mains
TABLE 1.1 and appearing on internal supply wires. These
Compoltent Common type offault

<'spikes" can easily lead to the breakdown of junco
tions in semiconductor devices.
Resistors High in value or open circuit.

Varia hie resistors Open circuit or intermittent
contact resulting from mechanical 1.3 Operating Principles of Common Active
wear. Components
Capacitors Open or short circuit.
The follOwing is intended only as a review; other
Inductors Open circuit. Shorted turns. Short texts should be consulted.
(including circuit coil to frame (iron cored

transformers) types).
(1) Semiconductor diodes
Thennionic valves Filament open circuit. Short
circuited electrodes (i.e. grid to These devices (fig. 1.2) have a low slope resistance
cathode). Low emission. when the anode is positive with respect to the
Semiconductor Open or short circuit at any cathode, a typical value being 25 n at a forward
devices junction. current of I mAo \\11en the anode is negative with
Diodes, Transistors, respect to the cathode, the resistance is very high,.
PETs, SCRs etc.
greater than 100 Mn for a silicon diode.
4 ELECTRONIC FAULT DIAGNOSIS
In order to pass current in the forward direction, Forward mA

curren!
a small forward bias is necessary, about 200 mV for
a gemlaniul1l device and 600 mV for a silicon diode.
---!t;:j::t---
TIle characteristics for small signal diodes are shown Anode Colhode
in Fig. 1.2. Note that, if a large enough reverse Low slope resislonc'e in
fOrwlrd COnducting
voltage is applied, the diode will break down and will slol~
lmA -
pass a damaging overload current unles~; limited by Reverse volloge Forword vollage
some external resistance in series wilh the diode. - ~'~====~---+"<O="-=-6-1/----~"'---- +
Very intense electric fields are set up in the junction
when the reverse breakdown voltage is applied
Reverse
and electrons are accelerated to such velocities that breakdown
they dislodge other fixed electrons from the crystal

lattice. TIlese electrons iI! turri dislodge others, so the
process is rapid and cumulative. ,(11e effect, put to
Reverse
good use in regulator diodes, is called avalanche. Cunenl p.A
Naturally, a normal diode has a maximum reverse
voltage rating that should not be exceeded, usually Fig. 1.2 Semiconductor diode
between 100 and 800 V.
(2) Voltage Regulator Diodes

The symbol and characteristics of a typical regulator
diode are shown in Fig. 1.3. They utilize the zener, or Forward mt.
current
avalanche effect. The device has a higher impurity
doping content than an ordinary diode, and this
results in a relatively small depletion Jayer. This
means that high field strengths (up to 107 volt/em) Reverse vollage
v.I Forword voltage
exist in the depletion layer for only small reverse -......, ,', +
voltages. By controlling the doping, diodes of varying
reverse breakdown voltages can be produced
typically from 3·3 V up to 150 Y, with power ratings
from 250 mW up to and above 75 W.
Reverse
The device behaves like a nonnal diode in the currenl mIl
forward direction, but in the reverse direction has a
very high resistance until the breakdown voltage is Fig. 1.3 Volt~ge regulator diode
reached. At this point the resistance falls to only a
few ohms.
A sImple ~pplication of the voltage regulator
diode is shown in Fig. 1 A, The voltage across the
diode remains almost constant even if the 10ild
current and supply voltage are varied over wide
ranges. +0----\
I
(3) Bipolar Transistors
Unslobi!ised
doc, input v, oz
SIObl~'
oulpul'
•
TIle transistor is best considered as a device in which

a current flowing between the' collector and emitter
is controlled by a much smaller current flowing
between the base and emitter. Fig. 104 Application of voltage regulator diode
E rniller Base
An n-p-n transistor is shown in diagram form in
Fig. 1.5, together with its symbol. 1l1e ootype
material has an excess of electrons, and p-type an
,,"~,
Collector
excess of positive charge carriers called "holes"_
~
When a junction of p and n type is fomled, a deple
tion region is set up in which no free charge carriers
;--__:___--I} Eo;'"••
exist. For correct operation the silicon transistor
requires a small forward bias voltage of approximately Emiller ,--_ _ _ _
n+_ _ _ ...J} Substrole
+600 mV between base and emitter to overcome the
junction potential set up by the flxed charges in the
depletion region. The collector base junction is Collector
reverse'biased. Electrons Dow across the emitter

base junction, but since dtere are by design many Fig. I.S Construction of n-p-n transistor
more electrons in the emitter than holes in the base,
only a small amount of r~ombination takes place. collector is 0·99 (since hFB == [c/Ie). This is the
TIlis recombination is ~ base current. 11Ie majority current gain when the transistor is connected in
or
of the elect~ons diffuse, spread, across the base
common base mode.
The current gain hFE between base and collector
untit they reach the depletion region of the collector
base junction. 11Iere they are swept up and collected is 99 (since h FE =Ie/!b). TIlis is the current gain
by the positive field. The electrons which make up when the transistor is connected in common emitter.
the cU,rrent flowing from collector to emitter are It is important to realize that hFE is a parameter
called the majority carriers since they outnumber that vades widely for transistors of the same type.
the small amount of «holes" in the base. A ~ance at any data sheet will quickly show this. A
The bper~tion of a p-n-p transistor is similiar typical spread in hFE may be from 50 to 500. Any
except that the polarities of the supplies arc reversed bias circuit has to be designed to overcome this large
and the majority carriers are «holes". spread.
1l1e transistor can b¥Perated in duee possible
_~--_+v«
modes, called common tlase, common emitter, and
common collector. The base, the emitter or the Rb
collector is made the common terminal for the input
and output signals. All duee connections have their
uses, but the configuration that gives the highest gain
is common emitter.
TIle relationship between the three currents 10 rnA
Fig. 1.6A Currents in a
flowing in a bipolar transistor, neglecting any leakage simple transistor circuit
current, can be written as
le=Ie+ 1b -1'------ +,IIC [
TIle base current Ib is much smaller than both

Ie and Ie. TIus is because most of the current carriers
crossing from the emitter into dIe base are rapidly
swept up by the collector. The base current is
typically only about 1% of the emitter current.
Take example as shown in Fig. 1.6A where
Ie = lOrnA, Ie == 9-9 rnA ami Ib == 0·1 mAo For this
transistor, the current gain I1FB between emitter and
Fig. 1.6B Circuit as 1.6A but Fig. l.6C Circuit as 1.6A bu
with the collector open circuit with base open circuit
6. ELECTRONIC FAULT DIAGNOSIS
If the transistor fails collector open circuit, there

Orain
will still be a base current flowing, and this current
flows in the emitter (Fig. 1.6B). However. if the
base or emitter fails open circuit, apart from a small
l~akage current, the effective current flow is zero
(Fig. 1.6C).
Further work on transistor faults ill a single stage SoU(C~
common emitter amplifier is dealt with in Chapter 2. Fig. 1.7 An n·chaullel junction FET
(4) Unipolar Transistors - the FET Family v.... ~

The operation of FETs differs from the bipolar
transistor in that the current flowing through the
~,~,
FET is controlled by a voltage input. TIle terminals
are called drain, source and gate, and the simplified /----0
O'I,lLF
construction of an n-channel junction FET is shown 0·1 J'F
in Fig. 1.7. TIlis is made from a bar of n-type material 0--11
to which contacts called drain and source are made
at each end. Two p-regions fonned into the bar
directly opposite each other are connected together Input
tV +
1M 10 J'F
and are caned tlle gate.
A c1,lrrent will flow between source and drain
when the· voltage between drain and sourc<e is positive. O-----~------~--~---OOV
However this current will fall if the gate voltage is Fig. 1.8 Typical junctiQn FET amplifier
made negative with respect to the source:When the
gate is negative, depletion regions are fo~ed, and must be taken in handling and soldering them a.~ str~y
this reduces the channel width between source and electrostatic fields can easily damage Ihe thin in
drain, thus the current falls.. When the gate is made sulating layer. ..
sufficiently negative, say -3 V. these depletio~

regions meet and the drain current is cut off. (5) Thyri:.wrs and Triacs
TIle important feature of a FET is that the drain The thyristor or silicon controlled rectifier (SCR) is
current is controlled by the voltage on a reverse biased another solid state device that acts as a high-speed:,
gate to source p-n junction. This means that it has a P?wer switch and is now used extensi·...ely to repla¢.e.
very high input impedance. A typical FET amplifier conventional relays and mechanical switches. The··
circuit is shown in Fig. 1.8. construction and symbol is shown in :::;-ig. 1.9 whete, .
With small signals the FET behaves like a linear it can be seen that it is made up of fOlu layers of· ...
reSistor, which can be switched from a few hundred semiconductor material in a pnpn sandwich.
ohms to severa] hundred megohms by means of the thyristor can be made. to act either as an opell
gate voltage. This can be useful in analogue switching
circuits and low drift chopper circuits.
Another type of FET is the metal oxide silicon Anode
field effect transistor (MOS-FET) sometimes referred
to as an IGFET or MOST. This device is different in
construction to the junction FET, in that the gate is
actually separated from the conducting channel by a
Gale
metal oxide insulating layer. TIle current flowing Colhod",
through the channel is controlled by the electrostatic
field between the gate and the substrate. Such devices
have an extremely high input impedance, and care Fig. 1.9 Construction Qf thyristor
I BASICS OF FAULT DIAGNOSIS
or as a reclifier depending upon how its gate is used.

Conduction between anode and cathode is blocked
in bolh the reverse and forward direction. 11le gate
has no control over the reverse characteristics, but
can be used to switch conduction in the forward Fig. 1.10 Triac symbol FIJ. J.I t Dix symbol
direction. When a small signal is applied between gate
and cathode the thyristor is turned on, and a large 1.4 Measuring Instruments and Testing Methods
forward current can flow v.ith only a small voltage
drop across the device. Once on, the thyristor can (1) Meters
only be switched off by reducing the current through To get information about the symptoms of a par
it to a value less than what is called the holding ticular fault, a set of voltage .!eadings at critical
current. TIle holding current is the specified minimum points in the circuit must be takeA. This inrormation,
current to ensure continued conduction, and this is together with additional information on the circuit
usually about ten per cent of the maximum forward perfonnance (i.e. distorted output, overheating
cunent. In a.c. power control circuits the thyristor component) is usually all that is neces.ury for correct
naturally turns off eveIY hillf-cycle when the supply fault diagnosis. So the only essential piece of test
reverses. equipment for fault fhiding is a good, general pUrpose
TIle thyristor can be switched into forward con multi-range meter. This should have a resistance on
duction by two ether means: (a) by exceeding the d.c. ranges of at least 20 kf! per volt. It is important
forward break-over voltage and (b) by applying a fast that the meter has a relatively high resistance, other
rising voltage Wavefonn between anode and cathode, wise the loading effect of the voltmeter could lead
typically greater than 50 volt/microsec, but normally to incorrect conclusions. Also when measuring
it is the gate signal that is used to control the point voltages in circuits that have fairly high resistances,
of switch on. . --the loading effect must be considered .
. The Triac (Fig. t.I~) is essentially two thyrist<!fS Take for example the potential divider shown in
connected in reverse p~rallel. This very useful device Fig. 1.12. The voltage across R2 should be 13·3 V.
can be switched into forward or reverse conduction If a meter of 20 kf! on the 10 V d.c.range is con·
by means of a control signal applied to its gate. 1t nected across R:z it will actually indicate neady 10 V..
finds it main application in full wave a.c. power If a higher range of the meter is selected. the metet
-col1t£Ol circuit... (See also Chapter 7.) current is reduced, and a more accurate indication
is given. It is always \vise to select the highest pos·
sible range when measuring voltages in high resistance
. (6) The Diac circuits .
TIlis component is often used as a triggering device
+1
for circuits using thyristors and triacs. Its symbol is
shown in Fig. 1.11. TIle diac will not conduct in
either the forward or the reverse direction until a
certain threshold voltage is exceeded, usually about
30 V. Once the threshold voltage is exceeded, the
diode exhibits a negative resistance as the current
increases, while the voltage across the diac falls. In
other words ahove a certain voltage the diac passes
a pulse oCcurrent.
Since the wac is symmetrical in operation it is
extremely useful in making economical trig.!!,er
Fig. 1.12 A voltmeter of 20 kn/V me3~llfi!lg the voltage
circuits for triac full wave a.c. controllers. Some output of a l,otcntial divider formed by two resistors of
devices are produced which incorporate a iliac and relatively high value. Voltmeter indicates approximately .
triac in one encapsulation; these are called Quadracs. 10 Von 10 V range whereas the true ou tput voltage is 1 3· 3 V
An alternative to the moving coil meter is the small, spot of light can be moved to any part of the Scr
portable digital multimeter. This displays the by applying signals to the horizontal and vertical
measured voltage, current or resistance on a three deflection plates. TIlese signals are produced frot
or more in-line digital display. TIle more digits used, V-amplifier and the timebase.
the greater the accuracy of the reading. TIle input A signal to be measured is applied to the Y-in
resistance of these instruments is typically 10 MU, of the CRG, is attenuated hy the switched aHem
which means that the unit lakes only a small current (the V-amplitude control), then amplified by the
from the circuit being measured. It seems that the V-amplifier and applied to the vertical plates of t
instrument will replace the moving coil type meter CRT. At the same time the timebase unit is trigg<
because of its accuracy, readability and high input to produce a sawtooth signal that, when applied
resistance. However, unless otherwise stated, in all the horizontal plates, causes the spot to move aCI
the exercises in the following chapters the measure· the screen at a uniform rate and then fly back an
ments have been made_ using a standard moving coil repeat the process. The result is that a bright trac
type meter, mainly because of the ready availability of the input signal appears on the screen.
of this instrument. This trace can only be held stationary if the
trigger control 011 the CRO tiniebase is correctly:
(2) The Cathode Rav Oscilloscope For a Single beam CRG there are two possible tri!
Among other useful instruments the next important ing modes: external or internal. the external pos.
from the point of view of fault finding is the cathode should be selected only when a trigger signal is
ray oscilloscope (CRO). Ulis is perhaps the most available; this feature call be extr.!mely useful wit
versatile measuring instrument available. With it, measuring the time or phase relationship between
it is possible to meaSUre d.c. and a.c., voltage, two signals, as win be seen later. 'nle normal mod
current, phase-angle, and a wllOle range of other for the trigger is 10 select internaL To hold the tf
quantities. The accuracy depends to a great extent switch the trigger select switch to INT and then
upon the care paid to the instruments' calibration, adjust TRIG. LEVEL(or TRIG. STABILITy) un
and in most modern oscilloscopes signals for .~ the trace locks.
calibration are built in. The typical input impedance Suppose we wish to measure 6e frequency anI
of a CRO is 1 Mil which has a capacitance of about amplitude of an unknown sine wave signal .. The
20 pF in parallel with it. The input impedance can eRO is set up with no input so that first of all the
always be increased by using a special probe unit. A trace is located (some instrument:, incorporate a
probe is simply a test lead which contains either a beam finder for this purpose). TIle BRILL and
passive or an active network at its end or at some FOCUS controls should be set to give a clear fine
point along the lead. The straightforward vs>1tage. line on the screen. TIle signal to be measured is
divider probe is a basic attenuator with good fre· appUpri to the V-input as shoWJl in Fig. 1.13 and t
quency compensation. TIle latter is usuaJly adjustable Y-amplitude control and the TIME switch set unti
and should be checked before use. A disadvantage of the signal can be easily measured. In the example 1
this arrangement is that the signal attenuation is high Y-amplitude control is at 2 V/cm and the time
typically 10: 1 or 100: 1, this is why the probes are switch is at 0·1 ms/cm. TIlerefore the unknown
called xlO or xlOO. signal has an amplitude of 5 V peak and a periodic
The «heart" of an oscilloscope is the cathode ray time of 0·20 ms. TIle frequency is then given by
tube (CRT). This consists of an electron gun, a
1 1
deflection system and a fluorescent screen. A high T= O---=.2'-x-'--:-10:--""3 - 5 kHz
velocity, finely focussed electron beam is produced

by the electron gUll. This beam passes between two As stated previously the eRO is a highly versati
sets of plates arranged at right angles. Voltages instrument but always make sure that it is calibrat,
applied to these plates deflect the beam both hori· correctly and sct to a calibrated position.
zontally and vertically. The beam finally strikes the Many modern CROs have double beams which
screen and a fine point of light is produced. 111is can be used to di=>play two time-r:!latcd signals. An
o
Brilliance
+
o
Focus
0-1 ms/cm
TIME8ASE
-Ve +Ve
o
Amplitude
TRIGGER
OSCILLATOR 'lnternol
log ponel
O 'Ext.
EXT.
0
TRIG.
Fig. 1.13 eRO used to measure a sine wave signal (rom an

oscillator
o
Brilliance
,--- ------- -l
o focus
:+6 V01>----...-_--1'
I
1
I
I
YI Amplitude
V/cm
,' ,
ms/cm
0
I , I ,
,
I
I
I
I TiME DIVISION
I y. SWITCH
I
! ,I.
Y2
l
I
L_A~~~LE
0 V·o------~,__....
O
TRIGGER
SELECT
'E.1.
Y:zAmplHude 0+ o
TRIGGER EXT.
MOOE TRIG.
Fig. 1.14 Double beam oscilloscope used to measure two
time-related wave forms from an astable multivibrator. TIle
y 1 beam is triggered on its positive edge.
example is given in Fig. 1.14 where the signals from

an astable oscillator are shown. Only one channel
can be used to trigger the TImebase so an additional Sig, gen.
1 ~Hl
switch is included that allows one to select either Low impedonce
Yj or Y2 for internal trigger. A single beam eRO can
also be used for measurements of phase between
Signals by applying one Signal to the external trigger
of the timebase while the other is fed to the Y-input. Fig. LlSA Simple laoorntory set-up to measure capacitance
Standard
(3) Simple Component Testing capacitor ~C. c.
~~-'----l~----~
When an instrument is being serviced and checks
indicate that a certain component is suspect, it is
then necessary to confirm the fault. Often simply
replacing the component is a sufficientcheck, but it
is always go.od practice to test the faulty component
to verify the type of fault. TItis is useful for a num
ber Qf reasons, the most importantbeing the collec
~--~ ~ r-------------------~
tion of data on component failures. A fault may be
caused by defects in component manufacture, a I kHl Audio oscillotor
design error, poor production metllOds, or ageing. Fig. I_IS8 Direct capacitance bridge. TIle ietector may be
Thus, for example, if a large number of components headphones, an osciUoscope, or sensitive 3.'';:. meter.
are failing open cirCUit, the manufacturer wiD need, At balance! Cx =} Cs
I
to be informed so that future defects can be avoided.
Tests to confirm open or short circuit conditions bridge as shown in Fig. 1.15B. to compare the un
can easily be made using the ohms range of a multi known capacitor against a standard. '
range meter, but while checking for an open circuit Tests on diodes, transistors and other semicon
it is usually wise to unsolder and lift one end of the ductor devices can also be made using the ohms
component before making the measurements, other= range of a multimeter.
wise other components that are in parallel with the
suspect component will give a false indication of the MVLTIRANGE METER
I
resistance. An alternative method of checking for
an open circuit resistor is to "bridge" the suspect
component with a known good one, and then re
check the circuit conditions.
"Leaky" capacitors can also be tested using an
ohmmeter, again by disconnecting one end of the
capacitor from the circuit. A good electrolytic
should indicate a low resistance initially as the
capacitor charges, but the ~esistance should rapidly
increase to approach infinity. Qpen circuit capaci
tors are best confirmed by placing another capacitor
of the same value in paral1el and checking circuit OA91
operation, or by removing the capacitor and testing or similar diode
it on a simple laboratory set-up as shown in Fig. 1.16 Using a semiconductor diode to det·~rmine the
Fig. 1. I 5A using a low frequency generator at I kHz polarity of a multirange meter when switched to the ohms
and two meters. Here ex = Ij2rrJVo with an accuracy range.
Ihe meter measures a low resistance, indicating that the
of better than ±10% for values from 1000 pF to black terminal is connected to the positive plate of the ,~. .
1 pF. An even better method is to use a simple a.c. internal battery.
11 BASlCS OF FAULT DIAGNOSIS
First it is necessary to detemline how the internal MULTIRANGE METER

battery in your multimeler is connected. For example,
in one typical instrument the common terminal
(marked black) has a positive voltage on the resis
tance range. If you do not know the connections for'
the particular meter you are using. the polarity can
be detennined by connecting the multimeter (on
oluns range) to an electronic voltmeter, or by measuring
the forward and reverse resistance of a semiconductor
diode of known polarity. See Fig. 1.16.
Having established the ohmmeter lead polarity,
you can discover a great deal about a transistor. First
identify the device leads if not known (see Fig. 1.17).
Measure the forward and the reverse resistance
between pairs of leads until you find two that
measure high (over 100 kSl) in both directions. These
must be the collector and emitter (provided the
is
transistor a good one). nle remaining lead is the
Fig. 1.17A Measuring tIle junction resistance of an n-p-n
base. Now measure the resistance from base to one
trnnsistor with a multirange meter. ForWard bias on base
of the other transistor leads; itsbould be low in one emitter. A low resistance (typicaUy less than 1 kn) should
direction (1 kU) and high (greater than 100 kSl) in be indicated.
the other. If the low re;tStance occurs when the
obinmeter lead with the positive voltage is connected
to the transistor base, the transistor will be n-p·n
type. Of course, it will be the other way round for
pen-po
The above check also tests that both emitter base MUlTIRANGE METER
and collector base junctions in tJle transistor are
good. If either junction shows up high resistance in .
bOtJI directions, it is open circuit; and low resistance
in bOtJI directions, it is broken down.
When testing components, and in partic~ar tran·
sistors, FETs and ICs ALWAYS
(1) Check for power supplies near the actual coin·
ponents, and in the case of ICs directly 011 the
appropriate pins.
(2) Do not use large test probes because they can
easily cause shorts.
(3) A void the use of excessive heat when unsoldering
a component and do not unsolder with the unit
switched on.
(4) Never remove or plug in a device without first
Switching off tJle power supply. Components can be Fig; 1.17B Forward bias on base collector. A low resistance
damaged easily by the excessive current surges. Oess than 1 kn) should be indicated. .
1.5 Fault Finding on Electronic Instruments and
MULTIRANGE METER.
Systems
111e previous sections dealt with componellt failures

and simple component testing, and the follov.ing
chapters are concerned with actual component faults
in one portion of a complete instrument, ill other
words for example in the power supply or the
oscillator section of an instnllnent. 1I0wev<:r, when
a complete instrument is returned for repair, the
service engineer must filst locate which blc·ck of the
instrument is faulty before he can locate the actual
component that has failed. TIIere are vario'ls methods
used to narrow down the fault to one bloc':,;, but
before these are discussed, it is useful to cc·nsider
some fairly obvious but often overlooked points:
1 (i) TIle service engineer must have a MAINTENANCE
~ MANUAL with up-to-date circuit diag:~ms. TIlis
~ manual should also give the figures of j,e per- .
~ 19. 1.17C Reverse bias on emitter base. A high resistance
i p-eater than 100 kn) should be indicated. formance specification.
I
(2) TIle engineer must have all the necessary TEST
EQUJPMENT. Usually a list of instruments and
any special instructions are in the maintenance
MULTmANGE METER manuaL
1I (3) The engineer then has to DEFINE THE FAULT
I
ACCURATELY.. lllis point is most important.
It is no good trying to locate a fault that is
vaguely defined. The symptoms must he accu
rately noted and tillS means that a functional
test must be made on the instrument.
For example consider that a signal gene::ator has
been returned for repair with a suspect power supply
failure. Before taking off the cover and checking
,
" power supply lines, the service engineer would
~
(a) check the mains fuses and if not blown
~
~ (b) check for sine wave output on all ranges,
~ (c) then make notes of tile fault symptams.
The complete circuit of most electronic im.truments
I~ . can be broken down into a series of functional

blocks; for example in a general purpose sine wave
ii g. 1.17D Reverse bias on collector base. A high resistance

:eater than 100 kn) should be indicated.
generator these would be power supply, variable sine
wave oscillator, !:::uffer amplifier, and output atten
uator. By treating the instrument in blocks, rather
~ than as a whole, it is possible to narrow down the
~. search for a faulty component first of all to one
block, then by measurements \\-ithin that block to
locate the actual faulty component. TIle methods
used to decide which block is faulty are Aerial

Frequency AF .
(0) Input to output (or beginning to end). (j) . changer IF, amplifier
(b) Output to input.
(c) Random.
Oeleclor
&&&&&0
Oemodulolor Loud
(€f) Half-split.
speaker
All of these have their particular advantages and
Fig. 1.18 Block diagram of superhet radio

uses. The RANDOM METHOD, which implies a
totally non-systematic approach, is rarely used. A
method based on the reliability of components can also (4)(IF). Output correct. TIlerefore the fault
be used when there is a wealth of service knowledge is somewhere in blocks (5) to (8).
and experience concerning a particular instrument. For (b) Split blockS (5) to (8) in half by checking
example a service engineer might make the reasonable output of (6). Input signal can be left at (I).
assumption that... because a particular electrolytic capa No output.
citor has been at fault in 60% of the instruments recently (c) Leaving signal at (I), check output from (5).
returned, it is a strong possibility that the next Output should be correct, indicating that the
faulty instroment also has a faulty electrolytic faulty block is (6), the demodulator.
capacitor. He would naturaUy check this first, and You can try this method for yourself by assum
in most cases save valuable service time. It must he ing that the fault is in block (3) for example, and
stressed, however, that this method depends upon you will find that the number of checks nel;essary to
the availability of a large amount of data on the locate the fault is still three. On average, four tests
reliability of the various components within an would be reqUired by using the input to output
instmment. Most service engineers would use a technique. The half-split method is most useful when
logical systematic approach to system fault location. the number of components or blocks in series is very
The INPUT TO OUIi>UT and OUTPUT TO . large, for example where several series plug and
INPlTf methods are examples of this systematic socket connections are used, or for heater chains in
approach. TIle method is fairly obvious. A suitable valve equipment. nlere are, however, several assump
input signal (if required) is injected into tlle input tions made for the half-split: (a) that all components
block and then measurements are made sequentially are equally reliable; (b) that it is possible and prac
at the output of each block in turn, working either tical to make measurements at the desired point; and
from the Input towards the output or from the out (c) that all checks are similar and take the same
put back to the input, until the faulty block is amount of time. These assumptions will not i1 ways
located. ll1is logical method is the one most service be valid and it is up to the serviCe engineer to then
engineers use on eqUipment containing a limited decide the best method of approach.
number of blocks. The half-split method can also be easily complicated
The HALF-SPUT method is very powerful in by
locating faults in instruments made up of a large (0) An odd number of series units.
number of blocks in series. Take for example a (b) Divergence: an output from one block feed
superhet radio receiver shown in block diagram form ing two or more units.
in Fig. 1.18. Since there are eight blocks it is possible (c) Convergence: two or more inputs bei~g
to divide the circuit in half, test that half, decide necessary for the correct operation of one
which half is working correctly, then split the non unit.
functioning section into balf again to locate the (d) Feedback: which may be used to modify the
fault. An example is the best way of really characteristics of the unit or in fact be a
understanding the metltod. Assume that a fault sustaining network for an oscillator.
exists in the demodulator of the receiver, tbe sequence When using any of the methods as described, try
of tests would be as follows: and use the method, or a combination of them, that
(a) Split in half, inject signal into the input of will locate the faulty block in a system in the shortest
(I) (the aerial circuit), and check output at possible time.
2 Single Stage Transistor Amplifier
---'t-----1'-------<,. Vee + 12V

2.1 Basic Principles
TItis section is concerned with the effects of indio
vidual component failures in a single-stage common
emitter amplifier. The circuit shown in Fig. 2.1
usually has eight components. Remembering that
resistors can fail high or open circuit; capacitors
either open or short circuit; and that the transistor
can fail open or short circuit between any of its
connections; it can be seen that a. total of at least
twelve faults is possible. For each of these faults a
unique set of conditions will exist.
~----~--~~~~~~~~'OV
Before considering any fault conditions, the
operation of the circuit must be understood. In a
class A amplifier a mean current flows througll the Fig. 21 Single stage class A transistor amplifier; normal
working voltages
transistor, and the input signal causes this current
to either increase or decrease. This change in collector and if we neglect base current, the d.c. b.1Se voltage
current then develops a voltage signal across the col can be calculated from
lector load resistor R3 • TIle operating point of the /"v,
. collector voltage, that is the d.c. voltage between V J>. CC R'
l collector and the 0 V rail, should be a value that
~ !!llows equal positive and negative swings of the
B-(R 1 +R 2 ) 2
and the emitter voltage VE is given by

~ :mtput signaL As a rougll approximation, VCshould :~"-'-'--I
~ ~e half the supply voltage. The whole purpose of B. ".___ ~J
VI:: = VB - VBE \1f: 0'1, -V'Ot
VC ::-tc!23 +\k>:.
j :he bias components (Rio R2 , ~) is to fix this where VBE is the forward bias voltage between base
~ )perating point, and to keep it stable. Stability is and emitter, typically 0·7 V for a silicon transistor.
~ lery important sillce a number of factors cause the Then
Ii.· )perating point to change. For silicon transistors
:he most important is the change in current gain
. VE
Em1 tier current IE = 7f;
;~. IFE. This can have a value from 50 to 500 for the
and as we are neglecting base current, IE ~ Ie so the

i\! lame type of transistor and, obviously without some
orm of stabilizing circuit, the operating point would

~ :hange drastically in the circuit every time the
d.c. voltage at the collector Vc is given b:1
l ransistor was changed. The bias circuit achieves

Vc = VCC!- ICR3
I tabilization by fixing the value of the base voltage Now, since VB is fixed, the d.c. current through the
I i'B and by keeping it constant irrespective of changes transistor will he fixed, and this gives the operating
~ 11 the base current. To do this the values of RI and point Ve.
;~ \2 must be chosen so that the current flowing In operation, the circuit is an example of series
t hrougll them is much greater than the transistor negative feedback. Imagine that the collector current
I 'ase current. 111ese resistors form a potential divider increases, thus causing the operating point to faIL
~
3 <1
~~
i
I
15 SINGLE STAGE TRANSISTOR AMPLIFI
'{he emitter .::urrent also increases, raising the emitter TIus shows the close agreement between the calcu
voltage VE. However, since VB is fixed by the lated and the measured values. When fault finding
potential divider, any increase of VE must reduce the on any circuit always try to make a rougll calculation
voltage between base and emitter of the transistor, of the voltage you would expect. This can be an
and this in turn causes a decrease of collector current.. invaluable guide as to which parts of the circuit are
This tends to counteract tne original rise to stabilize functioning correctly.
the operating point. Now let us consider the effect of component
Having set the correct bias with the resistors, the fallures, taking each in turn.
a.c. input and output signals must be coupled to and
from the circuit without disturbing the d.c. levels.
2.2 Resistor Faults
To do 111is capacitors C 1 and C1 are used. Both these
should be fairly high-value electrolytics, say 10 IlF, Rl OPEN CIRCUIT (Fig. 2.2)
to enable the circuit to amplify low frequencies.
Capacitor C3 , the decoup!ing capacitor, enSUres that
no a.c. signals appear at the emitter which would
TP
MR I~ No output
signal
reduce the gain of the circuit. Since the internal When Rl goes open circuit, the gurept flown" ill
resi<;tance of the emitter base junction is quite low, RZ and the base is zero. It follows that the tran sill2!"
C] must be rugh in value. A typical value is 100 pF. is cut ofT SO both the emitte ase volta are
.ibJ~ For the circuit shown in Fig. 2.1 the calculations
~. inee no collector current is flowing thevoltaie
of the d.c. bias voltages would 00 as follows:
dro2ped across the co1lector load Ra is zero and tAs;
collector voltage itself is the same as the supply
_ Vee
voltage V,e.
VB~(Rl+Cf3}·R2 _
- 47 k!2 + 12 kU 12 kU-2·4.
VE = VB - VBE=24-0·7= 1·7 V
. Ve= Vee-feR]
where Ie IE = VEIR4
'This is assuming 111at the current gain is high and
that the base current c.-m be neglected. This is nearly
always the case. Therefore Input
Ve = 12 -(3·05 rnA x 2·2 k!2) = 12 6·7'" 5·3 V

Putting these in table form we have, for the calcu
lated values:
TP
~~ In fact when the circuit is built, the actual voltages Fig. 2.2 R. open circuit
measured with a 20 kU/V meter will be slightly
different. This is to be expected since the bias resistors
used have a tolerance of 10%. R1 OPEN CIRCUIT (Fig. 2.3)
The actual readings w;';
TP
Meter
2
5·5
3
1·7
TP
MR 1:'6 I 2~51 Grossly distorted
output; negative
going signals
reading
clipped.
Without R2 in circuit the current lhat was flowinR

tJu()lIgh Ra nl)V/ trie~ tl) OIlY! into the ba"!.: of the
transistm. But the value of base ell! rent Wlll be
limited hy the tr3mhtc)r current gain, SC) less current
fI'J'VI~ ttlfe)".? R,. ·Il.i~ Jli<e<m~ Olat HI!: ba1-e v(J1ta!l;'!
must rise. TIle base current in fact rises to a value
that completely «turns on". or saturates the transistor,
Y,I.jrA:.'4 "~."":,'/J..:"'.~,;~;I.r: .., "...~4r~r:; Yt II,,'1.'Y/ t~/ ..~ ... t 1'}·11(~/ .." '--;
the emitter voltage.
Fig. 2.4 R. open circuit
R.t OPEN CIRCUIT (Fig. 2.S)
Input iJR I2~3 I 1~ I ~ I No output

signal
With an open circuit between emitter and 0 V,.u2
currents flow through the transistor_ Ollie collector
voltagssJhs;refore rises to Vee.: The v"Qltage at the
base is fixed by the potential divider"l~, and Rl and
since the base current is small in com]>arison to the
current through R2 this voltage hardly changes at all.
Fig. 7..3 R:a open circuit As with the previous example when the meter is
connected between emitter and OV. a small emitter
current flows so the voltage indicated at the emitter
R3 OPEN CIRCUIT (Fig. 2.4) is slightly higher than normal.
~R I 0~75 I :'1 I I No output o-~---<p-~---<p-----o +12 v

0\ signal
Without R1 in Siircyi' tim mllectQr Gllrucnt js zero,
so apy current tlpltld pg in the ,mWer rom! now be R,
supplied (rom the base. The base/emitter junction

acts like a forward biased diode placing ~ in parallel
with R2 • Since Ra
is a low value resistor (S60Q) the
C, 0)
emitter X2aaee £aUf t9 a yep' low value. The base o-=.JII-'+--t-~-H
§So
e<
voltage, as expected is ahpU$ mV greater than
II
the emitter voltage..

Input
It might be reasonable to assume that the voltage

reading at the collector would be zero since the
resistance is open circuit. However, when the meter
is connected it presents a high resistance path from
i the collector to a V. and the base/collector junction

acts like a forward biased diode passing a small current
through the meter. Pig. 2.5 R~ "!,,,,n circuit
17 SINGLE ST,l\GE TRANSISTOR AMPLIFIER
+12V
2.3 Capacitor Faults
C1 or C2 .OPEN ClRCUIT (Fig. 2.6)

TP 2 No output
MR 5·5 signal
WiUi this type of fault the bias conditions of the
circuit are ynchanged. The fault call only be an open
circuit coupling capacitor. A check with an oscillo
scope is necessary to determine which one is actually
Inpul
faulty.
------~~------~-------O+12V
o-------~----~--4---+-~ OV
. Fig. 2..7 C 3 open circuit
C3 SHORT ClRCUlT (Fig. 2.8)
~I
lopu! TP No output
MR
signal
The emitter resistor I\.q is shorted \jIut. so the
emitter voltage reads 0 V. The transistor heavily
forward biased saturates and therefore attempts to
ov pass a large curr~nt. However, the transistor current
is limited to a value given by Vee':;" R3 which preven1
Fir. 2..6 C, or C. open circuit
the transistor from being damaged. TIle base voltage
must be 0·7 V higher tllan the emitt;Z
C3 OPEN CIRCUIT (Fig. 2.7)
TP 2 +12V
LQw gain
MR 5·5
Again the bias conditions are unchanged. The
symptom that identifies this fault is the fact that the
amplifier voltage gain has fallen. With C3 open
circuit, irc. signals will appear acrosS I\.q introducing
nes,ative feedh3&k. The voltage gain will fall to a
value given by R3 -:- I\.q, Le. approximately 4.
Inpul
Fig. 2.8 C 3 short circuit

18 EU.CTRONlC FAULT DIAGNOSIS
2.4 Transistor Faults COLLECTOR/BASE JUNCTION SHORT CIRCUIT

(Fig. 2.10)
COLLECTQR/BASE JUNCTION OPEN CIRCUIT
(Fig.. 2.9)
TP
MR
No output
~R I! 2
3
3
2·3
No uutput
Ai with any short circuit a clue to the fault is given

Since the collector is open there can be gO Gg!1e&{g(;. by the fact that the voltages on the base and collec
current flowin~, so tile voltage at Tn rises to +12 V. tor are eQual. With this fault the ¥i[(~uit 4fectively
The base/emitter junction now acts as a forwarrl reduces to.Ra in series with the base/emitter diQd~
biased dipde in a similar wax as for the fault of Ra iW!..L, The resistance of this path is milch lOwer
<:?pen circuit. . than R, and R 2 • so tbe effect DOh, latrer wi#n"
can be neglected. TIle current flowing in ~ is given
-----..-----.-------0 -I- t2V by
- VCC":" VBE _12-0·7
J- 4mA
P; + R4 2·76 kil"
The voltage at the emitter will then be I x R4 =
2·3 V. The "X9Itaees at TIt Dud 2 will he 0·7 V Ngher
than this, sufficient to forward bias the hase/ewjttu
Input.
-
diode.
-------~-------~------o +12V
Equivalent circuit
Input
Sose emilter diode
ov
Fig. 2.9 Collector base junction open circuit Fig. 2.10' Collector base junction short circuit
19 SINGLE STAGE TRANSISTOR AMPLIFIER
EMITTER/BASE JUNCTION OPEN CIRCUIT

(Fig.2.11)
;IR 12~3 Il~ I~ No output

signal
With this fault there c.1.n be no cunent flowing in
$lw Usislgr . TIle voltage dro2s across R3 and &<
are zer9x s.) tJ;,m cpileetor voltage rises to Vee and the
emitter voltage is 0 V. 111J. v91tage 011 tbe base is
determiped hy ,h,
pgteptial divider RI and R2 and
therefore remains atm'. There is no djfference jp
lnpul
the sYmptoms if the _ or the emitter connecti2,tl

to the junctiog is open circuit.
+12V
Fig. 2.12 Emitterbasejunction·short circuit
COLLECTOR{EMITTER SHORT CIRCUIT

(Fig.2.13)
TP
Inpul
MR
11le voltage at the eWitter is equal to that on the
collector, indicating a short. The :value of the voltage
will be delegniQM hy Ra and & which pow form
l,Rotential divider:" 111e base voltage remains un
changed at 7-3 Y since the emitter ¥QUaie has risen,
thus cutting off the base/emitter diode. .
Fig. 2.11 Emitter base junction open circuit
•
+12V
EMI'ITER/BASE JUNCTION SHORT CIRCUIT

(Fig. 2.12)
TP
MR IO~13 Il~ I O~I31 No output
The Yo1tases at TPI and TP3 will be equal end at iI

low value sipce Jl.. , a low resjstance is placed directly
ill parallel With R? With a shorted base/emitter .
Inpul
junction aU transistor action ceases, so the colleelm"
voltage rises to V",.
Fig. 2.13 Collector emitter short cin;:uit

------.-------~ +30V
Questions
(l) Estimate the voltages that would be indicated
by a 20 kn/V meter between the test points and
o V in aU circuits in Fig. 2.14. In all case.'l silicon
diodes and transistors are used
(2) 11le circuit for this exercise is a common emitter
R3
amplifier (Fig. 2.15). It has a voltage gain of about RZ 18k
2k
80. First calculate the voltages you would expect
to measure between the test points and 0 V. Follow
the procedure as outlined. OV
111e following table shows the voltage readings Fig. 2.14A
at the test points for various component faults. In -15 V
each case state which component is faulty and the
type of fault.
IOp.F
C2+
1 2 3 Additional - lJ-!-o
C,
symptom ~I------<~--t-l
No output 10p.F
Fault A 0·16 9 0·16
Fault B 1·5 9 1 . No outpUt
Fault C 0·85 1 9 0.15\ No output
Fault'D 1·5 lA5 145 No output
Fault E 1·5 4·5 0·8 Very low gain
Fault F .0 9 0 No output OV
Fig. 2.14B Trl is silicon
- _ - - +IOV
(3) The circuit shown in Fig. 2.16 is a common base
amplifier. With this type of amplifier the input is Tq is silicon wilh
~ applied at the emitter and the output taken from the on hFE 0" 40
i collector. TIle bias circuit is identical in operation to
~ the potential divider bias of the common emitter.
.j
!
1 Again calculate the voltages you would expect to
measure with a 20 kH/V meter between the test
:j points and 0 V. TIlen determine which component or
R,
33k
l components could cause the fonowing fault OV
I conditions. -IOV
j Fig. 2.14CTr, issUicon with anhFEof40
1
1 1 2 3 R,820n CD
4
.~
30V+ 0-----1
I Fault
;i
A
B
C
0
0
3·8
0
3
3
12
12
3·8
R~
560n
i~ F
D
E
1·1
S·2
3·7
1·7
5·9
4-4
)·1
5·9
3·8
--0 +IOV
~
~ .. oV - o--------------------'o---~.:_o
~ Fig. 2.14D
:?
i
~
;:!
~
"
~
;.
~
j
L 1 ::)1 Nl:iLt: ::; I Al:it: I HANS!S I UH AMPLI FI ER
+ fOV
o
c Output
002"F
n
Ahnosl ide:nticol
'0 input
Fig.2.17C
~Olf-,,_rp--~_-<> ~ ~
m{tu1 1kHz
'Squore woves.
R,
!O~
o.
1N914
Ov.p: Y T
Fig. 2.170
Fig. 2.15 Common emitter amplifier
_ _ _ _-+-_ _+--0 +12 V
Zero output
Qutpll'! 60kHz
B sine wove
0-----+-,----+----'-----0 0V
Fig. 2.l6 Common base amplifier
Fig. 2.l7E
Input
01 470kHz
Fig. 2.17A (4) In the circuit of Fig. 2.17 no d.c. bias vol tages
are given. Instead the input and output waveforms
are displayed. In each case one component can be
considered to be faulty. State which component and
No output the type of fault.
siqnol
Fig. 2.17B
I!
I 3 Power Supply Circuits
I
1
j
1
i
I1
j
j MQit'lS l40V r,m.s Slobl•
n '.
. 1 Basic Principles of D.C. Power Supplies .., 50 til d.C_ OU1put
L filler RegUI.lot
ractically all electronic instruments require a source
T"'T j--.;,..YLOOd
f d.c. power before they will operate. Sometimes LJ
le source is a battery, but more usually the power is
btained from a unit that converts the normal single Fig. 3.1 Block diagram of conventional power unit
hase a.c. mains supply (240 Vat 50 Hz) to some TIle RECTIFIER unit converts the a.c. voltage
ifferent value of d.c. voltage. from the transformer secondary winding into pulses
The function of the power supply is to provide of unidirectional current. Three types of rectifier .
le necessary d.c. voltage and current, with low circuit are used for single phase: the half-wave, the
:vels of a.c. ripple (mains hum) and with good full-wave, and the bridge. 111ese, together with their
:ability and regulation. In other words it must output waveforms, are shown in Fig. 3.2.
rovide a stable d.c. output voltage, irrespective of TIle half-wave rectifier, althougl1 being a simple.
langes in the mains input voltage and of changes circuit, has the main disadvantage oflow'~fficiency. '
I the load current.
TIle diode conducts only on one half of the cycle, so
A further important requirement of a modern unit the efficiency cannot be greater than 50 per cent.
that it should be able to limit the available output TIle full-wave rectifier uses two diodes, each conduct~.
Jrrent in the event of an overload (current limiting) ing on alternate half cycles to give mu-ch higller i,
1d also limit the maximum output voltage. Damage efficiency. However, to achieve this, a transformer .
} sensitive components, such as ICs, in the instrument with a centre tapped secondary winding i:; necessary..•.
m easily occur if excessive voltages appear on th~ lhis means that twice the number of tums is recluiI'eO:
::lwer supply lines. on the secondary winding. This circuit was
There are various methods of achieving a stable when valve rectifie(s were ill use, since it was _•. ' __"',_._
.c. voltage from the a.c. mains, but only two methods to wind extra turns on the transform'er than to use
'e commonly used. These are more valves. The bridge rectifier, noW the circuit
(i) Using a linear stabiliser
choice, uses four diodes to achieve rectification ,ver·.'.it".P~~~);;
f'.
. (ii) Using a switching mode stabiliser

the whole cycle, and no centre tap is required. The·
, Both have their advantages and disadvantages as
four diodes can now be supplied in one
I
ill be seen. TIle switching mode power supply unit unit, which is more convenient and somewhat
iMPU) is a relatively new inilbVa'tionand finds its than wiring in four separate diodes. However
lain 'use in high'power,applications (I 00 W upwards). one part of tile encapsUlated bridge circuit fail;
~,,-,- .. .
2 The linear StabilIzed Power Untt whole unit then has to be replaced. .
Ii he block diagram of a conventional power unit is

Following the rectifier is the FiLTER which.
serves to smooth out the pulses received from.
ft lown in Fig. 3.1. The TRANSFORMER serves two rectifier. Th~ cirCUit can have either a Cal)aCllnre
~ ain purposes: it isolates the equipment d.c. power an inductive input as shown in Fig. 3.3. The in
~ leS fro111 tIle mains supply, and it changes the level
~~.' . the a.c. maillS voltage to some lower or higller filter, or choke input filter, is more commonly
used when the power unit has to supply a large
J lue. TIle ratio of the secondary voltage to primary
t lltage is determined by the number of turns on each
current. On low power eqUipment a capacitive
filter is more typicaL TIle input capacitor, called
~ inding.
23 POWER SUPPLY CI RCUITS

(i I Holf wove
~ A---- "
. Yo .... Nole thol Vo !!. 0·7 Y lor
"';'"0 ",,;0;,,
(0 I Copocitor input
~~~)
Oulpul woveform ossuminq 50 Hz o.c. input
( b I Choke inpul
V~ 10 40 50 60 70 Time in ms
~
(ii) full WDve rectifier
0_ _ -1__'.
Fig. 3.3 Filter circuits
"reservoir", is used as a storage device for electric

charge. Let us suppose a reservoir capacitor is con·
neded to the output of a half-wave rectifier as in
Rg. 3.4. When the diode conducts on the positive
half cycle, the capacitor is charged and a large pulse
of current is taken. The voltage across the capacitor
then rises to nearly the peak value of the a·.c. second
Oulpuf. wavefQrm Dssuming 50Hz D.C. input ary voltage. When the secondary voltage begins to fall,
L,Q---, +
+
Vo
+ !vp~=fi
70'
V. -Vd '
lime in mS
no-------'
(iii) Bddge reclifier

+
H--t-'I----+t--+:--+----jl----+::--~ Time
in msec
C discharging via Rl
C charging vio 0
I
Current
Fig. 3.2 Single phase rectifier circuits through
C
Fig. 3.4 Reservoir capacitQr connected to the ou tpu t of a

half wave rectifier. Showing typical wave forms, assuming
50 Hz a.c. input.
the diode becomes reverse biased, and the capacitor
f-~·I~
now discharges through the load resistor. The voltage

;:cross the load now more nearly represents a d.c.
level, but superimposed upon it is an alternating wave f/I
form, caUed the ripple. The value of the ripple ampli d,v,de;Ol \
UnstQbliz~ input • Stabilized
tude depends upon the size of the capacitor and the
load resistance_ To achieve low values of ripple a high
value electrolytic capacitor, typicaliy 500 pF or more.
"~ "",,\~, )~'
has to be used. A number of points should be noted

concerning the reservoir capacitor.
(0) Since it is an electrolytic, it is polarized and ~--------------------~----4~ _
must be connected correctly in the circuit. Fig. 3.5 Basic block diagram of a linear regulator
(b) Its d.c. working voltage must be greater than
the peak of the transformer secondary voltage. error amplifier to give power supplies of ·~xcellent
(c) It must be physically large since it has to absorb performance.
large pulses of current when it is cbarging, the The main advantage of the linear regulator is that
peak values of which may be scveral amperes. the output is continuously controlled to give good
If too small a capacitor is fitted, it may over stabilization against mains input changes and good
heat and possibly explode! regulation against load current changes.
Check the ripple current rating. Atypical specification for an output voltage of
TI1C other components of the circuit form a low +15 V @ 100 mA load current is
pass mter, which reduce still further the output line stability 10000:]
ripple voltage. Typical values for the iron-cored in (a 10 V change in mains
ductor are 1 to 5 Henries and for Cz , 500 pF. In some supply giving 1 1 mV
circuits these components may be omitted, especially change in d.c. output)
when an efficient regulator is used. TIle inductor Output ripple 0·1 mV pk-pk at full load
also is often replaced by a wire-wound resistor oflow D.C. output impedance 0·05 ohms .
value, say 22 n, in which case there will be a voltage Temperature coefficient 200 p.V per degree C
drop across this component resulting in a lower out Load regulation 0·033% from ~ero to full
put voltage. load (i.e. an output change
The last block is the REGULATOR, which is used of 5 mY)
to keep the output voltage constant irrespective of TIle limitation in the linear regulator drcuit is
changes in the mains input voltage and of changes in that good performance is achieved at the expense of
the load current. TIlese two functions are called inefficiency. Power is dissipated and lost in the series
line stabilization and load regulation respectively. All control transistor and this power loss increases with
linear regulators comprise load current. A large heat sink is required to ensure
(a) a control unit tllat the junction temperature of the series transistors
(b) a reference element (usually a zener diode) is kept within its rated value. For power units supply
(c) an error amplifier ing above about 100 W, the Switching mode regulator
as shown in Fig. 3.5. becomes a preferred alternative.
In operation, the circuit compares a portion of the
d.c. output voltage with the reference voltage. Any
3.3 Switching Mode Power Supplies (SM?U)
difference between the two levels is amplified by the
error amplifier, and the output fed to the control lltere are two main variations of this type. In one,
unit. TIle stability and regulation of the output a {,1St switching transistor is uscd as the control
voltage depends upon the stability of the reference element in the regulator (Fig. 3.6). TItis transistor is
clement and the gain of the error amplifier. High gain switched on and off at a frequency abov(! audio
Op-Amps in Ie form are now commonly used as the (usually 20 kHz). TIlc d.c. output voltage, after being
25 POWERSUPPLYCiRCUITS
l included to reduce Ule switching spikes that would
Uns10bJlized Fitter otherwise be fed back into the mains supply: TIns
"v SIIppi,
SMPU offers considerable advantages in terms of
N
efficiency, reduction in heat loss, and reduction in
, overall volume. However it does not possess the

regulation performance that can be achieved in the
linear circuit. Switching mode supplies are now
commonly used where large currents at low voltage
are required; as in equipment using many digital ICs.
3.4 Power Supply Protection Circuits

Fig. 3.6 Switching mode power unit using secondary
s...itching
Some fonn of protection must be incorporated in
smoothed by a low pass filter, is controlled by vary even the simplest power supply. A common form is
ing the mark-to-space ratio of the switching signal. the standard fuse which serves to disconnect the unit
.... Sllch techniques are known as secondary switching. from the mains sllpply when an overload or short
The error signal, generated by comparing the d.c. occurs. A power unit may have fuses in the line and
output with a. reference levet, is used to control the neutrat mains wires, and also a fuse in the d.c.
duty cycle of a free-running oscillator. TIle advantage unstabilized line. Fuses usually do not blow soon
of this type of circuit is that tbe series transistor heat enough to protect the series transistor iIi tlle regu·
dissipation is greatly reduced, hence greater regulator . btor if the output is shorted,and so some form of
efficiency. current limiting device is used. A simple circuit for
Another form of the SMPU is shown in Fig. 3.7 achieving this is shown in Fig. 3.8 where the load
and uses a principle called primary switching. TIle current flows through a low value current monitoring
mains supply itself, after rectification and smoothing, resistor. If the load current increases beyond a pre
is switched at high frequency by high voltage switch· determined value, the voltage developed across this
ing transistors. With this method tlle transformer resistor turns on Tr2 which in tum tends to turn off
following the switching transistors can be much Trl, the series transistor.
smaller than the bulky 50 Hz transformer required Over-voltage protections can be provided by a
in conventional supplies. Regulation is achieved by circuit wInch senses the d.c. output voltage, and
again varying the Switching duty cycle of the tran· compares it with a reference level as in Fig. 3.9. If
sistors. Naturally RF suppression circuits must be tlle d.c. output voltage rises above Vz a signal is
generated which triggers the thyristor and this short
Tr, and Tr~ are
high YOlloge circuits tlle output, either blowing a d.c.line fuse or
switChing frons is tors
L operating the current limit. Such circuits are called
+ "crowbars". Naturally the fault must be cleared befoie
H+----~Ic::r3
L
Siobilized
de the circuit can be reset.
rCJ •. output
Slobi1ized
. Tr3 output
Error
ompJlfier /
~----1e~ . .- Reference
""lIoge I
DZ
Referen~e
zener
- 00---------"---_ _ _ _ _ _-<._ _0
R4
I
Fig. 3.7 Switching mode regulator using primary switching Fig. 3.8 Linear regulator with simple current limiting circuit
ELECTRONIC FAULT DIAGNOSIS
l output vollage noted for say a ±10% chanf.e in the
Powe< supply a.c. input. TIle mains input can be varied using an
SCR
wiih cunetll
limd 0< fuse
adjustable auto·transfonuer as shown. l11cn if, for
in d.c.hne R example, the d.c. output changed by 50 mV from
N 10 V, i.e. an output change of 0·5%, then the line
Fig:. 3.9 Over-voltage protection circuit (crowbar)
stabilization would be 40: I.
Load regulation is measured, keeping the a.c. input
t5 Testing Power Supply Circuits constant, by noting the change in output when the
load is varied from zero to full load.
nle main parameters which Qught to be measured
:ither in a test department or by the service technician · = 01ange in d.c. output x 1000f.
Load reguI a tIon 10
tfter he has repaired a power unit are the following: D.C. output on no load
(0) D.C. output voltage For example suppose the output changed hy 20 mV
(b) Available d.c. output current frOIl. 10 V. 111e load regulation is
(c) Output ripple voltage at full load
3
(d) Stabilization against mains supply changes 20 x 10- X 100%= 0.2%
(e) Regulation from zero to ful~ load 10
These can all be measured using a standard test set-up To obtain fuller infomlation on a power supply's
is shown in Fig. 3.10. performance it is often necessary to plot the load
Vo<ioble
regulation curve. TIlis is a plot of output v )1tage
1
r ·--4 outo-tronsformer
l
,,
,
f;,,
<>------+- : Fuse +
'40V Power supply
io Hz under lest
III
Fig:. 3.10 Laboratory test set-up fOI measuring the per·
formance of a power supply unit
The d.c_ output voltage should be measured, and against load current. A typical result for a unit with
i.f ne.,::ess~.ry adjusted, ,·:hen the unit is iuUy I03ded. current limiting is ",ho\;;n in 3.! !.
Howeyer it is sometimes advisable to measure the
output on a low load and then gradually increase
the load current to maximum. TIlere should, of +
course, be little change in the output voltage:
TIle peak-to-peak ripple amplitUde can be checked
~ best by measuring at the output with an oscilloscope..
i
L -______________~~_ _ _ _ _
A sensitive a.c. range must be selected because the
ripple should be quite low, typically less than 20 mY. No Full 1)
logd load
Measurement of stabilization and regulation
requires that any small change in d.c. output be
carefully noted, and therefore a digital voltmeter is Fig:. 3.11 Typical load regulatiml plot for a. power unit with
a current limit
often necessary. For stabilization measurement, the Between no load and full load current the chan:le of output
unit should be fully loaded and the change in d.c. voltage ~hould be very 51TIail.
~7 POWER SUPPLY CIRCUITS
Ui Fault finding Techniques and Typical Fault
Conditions
To regulotor
Nhen a faulty power unit is returned for repair, the DZ
:3ult has to be isolated to SQme particular portion of Multimeler sal
to 'ohms' range
:he unit- TIle fault may lie ill the transformer, the
:ectifier, the filler section, or the regulator, and
;ne<lsurerncnt with a voltmeter will be necessary to
.ocate the fault.
However its probably best to start diagnosis with
i few rather obvious but often overlooked checks.
First measure the d.c. ou tput voltage. If this is
Fig. 3~12 Using an ohmmeter to measure the resistance
lero; the next check should be on the mains input. across the unstabilized line
Is the mains supply reaching the transformer primary?

If it isn't, there is the possibility of a faulty plug
:relatively simple to repair!), open circuit mains there will be a low resistance ·path through the
Nires, or a blown fuse. If the fuse is suspected, always rectifiers and a leakage path through tlle capacitor-
:est its continuity with an ohmmeter, never rely on Returning to the faulty power unit, suppose how
list a visual inspection. Jt is also worth noting that ever tIlat the fuse is intact, and that the mains is
roth the live and neutral wires may have a fuse in reaching tlle primary. TIle next step is to measure the
:ircuit, so make sure both are checked. secondary a.c. voltage, the unstabilized d.c. voltage
If tile fuse is blown it has done so because of some then the d.c_ voltage in the regulator and so on, until
:ault condition and tlle fault mtist be cleared before the fault is located.
1 new fuse is fitted. Resistance checks (with tIle mains Table 3.1 lists some typical faults together with
JnphIgged!) must be, used to locate such a fault. Use the associated symptoms. The faults are only a sample
In ohmmeter to measure the resistance of the trans of those which may occur. Locating a fanlty com
former primary, the secondary, tIle rectifiers, and so ponent from a given set of symptoms win come with
:>0. The winding resistance depends, of course, on practice and the following exercises are designed for
the size of the transformer. TIle primary resistance, that purpose.
for a medium size transformer, should be low,
typically about 50 Q. The secondary, usually supply
ing a lower voltage, may have a resistance of only a
few ohms. Detecting shorted turns on a winding can
therefore be quite difficult. Wherever possible com
pare the ID<!3.Sured resistance \\ith any a\'ai13ble data
on the type of transformer being used. Another use·
ful check i.s to run the transformer off load and test
for overheating.
When using an ohmmeter take care to use the
correct polarity for resistance checks where diodes,
electrolytic capacitors and transistors ate present. It
is all too easy to get misleading results. For example,
in Fig. 3.12, if the meter is used to measure the
resistance of the unstabilized line, tll.e positive prod
(connected inside the meter to the positive plate of a
battery) should be placed on the positive line and
the negative prod to eart.h. If tIle meter is reversed
I 28 ELECTRONIC FAULT DIAGNOS!S
TABLE 3.1. Typical Faults on }'ower Supply Units 3.7 Exercise: Power Unit with a Simple linear
Regulator {Fig. 3.131
FAULT SYMPTOMS
Mains transformer, open D.C. ou tpll t zero. Second· This unit incorporates most of the features discussed
circuit primary or secondary ary a.c. zero, high resistance earlier and is designed to give an output of 12 V at
primary or secondary.
100 mAo 11le output resistance is less than 0·5 n
Mains transformer, shorted Two pos$ibilltie~: (a) mains the load regulation better than 0·5%, and the rip;le
turns 011 primary or fuses blown or (b) low dc.
secondary output and transformer over less than 5 m V peak to peak on full load.
heating becallSE' of exce.~\'e TIle unstabilized d.c. is obtained frOf;} a bridge.
current drawn. rectifier circuit and a reservoir capacitor of 3300 JJ.F.
Mains transfonner, windings Fuses blown. Low resistance The trallsfomler has a secondary voltage of 12 V
shorting to frame or screen between windings and earth. r.m.s. so the unstabilized voltage across C 1 will be
----~----.--------------
One diode in bridge open Circuit behaves as a half- approximately 12"\1'2 V, i.e. about 16 V.
circuit wave reet.iCier. Lower d.<=: TIle reference voltage is provided by a 5·6 V zener
output W1U,1 poor regulatIOn. diode' a 400 mW device such as a BZY88 C5V6 ' .
Increased ripple at 50 Hz noL.i_ . ' . ' . IS
100 Hz as sJlould be ttle ca"e. IdeaL Tr 1 IS the d.c, error ampbfier WhICh compares
----~-------------- a portion of the d.c. output voltage, the voltage
diode in bridge short Maim: fuse blown, since
circuit secondary winding will be across ~, with the reference. Any difference between
practically shorted every other tIle two voltages is amplified by Trl an~ the amplified
half cycle. A resistance check signal is fed to the base of Tr:z. Consider for example
across e~cl1 arm of bridge is
required, mea..~ing the the case when the d.c. output falls wheIl more load
resistance of each diode in current is taken; the base voltage of Tr1 decreases
the forward and reverse and TrJ conducts less current. Therefore TrJ collector
direction.
voltage riseS, and this rise in voltage is coupled
Reservoir capacitor <>pen Low d.c. output with very through Tr:z. which acts as an emitter fdlower to
circuit high values of a.c. ripple on
output. counteract the original faU in output. TIluS the
circuit operates to maintain the output as nearly
Re.~ervoir capacitor short Fuses blown. D.C. resistance
circuit orun~tabilized line low in constant as possible.
both direetions. Since th€' n'Jtput load current is only 100 rnA,
Error amplifier in regulator High d.c. output that is then Trz (BFY51) need not be mounted on a heat
open circuit unregulated. No control sink. 111e circuit is not provided with a current limit,
signal for the series element. so should an excessive current be drawn. say by
Series transistor open Zero d.c. OlltpU.t. TIle un shorting the output, then Tr2 would undoubtedly
circuit base emitter $tabilized d.c. will be slightly
higher than normal since no bum out. A current limit can be added if required as
current is drawn. shown in Fig. 3.8, but for the purposes I)f the exercise
Reference zener short Low d.c. output. Possibility we shall assume no current limit.
circuit of l'cries transistor TIle normal d.c. voltages measured W'.th a standard
overheating. multirange meter are as follows:
Test point 2 3 4
Voltage 13 5·8 12·2
First let's consider the following fault condition:
11~.5 I~ I~ I
TP
v 17·5
The d.c. output is zero, but the unstabiJized input
to the regulator has risen, indicating that little curren(
29 POWEFl SUPPLY CIRCUITS
L 0 S;]]tChT;
500mA :: !
+
240 V
50 Hz
r.m.s::
/I
l,IB
II ( I
/I \ I
I
" I
3300p.f
L-------~~--~~--~----------~__o OV
Fig. 3.13 II. 12 V 100 mil. power lIupply

Transformer: primary 0-240 V a.c.
secondary 12 V 250 rnA
R.S. type 196-303 or simil:.;r
Rectifiers IN4001
is being drawn. Also TP2 is at the same voltage as

TPI. ntis further shows that no current at all is
flowing through Rl into the base ofTr2' The only
possible fault is that. Tr2 has an open circuit base
emitter junction. Note that if RJ were open circuit
TP2 would be at zero volts ..
Consider a fault condition when all test points are
at zero volts. Further inspection shows that the fuse
has blown. Resistallce checks give the primary resis
tance as 43Q., the secondary as 4Q., but TPJ to 0 V
is zero ohms. TIle·fault in this case can only be C 1
short circuit.
Questions TIle following table lists a series of
fault conditions; in each case state which component
or components could cause the fault and give a
supporting reason.
1 2 3 4 Other symptoms
Fault
A 16 15 14-5 ]4·5
B 11 6 4·8 5 Increased ripple
C 16 15 5·8 14·5 Poor regulation
D 17·5 0 0 0
E 16·5 2·1 0 1·5
F 17·5 17·5 0 0
G 16 7·5 5·8 7
H 16 5·9 5-9 5·2 Poor regulation
Tr,,2N3055 A~ln
R,
lk
+
C,
6800f'F
240 Vr.m.s.
50 Hz
Fig. 3.14 10 V 1 A stabilized power supply with current

limit voltage developed across Rs, caused by the load
Transformer RS207-267
Rectifiers IN5401 ourent passing through it, is applied via Rs (a base
current limiting resistor) to Trs. If the load current
is sufficiently high to cause Trs base voltage to exceed
3.B -Exercise: Stabilized Power Supply with Current
about 600 m V, then Tr s begins to conduct and it
draws current away from Tr3 base, thm: causing the
limit Circuit (Fig. 3.14)
series elemenj to limit its conduction_ 111e output can
This unit is designed to the following specification: be short circuited and the load current will be
D.C. output: 10 to 15 V adjustable, at 1 A.
limited to about 1 A. TIle actual value of load current
Current limit: adjustable from 500 mAo
limiting is determined by the setting of RV 1 • The
Output ripple: 20 mV pk-pk on full load_
output characteristics are shown with the circuit
Load regulation: better than 1%.
diagram.
Output resistance: 0-1 n
An over-voltage crowbar protection drcuit can
In this circliit the error amplifier is formed by Trl
be included as shown in Fig. 3.9. If the output voltage
and Trl which are wired as a differential amplifier. exceeds the ";;::lue of zener ZD2 the thY;'istor is
TIlis gives improved stabilization of output voltage triggered into conduction and this short circuits the
against changes in ambient temperature. TIle reference output voltage bringing the current limit into
element, the 5·6 V zener is connected to Trl base, operation,
while a portion of the output is applied to Tr2 base. The normal voltages measured at various test
Any difference between th~e two levels is amplified points are as follows (RV2 is adjusted tl) give an
and fed to the base ofTr3 and so to Tr4 to control output of 10 V).
the output level. Tr3 and Tr4 are wired as adarling
ton amplifier to give high gain round jhe circuit. To
maintain a safe junction temperature for Tr4, at ~ 11!.515~9 !1~-911~ 15~91
ambient temperatures up to 50°C, this transistor Suppose we had the follOwing fault (:onditions:
should be mounted on a heat sink which has a thermal
resistance of not greater than 10°C per watt. A
suitable type is RS149.
~1) II!.211~'11 ]~-111:.719~21
TIle current limiting is provided by the circuit TPl has fallen slightly, indicating that more currell!
comprising Rs, RV 1, Rs and Tr s. A portion of the is taken, while all other voltages have rben much
31 POWER SUPPLY CI RCUITS
higher. If the zener were working correctly TP2 would

be 5:9 V; therefore the fault must be an open circuit
zener diode. Under these conditions the error ampli·
fier loses control over the out.put since Trl conducts
heavily and Tr2 ceases conducting. Thus base
voltage of Tr3, and the output, must rise.
Questions With each of the following fault

conditions, identify the faulty component and the
type of fault, with a suitable explanation.
1 2 3 4 5 Additional symptoms
Fault
A 19·3 0 10·4 8·6 4·2
B 19·5 5·9 19·1 16·7 0
C 20 5·9 20 0 0
D 29 5·9 0 0 0
E 19·5 5·9 11·9 10 5·9 Current limit will not operate
F 19·5 5·9 18·9 17-4 9·1 Poor regulation
G 19·3 5·9 6·1 18·6 6 Poor regulation
H 0 0 0 0 0 LP I glowing
I 0 0 0 0 0 FS I fuse blown. Resistance of
primary 48 Q. Resistance of
secondary 6 Q. Resistance TPl to
ground + ve prod on TPI 6800 Q
-
I
!!
+IOV
I)Z,
8ZX61 CIO
Fig. 3.15 Switching mode regulator, 20 V at 2 A

techniques involved in switched power units. TIle
Transformer: primary 0-240 V
operation can best be understood by referring to
secondary 25 V T.m.s. @ 3 A
R.S. 196-145
Fig. 3.15. The various operations are performed by
Bridge rectifier: R.S. 261-457
the following transistors and components:
Square wave oscillator Trl and Trl fr-!q. approx.
5kHz
Duty cycle control Tr3 and Trll
3.9 Exercise: Switching mode power supply Comparator Trs and Tr6
(Fig. 3.15) Reference DZ 2 a BCY88 7·5 V zener
Switch drive Tr7 and Trs
This unit, designed to give a stable d.c. output of
Series switch Try
20 Vat 2·5 A uses secondary switching. This mode
of operation was chosen for two reasons: firstly the An unstabilized d.c. voltage of approximately
components reqUired are fairly easy to obtain and 35 V is deve1<:";,,:,d across C 1 by the bridge rectifier.
relatively cheap and secondly the resulting circuit The transformer has therefore a secondary voltage
is not too complex. Primary switching regulators, of 25 V r.m.oS. and a rating of about 100 VA.
although being superior, require relatively expensive The unstabilized voltage is switched at a frequency
high voltage switching transistors, and a specially of about 5 kllz by Tr9' The switching signal is sup·
wound high frequency transformer. plied via Trs, Tr7 from the astable multivibrator
In this design 'the switching transistor is a BD132; formed by Trl and Tr2' The duty cycle, or mark/
this, although being desjgnat~d an audio transistor, space ratio, of this astable is controlled by the con
has an IT of 60 MHz and can pass a m:lxinlum duction of Tr3 and Tr4. A relatively low frequency is
~~ current of 3 A. If a higher output current is required chosen m;linly for demonstration of openltion, since
i a transistor such as a TIP 2955 could be used; this tIle signal is just audible when the circuit is operating.
~. has a maximum collector current of 15 A. TIle DecreaSing C] and C4 will increase the frequency to
regulator is designed to give an introduction to the above the audio range.
UU
33 POWER SUPPLY CIRCUITS
TIle switching signal duty cycle is changed by a

d.c. level that is the amplified difference between the
reference voltage(DZz) and a portion of the d.c.
output. Imagine that the output voltage falls, caused
"OV J
say by an increase in load current. 111is faU is fed to
ov --------------------------
Trs base via R9 and RV t • Trs and Tr6 form a I• ., ,......
1601'S 'lOftS
differential amplifier and when the base voltage of
Trs falls, Tr5 conducts less and Tr6 conducts more. Fig. 3.16 Switching wave form at Trl collector when a hea\'}'
This is because Tr6 base voltage is held constant by load is connected to the SMPU output .
the refer.ence voltage. TIle outputs ofTrs and Tr6 are
connected to Trs and Tr4 base. TIlerefore TrJ tends Questions
to tum off and Tr4 on. Now these transistors control
the discharge time of the astable timing capacitors (1) In the following faults the power unit fails so
C3 and C4 . TIlUS with Tr3 conducting less,equivalent that the output at TP2 remains zero after the start
to a higher resistance, and Tr4 conducting more, button is pressed. The oscillator, however, is observed
equivalent to a low resistance, Trl is held off for a to start. A resistance check gave the following results.
longer period, and Trz for a shorter period. The In each casejhe positive prod of the ohmmeter was
collector wavefonn at Trl collector is thus as shown connected to the measured point and the resistance
in Fig. 3.16, a signal that is positive for a longer measured with respect to chassis. State, ''lfith a
period than it is negative. Therefore Tr9 receives a supporting reason, tlle faulty component or
switching signal that turns it on for a longer period components.
of the switching cyc1e; this tends to counteract the
original fali in output level. 1 2 5 6 Tram. TraIlS
Since the series transistor is switched on and off prim sec.
continuously its collector dissipation is quite low Rl'-s.
and only a small heat sink say 5 cm 2 is required. 'nle A 7kn 10 n 160kn 12n 33 n 2n
switched output from Trg is filtered by L\C2. Durin& B 7kn 10 n 7kn 500kn 33 n 2n
C 7kn 10 n 7kn 12 n 33 n '2n
the OFF period for Tr9, D4 and RI7 minimize the .
collector leakage current, and D, transfers the stored
(2) Sketch the time related waveforms you would
·energy in Ll to the load. Therefore both D4 and Ds
expect to measure using a dOUble beam CRO at Tr2
must have a peak current rating equal to the maximum
and Tra collectors, for the conditions of low output
coUector current of TrJ.
load cutrent. Assume that the CRO is triggered from
In order to start the circuit a switch is included so
the Tr2 wavefoml on the Y 1 beam.
that power can be supplied from the unstabilized
output to the muItivibrator, however once the (3) State the symptoms for the following faults:
circuit is running the ~tabilized d.c. output is used
(a) C) short circuit.
to power the multivibrator.
(b) Tr8 base emitte! short circuit.
111e normal working voltages for a load of 2 A are
as follows:
(c) Tr4 base emitter .open circuit.
(d) R16 open circuit.
TP 2 3 4 5 (e) R14 open circuit.
MR 20 OSC OSC OSC (4) TIle power unit fails so that the fuse is blown. A
d.c. resistance check gave the following results. State,
9 with a supporting reason, the probable fault.
6 7
OSC 7·2 5
1'1' 1 2 5 6 TrailS. Trat:
prim. sec
Res. 7 kn 10 n 7 kn 12n 33 n 2!
4 Amplifier Circuits
4.1 Types and Classes of Amplifier

r-------'---i +
Since there are so many different types of amplifier Power
used in electronics, before going into circuit details supply
Lood
it is a good idea to define what is meant by the term resistor
amplifier. An amplifier is any device where a small

input signal is used to control a larger output power.
It follows from this that an amplifier must consist of
ii Active
some active device, such as a valve or transistor; a
source of d.c. power; and a load resistor. This is
shown in Fig. 4.1 A. Here the inpu t signal is used to :Vi ot_ _ _ _ _d.....ev.....ic_e_ _----l
control the current that flows through tbe active
device. lhis current then develops a voltage change
Fig. 4.1 A Basic amplifier circuit
across the load resistor, so that the outpu~ power is
Po = Voio watts
The input power Pi = "iii watts. P
POWER GAIN, or power amplification, is given tV 0 - - - '
by the ratio of output to input power:
A =Po
p Pi Fig. 4.1 B General amplifier symbol
Power gain =: Po/Pi
A more cQmmon symbol for the amplifier is shown

in Fig. 4.1 B, the signal flow being in the direction of of a voltage amplifier is 2 V peak when its input is
the arrow. 100 mV peak, then voltage gain is
Any amplifier increases the power con tent of its 2V 2 V
input signal but this is not always the chief con fOO~l~lV = 0.) V = 20
sideration. An amplifier may be designed to give- .

primarily voltage gain, current gain or power gain. Often the figures involved in working with
TIluS the first classification of amplifiers is one which amplifier gains as ratios can become unwieldy. This
divides them into types designed primarily for power, is the case when an amplifier exhibits large changes
voltage or current amplification. We have already of gain with signal frequency and when these changes
seen that power gain Ap = Po/Pi. So it follows that have to be recorded graphically. For this reason a
logaritllmic unit for gain is often used. This is called
VOLTAGE gain AI1 == ~~ the Be1.
PQ'.Ner gain Ap =lOgJO(~;) BeJs

CURRENT gain Ai == i;>
Ii A Bel is usually too large a unit for elxtronic
It is important to realize that these are all expres measurements, so tenths ofa Bel or decibels (dB) are
sions of gain as ratios. In other words if the output commonly used. TIlen
34
35 AMPLIFIER CI FlCUITS
----t~----I'---- + vee
Ap = 10 lOglO(~~) dB
By using decibels for gain units, very large changes
in gain ratios can be compressed. The figures are much
d.c.
easier to handle. This can be seen from the follow
ing table.
Power gain as ratio Power gain in dB

- --_._-_._---
10 10 dB
100 20 dB
1000 30 dB
10000 40 dB
100000 50 dB
OV
1000000 60 dB
Voltage a!1d current gains can also be expressed in ib signaicurre·

dB, as follows:
All = 20 loglO(~ ) dB
ie .. Bias
level
\
, ;
,I
"--,, ' '
~-
it: : h'e itt
'-'
Ai = 20 logl~~; )dB
,, ,
I
,
\ ,
TItis is only strictly true if tlle amplifier has equal Bias I
I
\
I
input and output resistance. This is rarely the case, "0 level
\
, I
I \
,, I
I
v. V..- tcRl
\
,, I I
,
I
, I
but it is often assumed.
TIle reason why the multiplier 20 is used can be
'
'
I
seen from ~he following:

Fig. 4.2 Diagram to explain how inversion takes place
. -!JJ.- (VO/RO)2 between output and input of a single stage common emitter
Power gam - Pi - (Vi/Ri)2 amplifier
where Ro = output resistance and Ri = input resistance.

If Ro = Rj, then Ap = (Vo/vii. TIlerefore
tJle fact that, as the input voltage increases, the
Voltage gain = 10 log (~;r dB current flowing into the base increases, which in
creases the collector current. This current flows
Now logx'l =11 logx through the collector load and therefore the collector
voltage falls_ At high signal frequencies, the phase
Voltage gain 2)( 10 log (X£..) dB
\Vi
0
shift does not remain exactly at 180 ; this is because
(a) the current carriers in tlle transistor take a finite
As well as possessing gain, any amplifier also has time to reach the collector region and (b) reactive
PI·IASE SHIFT between its output and input. For components in the circuit produce additional phase
example at low frequl!llcies a single stage common shifts. Phase shifts in an amplifier at higll frequencies
emitter transistor amplifier produces an output lead to phase distortion and possibly instability if a
signal which is an inversi.on of its input. 11us is negative feedback loop is used. This will be seen later
shown in Fig. 4.2, and is explained quite simply by when negative feedback is discussed.
36 ELECTRONIC FAULT DIAGNOSiS
TABLE 4.1 Broad Classification of Circuits
Frequency response Go.<s ofoperotiOIl ond tJ'picoluse

Voltage. Audio and low frequcncy (lass A Small signal voltage and
Current Radio frequeue}' (tuned) current amplil1ers
or Wideband or video
powcr Pulse (.lass II - I'o,..,.er output amplifiers
Oirect current
Class C - Transmitters and pulse
switches
A further classification of amplifiers can be made Gain .t\

dB
,y considering the RANGE of signal frequency over
vhich the amplifier has useful gain. An audio fre
tuency amplifier, for example, should amplify signals
.ver the range from 15 Hz up to perhaps 20 kHz. A
-'''! \
:~ _ _ _ JJond!.~~~.~ ____._._:
I I
raph of amplifier gain against signal frequency is
alled a frequency response curve. A typical frequency
I
!sponse curve for an audio amplifier is shown in ~--:7-T--::!=:---::--~----L---L.!.--+-
ig. 4.3. Note that gain is usually plotted in dB on the
10 ! 100 1k 10 It 100 kit MHz Ftequency
~ in Hetll
i
I
~rtical axis and frequency on the horizontal axis.
J ,
fj f~
~ requency is plotted logarithmically so that a large
j mge can be accommodated.

Fig. 4.3 Typical frequency response CUIVe
~ 11le gain of any amplifier will change because of
i Ie reactive components in its coupling and decoupl
t g circuits, stray circuit capacitance and inductance,

TIlere are baSically three classes of oper<.tion.
4 ld because of the frequency limitations of the
Class A The active device (the transistor or valve) is
tive devices used.
biased so that a mean current flov.'S all the
TIle BANDWIDTH of an amplifier is usually
time. This current is either increased or
:fined as the range of frequencies over which the decreased about this mean value by the
, in has not fallen by more than 3 dB from its mid input signal. TIus is the most commonly used
i ~
~quency gain. If the response is flat this is equivalent dass, typical examples being small signal
! 50% of maximum gain for power amplifiers (half amplifiers.
·wer points) and 70·7% of maximum gain for current Class B TIle active device is biased just to eut off
voltage amplifiers. From Fig. 4.3 it can be seen that and is switched into conduction by one half
~ bandwidth = f2 - f1 . cycle of the input signal. This c1asf. of opera
Amplifiers can therefore be classified as tion is widely used in push'pul! power output
(a) Audio frequency (AF or LF) amplifiers.
(b) Radio frequency (RF). Tuned with narrow Class C TIle active device is biased beyond the point
~
~ bandwidth
of cut-off so the input signal must exceed
~ (c) Wideband or video.
a relatively high value before the device can
• (d) D.C. amplifiers. be made to conduct. This class is used in
, TIle basic response curves for these types are shown
I Fig. 4.4. For a d.c. amplifier, the active devices
~ 1St be directly coupled so special techniques are
pulse switching and transmitter circuits.
The preceding discussion of amplifier c.!,:tssifica
tion can be rather confusing, but it is impo:tant
. luired to ensure correct biasing. This is discussed before attcmpting any repair work 011 an awplifier
, _ ~r. that the type of amplifier and its purpose i! fully
TIlere is yet a further classification to deal with, appreciated. The information in the last few para
~ nely the CLASS OF OPERATION and the intended graphs is gathered together ill Table 4.1 that scts out
'~; : of the amplifier. the main types of amplifiers in use.
;·f
37 AMPLIFIER CIRCUITS
Gain f.i1 X 1+\
Low frequency or audio Wide bond Amplifier with voi1oge
____-A-__~or t
L_+:::::===~ video
Vi v,t f+\ goin Ao Vo
f
\,. \ )
0-
Vf
-'.'.
1
In'
Y
RF
'V
P
Frequency
Fig. 4.4 Various amplifier response cmves

0
REO
Feedback
network ~
4.2 Negative Feedback Fig. 4.5 BI()Ck diagram of amplifier with negative feedback
No discussion on amplifiers can be complete without

mentioning negative feedback. The study of this
subject can be very involved, but the basic principles The loop referred to is the connection between
are not too difficult to grasp. the output to the input via the feedback network (I
An amplifier is said to have negative feedback TIle product Aof3 is called the LOOP GAIN. It is
when a portion of the output signal is fed back to the gain of the circuit from the amplifier input
oppose the input signal. TIlis is shown in Fig. 4.5. tenninals X-Y to the feedback terminals P-Q. Now
Here an amplifier with a gain Ao has a portioll of if the loop gain is much greater than unity, the gain
its output signal Vo , fed back in series with tlte of the amplifier with negative feedback can be re
input in such a way that it opposes the input. TIle written as follows:
feedback circuit has a fractional gain of {l Therefore
A !,:>!Ao
the feedback signal is given by c. Aofl
Vf= J3Vo 0) (The 1 in the denominator can be ignored since it is

The input signal to the circuit is small compared with Ao(3) Therefore
Vi=: Vs + Vf (2) A !':>!!
c IJ
But substituting (1) in (2) we get
l1lis is an important result since it means that the
Vi V$ + {3Vo gain is now dependent only on the characteristics of
Now the gain of the amplifier is Ao == Vo/Vs. Thus the feedback circuit. Thus if the feedback network
is a potential divider made by two resistors as shown
Vo=AoVs
in Fig. 4.5 then the amplifier gain is
TIle gain of the whole circuit. which we shall call
Ac,. is given by
The gain is therefore given by a ratio of two

resistors, and is independent of changes in circuit
components, such as .current gain changes in transis
Le. tors. This holds true as long as the loop gain AofJ~ 1.
Negative feedback is widely used for the following
The gain of the amplifier without feedback is called reasons:
the OPEN LOOP GAIN Ao. . (a) It stabilizes the gain of the circuit, making the
The gain of the amplifier circuit with negative gain independent of changes due to com·
feedback is called the CLOSED LOOP GAIN Ac· ponents, temperature and power supply lines.
(b) TIle frequency response is improved, and the the amplifier introduce additional phase s:lifts, tbis
bandwidth widened. TItis can be seen from in tum changing the phase of the feedback signal.
Fig. 4.6. At some frequency the total phase shifts ',vi.ll be
(c) The way in which the feedback Signal is such that the feedback signal is adding to input not
derived from tile output and applied to the opposing it. TIle result is tl1at the circuit will oscillate.
input can be used to modify the input and 1llis situation can be avoided by ensuring that the
output impedance of the circuit. loop gain (Ao(}) is less than unity when the total phase
(d) Non-linear distortion and internally generated shift round the loop is such as to produce positive
noise in the amplifier is reduced. instead of negative feedback. This is why op-amps
TIlese reasons show that a manufacturer making, such as the 709 have to have a frequency:ompensa·
y. wide band linear amplifiers, can by using negative tion circuit to limit the bandwidth. For ttis reason
edback ensure that every amplifier produced has also, direct coupling is often used since it eliminates
:arly the same characteristics. phase shifts due to coupling capacitor. Fig. 4.8 shows
- t - - - - - < p - - - - - - - < r - - - - - - 1 P - - - - - Vee
Bandwidth without
feedback
Output
IIIPut
Bandwidth with feedback
Frequency
~4.6 The effect of negative feedback on the bandwidth

an amplifier
Fig. 4.7 Two-stage amplifier with negative feedback (a.c.

coupled)
A typical example of a two-stage amplifier using
- - - . . . . - - - - - - - - . . ,....- - 0 \Icc
gative feedback is shown in Fig. 4.7. A feedback
istor RF is connected from the output to the
litter of the first stage. The overall gain of the
]
cuit \¥ill be given by
Oulput
RF+RE o-J I'J------.--+---
i.e: a gain of 30
RE
This type of feedback is said to be shunt-derived.
ce the feedback network is in parallel with the
tput load, and series-applied since the feedback
tage VI is effectively in series with the input. TIle
tput impedance is reduced, while the input
pedance is increased. Fig. 4.8 Two-stage amplifier with negative feedback (d.c.
One of the problems associated with negative coupled)
dback is concerned 'with phase shifts round the the amplifier of Fig. 4.7 Ilsing direct coupling. An
'p. In a two·stage voltage amplifier the output is additional feedback path is provided to stJbiiize the
I ;>hase with the input; this is why in Fig. 4.7 the d.c. operating point via Rs.
i dback is applied to the emitter and not the base_ Feedback can, of course, result from [,mit con·
I higher frequencies the reactive components within ditions when for example a decoupling ca~acitor
I
39 AMPLI FI ER CI RCUITS
goes open circuit. In this case the gain will be reduced -f. -
drastically.
A
t t
Soqnol
Testing amplifiers witJl negative feedback loops is qeNef'Olot'
Voriable
Amplifier
10 Hl 10 1MHz ol1,,;nuolof A
discussed in the next section. uodef tesl
sine wove 01060 d8
4.3 Testing Amplifiers: Basic Measurements Fig. 4.9 Laboratory set-up to measure the voltage gain of an
amplifier
TIle various tests that should be made on an amplifier
system obviously depend upon the type of circuit
under consideration. The basic measurements that oscilloscope controls as set, the attenuation is
should be made are those of gain, frequency response, increased until the output is exactly the same height
and bandwidth. In addition it may be necessary to as with the first measurement. The gain of the
measure the input and output impedance, the maxi· amplifier is now equal to the setting of the switched
mum power output, and the efficiency. The latter attenuator. The advantage of this method is that the
would apply only t6 power output stages. All of measurerllent does not depend \lpOn the accuracy of
these measurements can be made, with reasonable the oscilloscope. If the variable attenuator has
accuracy, using the instruments listed below: switched ranges down to o·} dB, then the result will
(a) Stabilized power supply be obtained to within ± 0-1 dB.
(b) 20 kil/V multi range meter
(e) Signal generator with sine and square wave Measurement of Frequency Response and Bandwidth
output _ Using the same set·up as in Fig. 4.9, the gain of the
(d) Variable attenuator, calibrated in dB.
amplifier can be found at any frequency. The gain, in
Ie) Oscilloscope.
dB, is then plotted against a frequency on linear/log
For tests to measure distortion, noise, stability
graph paper. For an audio alpplifier 4 cycles of log
and pulse response more specialized equipment is would be required to cover the frequency range
required, which may include 10 Hz to 100 kHz.
(a) Distortion meter 11le bandwidth can be qUickly determined by
(b) Noise measuring set noting the two frequencies at which the gain falls by
(c) Spectrum analyser 3 dB from the mid-frequency gain.
(d) Phase meter
(e) Function generator. Measurement of Input Impedance
It is beyond the scope of this book to detail the TIle input circuit of an amplifier c~n be represented
highly specialized tests, but the following is intended by a resistor in parallel with a low·value capacitor.
as a guide to basic measurements. At low frequencies the input impedance is mostly
resistive since the reactance of the capacitor is stich
a high value. A circuit for measuring input impedance
Mea:>'<Irement of Gain
at 1 kHz is shown in Fig. 4.10. A variable resistor,
TIle layout of the measuring circuit is shown in
usually a decade resistance box, is connected between
Fig. 4.9. Suppose the amplifier's voltage gain at a
the signal generator and the amplifier input. Tlus
frequency of 1 kHz is reqUired. First the signal
generator is set to give an output of say 500 mVat
I kHz, with the attenuator switched to zero dB. TIus
signal, at the amplifier input (point A), is connected·
to the V-input of the oscilloscope and the oscilloscope
controls are adjusted so tila! the trace displayed uses
a large portion of the screen and has its peaks just on
graticule lines. TIle oscilloscope leads are then moved Fig. 4.10 Measurement of the input impedance of an audio
to the amplifier output (point B) and, leaVing the voltage anlplifier
!l0 ELECTRONIC FAULT DIAGNOSIS
:"esistor is set to zero and the amplifier output is harmonic distortion. 'TIlis may be a value of total
;onnected to the measuring instrument, an oscillo· harmonic distortion of 0·05% of the outpH signal.
:cope or a.c. meter. The controls are set so that a V2
arge deflection is indicated. Tlte resistallce of the I)ower ou t pu t-.-!L
- R
L
lecade box is then increased until the indicated
)utput signal falls by exactly a half. Since the
wbere Vo is the r.m.s. value of the oUtI'll t signal.
'esistance box and the amplifier input impedancc
'orm a potential divider \vhen the output is halved, _ peak to peak value
he setting of the decade resistance box is equal to
Remember Lm.s. - 2.J2
he input resistance.
The efficiency of the amplifier can be checked by
measuring the d.c. power taken by the amplifier
from the supply.
?"'ag. 4.11 Measurement of output impedance of a voltag~ and

Implifier
. r.moS. output pc.wer 100 m
Power effiIClency = . x 70
d.c. mput power
Aeasurement of Output Resistance
Jle circuit shown in Fig. 4.11 is used for this 'TIle sensitivity of the amplifier is the input voltage
neasurement. The tectmique is similar to that of required at the input to produce maximum undis
neasuring the input impedance. A signal frequency torted output power.
If 1 kHz is used and initially RL is disconnected lind
large deflection obtained on the oscilloscope. The
xtemalload Ri is then connected and reduced in
alue until the output falls by exactly a half. 'TIle
alue of R L at which tIlis occurs is equal to the
esistance. ~~\7:~.r.m.'
Y CRO
"easurement of Power Output, Efficiency, and 07
'ensitivity for an Audio Amplifier
~Qr these measurements the loudspeaker should be Fig. 4.12 Measurement ofpO\ver output, effici-ency and
eplaced by a wire·wound load resistor of the same sensitivity of an audio output stage.
RL is a wire-wound resistor of the same value as the loud
alue as the loudspeaker impedance, and the tests spenk"r impedance.
hould be carried out a~ a frequency where the

)udspeaker impedance would be mostly resistive,
ypicaUyabout 1 kHz. The diagram for the measure· 4.4 Transient Testing of Amplifiers
nent is shown in Fig. 4.12. 'TIle WlI ttage rating of
he load resistor should be higher than that ortbe All the tests previously described are mac;e using an
naximum output power. The input voltage should input signal at one frequency. By applying pulses or
Ie adjusted until the output Signal indicated by the square waves to an amplifier it is possible to acquire
,scilloscope is a maximum undistorted leveL 'TIlis information about the amplifiers frequen:::y response,
; when there is no clipping of the positive and phase distortion, and any tendency to instability.
legative excursions of the output signal. Naturally A square wave is made up of a series of pure sine
f a distortion meter is available then a more accurate wave components, which are a fundamental, having
heck on distortion levels can be madc. Then the the same periodic time as the square wave, and all odd
naximum output power should be recorded without harmonics. Thus by applying a square wa-"e or pulse
:xceeding the manufacturers speclJlcd value of to an ainplifier, a large range of signals at different
41 .6.MPLIFIER CIRCUITS
frequencies have to be amplified by the same ratio

and without phase shift if the output is to be a per
fect replica of the input.
For te."iting low frequency amplifiers, a square
Line pulse Line pUlse
wave of 40 liz or I kllz is suitable and the output
signal can be observed on an oscilloscopc. Departure
from squareness in the output signal gives a good
indication of the transient distortion that is present
in the amplifier. Variolls conditions are shown in (I)
Fig. 4.13.
Video and wideband amplifiers can also be usefully
tested in this way> but usually a special form of signal
called a pulse and bar i.~ used. this, together with
various possible outputs, are shown in Fig. 4.14.
(1) Square wave input signal
Possible outputs (3)
(2) A mplifier deficient in

low frequency gain. No
phase error. Fig. 4.14 Pulse and bar testing of video amplifiers
(J) Lack of high frequency gain
(2) Excessive high frequency gain
(3) Amplifier deficient in I.f. (3) Poor low frequency response
gain with phase error.
4.5 Distortion Measurements
(4) Amplifier wifh excessive Various types of distortion can effect the shape of
If. gain. No phase error.
Ule output signal from an amplifier.
Amplitude Distortion
(5) Amplifier with excessive
l.f. gain and phase error. The output signal is flattened on one or both of its
peaks as shown in Fig. 4.15. This type of distortion
occurs when the amplifier is overdriven by an
excessively large input signal, or when the bias
conditions change, or because of some non-linearity
J~
(6) Amplifier wi.th poor
11igh freqllency Tl'Sponse in the characteristics of a transistor or valve.
and phase error.
h +~
I
~ (7) Amplifier with excessive

Inpu!
sigool
~ ~ Ju,. freq"'''', g"o"
(mV)
Fig. 4.13 Square wave testing of an amplifier Oulput +

signol
(Volts)
N . .".. 10islorlion
'0.
Mox - 'Ie swing
Fig. 4.15 Amplitude distortion resulting from an overdri.'1en

amplifier
42 ELECTRONIC FAULT OIAGNOSIS
Frequency Distortion bias components is to overcome the distOttion by
ll1is results when the amplifier gain changes drasti providing a very small amount of forward bias.
cally with frequency within its passband. Suppose

all amplifier has a frequency response as shown in Intermodulation Distortion
Fig. 4.16, which is reasonably flat over the passband, When non-linearity exists in an amplifier circuit, two
but that the actual response is as SI10\'1111 in Fig. 4.16B, signals of different frequencies, say 400 Hz and
then the amplifier is said to have frequency distor 1 kHz, as well as being amplified will be mixed, and
tion. 111is can take the form of loss of gain at low or so the'butput will contain small amplitude signals
jigh frequencies or increase of gain at low or high of the sum and difference frequencies, i.e. at 600 Hz
frequencies. and 1·6 ~Hz and harmonics of these frequencies.
Measurement of distortion levels is usually made
~:!." "'' l r~
using a distortion meter, an instrument which sums
tile power in all the harmonics and gives the result
response . \ as a percentage of the output power. 111is gives the
value of the lotalizannonic distortion resulting from
~ amplitude and non-linear distortion, but does not
Frequency
. include frequency, phase or intermodulation dis
Goin tortion. A frequency of 1 kHz is nonnally used for
(b) this measurew~n t.
Response Total harmonic distortion can also be measured
with frequency
distortion bY passing the 'output voltage signal through a ftIter
which attenuates the measurement frequency (I kHz)
Frequency
but passes all harmonics. A good circuit for this is a
twin-tee ftIter as shown in Fig. 4.18 since Uus has
1ig. 4.16 Frequency distortion
maximum attenuaqon at one frequency. 111e output
can be measured using a sensitive r.Ill.S. mJlivolt
'base Distortion meter.
\s the signal frequency is increased so the phase of c c
he output signal relative to the input will change.
11is type of distortion is troublesome when the input
ignal is a complex waveform, made up of several
ine wave com ponents all at different frequencies. If
:lese all suffer different phase shifts through the
mplifier, the resulting output willilot be identical
1 shape to the input. Fig. 4.18 Twin-tee filter
"ross-over Distortion Intermodulation distortion can be measured by

ijstortio!l of this type occurs in class B push-pull feeding two signals of 400 Hz and 1 kHz into the
utput stages. In a complementary transistor output amplifier usually with a ratio of about 4: 1. Then
:age for example, unless some fprward bias is applied, using a filter at I kHz the result o(any intennodula
le transistors win not conduct until the input signal tion will be indicated using the method detailed
) their bases exceeds about 500 mV (this is for a previously.
licon transistor ). See Fig. 4.17. TIle purpose of the A method tllat can be used to display @nplitude ..
distortion, phase shift distortion and harmonic dis
tortion for an audio amplifier is shown in Fig. 4.19.
TIle Signal generator set to 1 kHz is fed to the amp
lifier input at a suitably low level and to the X-input
Fig. 4.17 Cross-over of the oscilloscope. 11le output from the ampl.ifier
distortion is fcd to the Y-input of the oscilloscope. 111C Qscillo
43 AMPLIFIER CIRCUITS
scope trace will be a straight line at an angle of 45° its nominal 82 kil to say 150 kil the operating
if tJIC amplifier output is undistorted. Naturally a point will now rise to approximately 8 V. The out
high-quality oscilloscope must be used for this tcst, put signal is now distorted on its positive excursion
since any non-linearity in the X and Yoscilloscope as shown in Fig. 4.20B.
amplifiers will also be displayed. Various outputs For fault finding on amplifier systems it is best
for different types of distortion are shown in Fig. to follow the standard procedure and inject a signal
4.19 also. into the input, and by using an a.c. meter or an
oscilloscope, check each stage in tum until the
(0) Normal
AF oscillolor
1kHz
0
No dislorlion
([)
Odd harmonic
0
Cross - over
No phose shifl dislorlion dislorlion
+IQV - - - - --
o o
No distortion
some phose shift
Even harmonic
dislorlion
Odd harmonic
distortion with
+5VI----,
Oulpu! signOI
(b) Foul! conditions R~ nigh to 150 k
phose shiff -....,..----f--- + 10 V

Fig. 4.19 Method of displayjng distortion using aeRO
tv~:~
4.1:> Faults in Amplifiers
It would not he possible to detail all the possible

faults (hat could occur in aU the various types of
R~
amplifier circuits. Instead the follOwing is a general 22k
guide to assist in fault location. Before considering. -_6_--'_--4--4-_
some typical faults it should be noted that, as well
as the d.c. bias levels, the output signal itself is often +IOV
an invaluable guide to the type of fault. +8·3 V 1-'--____
The previolls section detailed the types of

distortion that could occur. Let's consider a simple
example of how changes in bias components can
cause large amounts of amplitude distortion. In
Fig. 4.20A the operating point at the collector is Outpul signol
about +5 V to allow equal positive and negative Fig. 4.20 Distortion resulting from a change in bias
swings at the output. If RI goes high in value from cOllditions
faulty stage is reached. Then measure the d.c. levels 4.7 Exercise: Two-stage Pre-amplifier (Fig_ 4.211
at this tage.
Specification .
Table 4.2 lists some typical faults together with
Voltage gain 34
the expected symptoms.
Input impedance lOOkS"!
Output impedance 500n
Frequency response 20 Hz to 30 kHz
TABLE4.2 Typical Faults on Amplifier Systems Sensitivity for I V Lm.S.
output 30 mV f.m.s.
Small signal amplifiers
l1~is c;rcuit, using negative feedback loops to
FAULT SYMPTOMS
stabilize both the a.c. gain and the doc. operating
Bias conlponent failure open Results in a large change in point, is a typical example of a design as discussed
circuit or high value resistors operating point usually tend·
ing10 cut transistors off. 11lis in section 4_2.
gives either gro~sly distorted A low noise high gain transistor is used for Tr I ,
output or 110 output at al1. and since series negative feedback is used the input
Short circuit decoupling or Again a largechallge in impedance is almost equal to R3 , which is 100 kil.
coupling capacitors operating point usually tend l11e a.c. gain of the circuit is determined by the
ing to f Qrce tra nsistors to
conduct much harder. negative feedback loop from Trl collector via Rs
Grossly distorted output. to the emitter of Trl and R 2 • Since Rs and Rl form
Coupling capacitors No tran~fer of sigll31 from a simple potential divider, the feedback fraction (3 is
open circuit one stage to next. AU d.c. given by
bias levels normal. No output
signal. R').
Signal deeoupling capacitors Low gain, since series negative P=R 2 +Rs
open circuit feedback is introduced.
Now if the loop gain A Q {3 is much greHter than
Power line decoupling Increas~ in "hum" level unity, as is the case in this circuit, the gain with
capacitors open circuit (lOO Hz) at amplifier output
lbe rust stage of 3 pre- ..• feedback is given by
amplifier is normally supplied
from a decoupled line. A. ~.!.=R2+Rs=34
----------------------
Open circuit feedback Excessive. gain with instability
C (3 R').
line and possibly oscillation. When you build this circuit, you can easily verify
---~----------- ...--.- .. _ - - -
Noisy transistor or Poor sigmlHo-noise ratio. by measurement that the gain is 34, and the open
resistor.at input (Always chedc early stages loop gain can be measurcd by decoupling R2 with a
fust.)
large-value capacitor. This reduces the feedback
Change in coupling and Reduction in bandwidth. signal to zero.
decQul'ling capacitor values Poor low frequency response.
to lower value Since the cirwit uses direct coupling thc calcula
tion of the d.c. bias levels is relatively difficult. In
Power amplifiers
designing such a circuit one starts with tYe required
FAULT SYMPTOMS operating point at Tr2 collector. To give undistorted
Bias resistors open circuit For class II llI11plifiers, the output this should be about half the supply voltage
or high in value type in common use, there
Le. 7-5 V If we assume that ~, Tr2 colkctor load is
will be a large amoullt of
cross-over distortion_ 2 k2Q, Rs is 33 kil and R2 1 kn to give the voltage
Output capacitor short
gain required then the other component values can
Output fuses blown or
circuit lrallsist()r~
overheating. Use be calculated as follows.
resistance check to find
faulty C<1lnponcn t. Vrr- VC2 _ 7-5 _
--------.--------------Either (i) increase in cross (a) Current through R, =.:; -)"--"-R---·- _. -2·-k·7{-)- 3-4 mA
~4 -"~
Bias potentiometer
incorrectly set over dislQ;tiol1 or (ii) over-
of transistors.
45 AMPLI Fl ER CI RCUITS
_ _....-------T----<::J + 15V (n We made 10 to be 100 I1A in the design, but in
addition to this current flowing through R J there
is also the base current for Trz. Now
"-
Iez 3·18 mA
I 82 - 5311A
II FE(min) 60
l11erefore
15 3·5 11·5
15 3 11A 0.153== 75 kn
A value of 68 kn was chosen.

TIle d.c.levels at the various test points are shown
for the calculated and measured values below. A
20 kn/V meter was used.
TP 1 2 3 4 5
Fig. 4.21 Two-stage pre-amplifier Calc- 0-32 3-5 7-5 2·8 I
ulated
Vo 7·5 Meas 0-39 34 7-5 2·8 1-06
(b) Current through Rs == R + Rs = 34 kn = 0-22 mA ured
z
and the current through Trz collector 3·4 - 0-22 It is important to realize that the d.c. operating
3·18.mA point is stabilized by both feedback loops and tnat
(c) Assuming that trl current is low, say 100 I1A by any tendency for the operating point to change
design, then the voltage at Trl emitter is given by would be counteracted by more or less d.c. feedback.
C:z is included to limit the hi&ll frequency gain;
VE =R 2 (IE+IR)
above 30 kHz its reactance is less than Rs and there
= 1 kn (0·1 + O·22)mA "" 320 mV fore the feedback signal increases, thereby reducing
the overall gain.
(d) Voltage at Trl base =: Vi!' + JIBE:!!:; 1 V and if we
Let's consider the effect of some component
make the simplifying assumption that Tr 1 base
current can be neglected then the voltage at the failures on the operation of this circuit. Imagine that
the d.c. feedback loop via R3 goes open circuit. With
junction of R1 and R6 will also be I V. The value
for R, is now given since
R3 open there can be no bias current into Trl and
thus Tr 1 is cut off. Normally we would expect the
VR, 1 collector voltage of a transistor that is cut off to rise
IE'l =3-18 kn=314U to the supply voltage, but in directly coupled circuits
(330 n_p.v.) this is not possible. In this case, RJ reeds bias current
(e) By making R6 a 560 n resistor the voltage at to Tr2, so with Trl off the bias to Tr2 increases and
Tr'l emitter will be about 2·8 V. we must expect Trl coUector voltage to fall and its
emitter voltage to rise.
VEl (R 6 +R 1 )/ez TIle actual readings with R3 open circuit are:
and the base voltage ofTrz is TP 2 3 4 5.
VEl + VBE "" 2·8 + 0·7 "" +3·5 V Vn Meter 0·15 5 4·3 4·25 1·6
reading
This will be the \loltage at TIl conector which
:;hou1d aHow a reasonable voltage swing at that ft is intercoting to note that these readings also result
point. ifTr2 develops either a short circuit base emitter
junction, or if Tr:z collector is open circuit. Verify
these for yourself.
Consider now the effect ofa base emitter short on Since Trl is passing no coUector current, test
Ttl' Obviously test points (2) and (4) will give tlle points (3) and (1) will also be (htermined by the
same reading, and we would expect test point (3) to potential divider formed by ~ , Rs' and Rl _Try and
rise since Trl can no longer act as a transistor and calculate the values you would expect to measure
therefore collector current ceases. Since RI is such before looking at the next table.
1 high value in c.omparison to R.; and R, , the vol tage In fact the measured values \vith Trl t·al<e emitter
It (2) and (4) will be low. It can be qUickly calculated short are:
;ince Rh ~ and R7 wiiI fonn a: potential divider.
Test point (4) and (2) with Trl b_e. shorted is TP
MR
I 1
04
4
0-2
5
0-08
V
4
=15x(R 6 +R 7 )_15xO-89kfl -194mV'
Questions All the readings were m<lde using a
RI + R6 + R, 68·89 kfl
standard 20 kn/V multirange meter_State which
Test point (5) will be tllen about 70 mV_ component or components could cause the symptoms
ami give details of the type of component fault. The
output was observed on an oscilloscope with an input
ofJO mV r.m.s. at 1 kHz.
I 1 2 3 4 5 Output signal
I :=ault
I A
B
0
0-2
0·75
2·5
0-1
9-2
0
2-3
0
0-9
Zero -output
Output 12 V pk-pk gross distortion
C 0-6 0-6 14 0-06 0-02 .Zero output signal
1 D 2·7 3-3 2-7 2-65 I Zero output Signal
i E 0-15 0-8 14 0-2 0·08 Zero output signal .
I1
'
F 0-39 3·4 7·5 2-8 1·06 Much reduced output signal
l G 0·42 0 14 0 0 , amplitude
Zero output signal
.8 Exercise: Pre-amplifier with a FET Input

=ig.4_22)
ne of the important features of Fieltl Effect Tran Another useful advantage of using a FET at the
;tors is that under normal operating conditions they input stage is that the noise generated is lel<s than
lve a very high input impedance_ This makes them that from a bipolar transistor. FETs use only one
eal devices for amplifying signals from transducers type of cbarge carrier and this tends to result in
ch as the piezo-electric crystal and the cer<lmic lower electrical noise. In amplifier systems a low
ck-up, both of which must work into a high noise device is usually required at the input stage
sistance. The crystal pick-up can be simply con since this is where noise generated within the amplifier
lered as a low-value capacitor with a small voltage will receiver the greatest amplification_ Tie noise
nerator in series (Fig. 4.23). The voltage is induced factor of a device is a direct measure of how much
ross the crystal when a mechanical force is applied noise is added to the input signaL Suppose, for
it from the modulated groove on the record. At the example, that the input signal to an amplifier has a
\V-frequency end of the audio spectrum the small signal noise ratio of 40 dB (a vol tage ratio of 100: I),
~ies capacitor has a high reactance (re:J11ember then if the noise factor of the amplifier is 4 dB, the
, = 1/21Ifc) so that if the input resistance of the resulting signal-to-noise ratio of tlle amplifier's output
e-amplifier is not high, the low frequencies \,~l! be win be 36 dB. Noise factors for FETs can be as low
:enuated and the bass response lost. as 2 dB.
il7 AMPLIFIER CIRCUITS
- -__-----~-- + 12 V is directly coupled to Trz, a common emitter amplifier
which provides most of tlle gain.
TIle FET is provided with self-bias via R3 which
works in the following way. TIle gate is connected to
T<,
the 0 V raU via Rl so that when power is applied the
I/,-I'
<'N3~19 gate·to-source voltage is initially zero; this causes a
I------<>
Cz dmin current to flow, which in turn causes a voltage
lnpul CD A5
Oulpul
to be developed across R3 • TIle source voltage rises
lk2
positive, thus increasing the gate-to-source reverse
Ai
2M2 bias and causing the drain current to remain constant
at a reasonable value.
Negative feedback is provided by Rs and R...
Actually the feedback signal is series-derived, since
o-______+-_~~----------------oov
R.. is in series with the output resistor Rs , and series
applied since R.. is also in series with R 3 • The gain
Fig. 4.22 FET input pre-amplifier is given approximately by the ratio of Rs and ~.
C1 is a decoupling capacitor which is necessary to
ensure that a high overall open loop gain is achieved_
Its value determines the low·frequency gain point.
When you build this circuit you may find that the
d.c. bias voltages are not the same as those given in
the next table. This is because the parameters of
FETs do differ widely. What is important is that the
Fig. 4.23 Equivalent circuit of CI}'stal pick-up operating point of Trz collector (point 5) should be
such as to allow reasonable positive and negative
The specification for the pre-amplifier is
swing at the output. This output signal is suitable for
Voltage gain 12
driving a medium power output stage; a volume con·
Input impedance 2 M11. at audio frequencies
trol would be required.
Output impedance 130011.
. The voltages at the d.c. test points measured with
Frequency response 20 Hz to 30 kHz
a 20 k11.{V meter with respect to 0 V were
Output Signal 1 V r.m.s.
A low voltage gain is used since the signal from the
pick-up is usually 100 mV or so.
An n-channel junction FET (2N38 19) is used as a
;;R I ~ I Ii.3 I ~ ., 0:6 5
8
common source amplifier in the input stage; and this
Ouestions The next set of readings were taken

under fault conditions. Identify, with reasons, the
faulty component and the type of fault.
1 2 3 4 5 Output
I
Fault
A 0 11·9 3·3 3·2
3·2 Zero
B 0 11·2 1-2 1·2
11·9
Zero
C 0 11-3 3 0-6
8
Very low gain
D 0 12 0 0
0
Zero
E 0 10·5 3 0
0-1 Zero
----~
g ELECTRON!C FAULT DIAGNOSIS
9 Exercise: D.C. Amplifier (Fig. 4.241 output will drift by 50 mV because the 2 mV
change of VBE at the input.transistor will appear as
.C. amplifiers are used to amplify slowly varying
a d.c. input signaL For this reason single transistors
~nals from transducers such as thennocouples,
are rarely used as the input stage of d.c. <ilnplifiers.
ermistors, strain gaug~s, photocells, etc~ As weU as
111e circuit in common use is the diffcrelltial ampli·
:lplifying the signal, the amplifiers' output" must
>t drift. Drift in d.c. amplifiers is defmcd as any fier shown in 1'1g. 4.25, in which two transistors are
",tired together in a balanced arrangement. For this
ange in output signal when the input is sbort
type of circuit two sets of input conditi(.ns can be
'cuited or held at zero. Drift is caused by several
considered:
;tors but the most important are those that effect
(i) lnput of opposite polarity calkd differential
e input stage; these are temperature and power
mode inputs.
pply voltage changes. The latter can be minimized
(ii) Inputs of the same polarity..:. cared common
using a well stabilized power unit so we shall COll
mode inputs.
er the effect of temperature. 111is affects tran·
tors in three main ways: firstly it changes the

For the differential mode input signals a large
~meters. namely the current gain, secondly it
output signal is generated. Imagine point A going

ers the leakage current, and thirdly it changes thc
positive while B goes negative, Trl will conduct more
;e emitter forward bias voltage:
th,1\I Tr2 and the output signal will be.Iarge.
Olanges in /tFE with temperature can be mini·
A common mode signal on the other :'land will
?;ed by using negative feedback. Leakage currents
result in little change at the output since both
! be kept low by using silicon planar transistors,
transistors conduct more or less equally. Now as long

I so we are left with the main cause of drift being
as the two transistors are matched and situated close
change of VBE with temperature. In any tran
together, preferably in the same enclosure, then any
or. VBE changes by approximately -2mV per
changes 'in the VBE of both transistors with tempera
~1"ee C rise in temperature. TIlis may not sound a
ture can be considered as a common mode input,

. but imagine that a single transistor is llsed 3S the
with the result that little cllange in the output signal
ut stage of a d.c. amplifier with a gain of25. Theil
takes place. 111US by using the differential amplifier,
every one degree change in temperature, the drift with temperature can be kept to a low value. i
I
----~~----~--------------~r----o+lbv
Tr4 6CY70 1
I
• R,
Jk --.....---...,...-. +
+----l-___
___+)_ Oulpul
A 0------1---1
e
zener
----~-----~~---------~~---Q -IOV
4.24 D.C. amplifier Fig. 4.15 f:la~ic differential <lmpl.ifier
49 AMPLlF1EH CIRCUITS
A measure of the quality of a di fferentjal amplifier small enclosure_ \II'hen you test the circuit you can
is c.1l1ed COllUl1Un mode rejection mHo (CMRR) observe the effects of temperature by blowing on one
of the transistors and noting the large change in d.c.
CMRR = Differential gain.
~>utput.
A constant current of about 330 I1A is pro·
Common mode gam
vided from Tr3_ This current is determined by the
To achieve high values of CMRR it is necessary zener diode and R 2 .
to make the emitter resistor R£ as high in value as -nle output from Trl conector is connected to the
possible, since it is this resistor that provides the common cmitter amplifier Tr4. l1legain of the cir
negative feedback which keeps the common mode cuit is stabilized by a negative feedback loop from the
gain low. For this reason the current to the differential collector of Tr4 via R1 to the base of Tr 2' TIle
amplifier is often supplied from a constant current p(ltentiometer RV J in the differential amp, is used
source, and R£ is then equal to the high slope to offset any differences between Trl and Tr2 base
resistance of the output of this source. In the example potentials when the input is zero. Under these condi
(Fig. 4.24), Tr3, wired as a common emitter amp· tions it should be adjusted to give zero output volts.
lifier, supplies the constant current. 111e drift and stability of the amplifier i~ best
TIle d.c. amplifier in this exercise is relatively checked by a chart recorder. The voltage gain can be
simple and, the~efore, does 110t produce the best measured by applying d.c. signals from a stable
results in ~erms of drift and stability that can be millivolt source.
achieved. Nevertheless, if care is taken over selection In operation a smali positive input signal (d.c.) ",iU
of the transistors and in construction of the circuit, cause Trl to conduct more heavily than Tr2. thus
fair performance is possible. the collector voltage of Trl Wl~I fall cau~ing Tr4 to
The spedfication is conduct more. The output voltage rises, and a portion
Voltage gain 22 of this is fed back to Tr2 base which opposes the
Input resistance 10 kfi input. To alter the gain. change tile ratio of R7 and
Output voltage ± 5 V maximum ~.
Temperature drift ± 3 mV per °c TIlC voltages at the test points with respect to
Stability ± 10mV per hour o V were measured using a.20 kfi/V meter when the
Trl and Tr:z form the differential amplifier, and circuit was working nOIDlally witll zero input.
tilese transistors should be carefully matched for
current gain and mounted close together.-This will
TP
V
en~ug~ the hest drift perfonnance. It is al~o "1:'~,~o.
shield the circuit from draughts by building it into a
",.'
1 :.
, .'
Questions
(2) Write down the symptoms for the following faults:
(1) From the following set of readings, which were (a) Tr2 base emitter open circuit.
taken under fault conditions, determine which com (b) Rs high in value.
ponent is faulty and its type of fault_ (c) TrJ base emitter short circuit.
I 2 3 4 5 ---
Fault ,
A +9·6 +4-8 +4·7 +5-5 -10

B +9-6 -(5·8 -8 -7-4 -10
C +9 -0-6 -8 -7-4 +0-2 Output ~rifting
D +9·5 0 -9-6 -10 --10
E +9·6 !
-I -8 -7-4
--L--_ __ -10
50 ELECTRONIC FAULT DIAGNOSI,S
4.10 Exercise: Audio Power Amplifier (Ftg. 4.26) Trs conduct on the negative half-cycle::. Naturally
a small amount of forward bias must bl: supplied to
SpecificatiOIl \
the output transistors, otherwise the output signal
Power output 4 W into 8ll
would contain cross-over distortion. l1tis bias is
Harmonic distortion less than 2% at max
provided by diodes DI and D2 , and tlu: low-value
output
variahle resistor RVJ • To enable a small quiescent
Sensitivity approximately 1 V T.m.S.
current to flow through Tr'l and Tr$ a d.c. voltage
Frequency response 15 lIz to 20 kHz
of approximately 1·25 V must be set up between the
Input impedance I k5 n
base connections of Tr 2 and Trl. TIle <iiodes D 1
TIle circuit uses a standard class B complementary
and D2 , being forward biased by Trl collector
output stage with a matched pair of medium power
current. give about 1-4 V and RV 1 is used to adjust
transistors (type BDl31 and BD132): lllese two
the level. This potentiometer must be :Idjusted
transistors are fed via transistors Tr2 and Trl from
carefully during the initial setting up so that the
the output of a common emitter amplifierTrl' ~r2'
current flowing in Trs emitter is aboct 10 rnA.
Tr4 and Trl. Trs are wired as complementary pairs
To do this the link A is broken and a d.c. milliam
with very high current gains.
meter placed in circuit to monitor the current. TIus
Transistors Trl and Tr4 conduct on the positive
setting should remove any tendency .for cross-over
half-cycles of the signals across· It; • while Trl and
distortion to occur.
Fig. 4.27 Square wave test on audio

amplifier Input 4,) Hz square wave
+[]1
C 2500p.F
3
+14V~.J~
au
Load ~4V -------------
. _ . ~;;L._ _- ,
Fig. 4.28 Wave fOIm for fault E

,. if= 1 kHz)
Trs
1\
61)13'
Link A
RI
Ik6
Fig. 4.29 Wave form for fault D
if= 1 kHz)
Fig. 4.26 Audio power amplifier

51 AMPLIFIER CIliCUtTS
The t \YO OU lput transistors must be mounted on a Since transistors with high values of cut-off
heat sink, about lOO cm 2 of 3 mOl thick aluminium frequency are used the circuit may well tend to
is sufficient. When the Olltput transistors wann up burst into oscillation. Components Cl , C4 and Cs
during Itigh power output, or when the ambient are included to prevent this.
temperature changes, changes in the base emitter When measuring the power output. an oscillo·
voltages of Trl and TT3 take place. 111e diodes DI scope is used to check that an undistorted voltage
and D2 change with temperature' also so that the waveform is set up across the load, and a standard
overall bias conditions do not ~1ter. Dl and D2 then multimeter on a.c. voltage range to measure the
provide a degree of thermal compensation. r.111.S. value of the output voltage. Then
A stabilizing negative feedback loop is provided
V2
via R:z, which feeds a portion of the output signal Power output =~
back to oppose the input. Since the circuit is
directly coupled, the feedback also stabilizes the where RL is the 8 n load resistor.
d.c. operating point at the junction of R7 and Rs· TIIC frequency response results are shown in
If the circuit is working correctly, Tr2 matched with Fig. 4.27 for a 4.0 Hz square wave input. Note that
Tr.h and Tr4 matched with Trs, then t."tis d.c. the low frequency components present in the 40 Hz
operating poi!lt shol,lld be exactly half the supply square wa¥e have less amplification than the higher
voltage. TIle rest of the d.c. voltages can be calcu· frequencies. This is not necessanly bad since the
lated by working from this value and remembering lower cut-off frequency of the amplifier needs to be
that the voltage between base and emitter of a work higher than the bass resonance of the loudspeaker.
ing transistor should be approximately 0·7 V. The In Qass B output stages, the output transistors
actual voltages with a supply rail of +24 V are are prone to damage if either the output is acciden
TP
MR
I I I I I I I
1
+1
2 3 4 5 Supply
11·1 12·5 11·8 11·8 (24)
tally short circuited or if an open circuit occurs in
the bias chain formed by D" D2 and RV J. To avoid
.this the power supply could be fitted with a current'
Capacitor C3 , the coupling capacitor from the limit of say 750 mA, or 750 rnA fuses can be fitted
output to the 8 n load, should have a working in place of R1 and Rs.
voltage of 25 V and a ripple current rating of at
least 600 mAo (Do not fit too small a capacitor, and
make sure that it is mounted well away from the
heat sinks.)
Questions
(1) All the following d.c. voltage readings were taken (2) What would be the symptoms for
when a fault existed, with an input of 100 mV at (a) Tr4 base emitter open circuit
1 kHz. Test points were measured v.-ith respect to (b) C3 open circuit.
o V \Ising a standard multimeter. Consider each in
turn and tty and deduce which component is faulty
and its type of fault.
1 2 3 4 5 Additional symptom
---
Fault
A 1·2 6·5 9·5 13 8 No output
B 0 0 0 0 0 No output
C 8·5 22·5 23·9 23·2 23·2 No output
D 0·8 11·5 12 11·7 11·7 Waveform as Fig. 4.28
E 1·0 1I·7 12·5 11·7 11·7 Waveform as Fig. 4.29
F 0 22·3 23·9 23·2 23·2 No output
_ _ _- - 1_ _ _ _
I
5 Oscillator and Time Base Circuits
5.1 Principles of Oscillators {(l' Sine wove

f= -,}Hz
An oscillator is any device or circuit that produces
an output which varies its amplitude with time. The + --
I.
output may be sinusoidal, square, pulse, triangular, \.. :PCOk value
or sawtooth as shown in Fig. 5.1. -- \ __ 1

TIlese circuits are used in all types of electronic
equipment, from radio and tv transmitters and
receivers, computers, oscilloscopes, signal generators, Periodic
..... ---
to digital frequency meters. lime T
Oscillators can be constructed using components (b) Square wave
:hat exhibit a negative resistance characteristic such +

lS the unijunction transistor and the tunnel diode.
nle operation of these will be discus~ed later. How o -- --- -- ._ -- -- --~- I
:ver, a large majority of circuits are based round an

Impljfier with a positive feedback loop_ When a por·
ion of an amplifier's output is fed back in phase {c 1 Pulse Iroin
(posilive pulse shown)
vith its input, the effective input is increased and so
; its overall gain.
For positive feedback
_ Ao
Ac -1-/M
o
(d) Sawfoofh
So, if the loop gain PA o approaches unity
Ao
Ac 40 ~oo
The gain with positive feedback thus approaches

finity. Such high gains will result in oscillatiolls, Ule
equency of which must be controlled by a frequency
:termining network.
TIle requirements-for a circuit to produce oscilla·
Fig. 5.1 Typical oscillator output signals
)flS are
(a) Amplification
(b) A positive feedback loop
(c) Some network to control the frequency
(d) A source of power.
This is shown in block form in Fig. 5.2. The net
lrk that determines the operating frequency may
part of the feedback circuit or external to it.
'pi cal circuits are combinations of Land C, or R
I~
53 OSCILLATOR AI'JD TIME BASE CIRCUITS

AJlother method of providing positive feedback
in a single stage amplifier, without using a transformer,
is to lise a PHASE smf-TING network from the
collector to the base. A typical example of a phase
·shift oscillator is shown in Fig. 5.4. Each CR network
0
from the coliector to the base providcs 60 of phase
shift so th:::.t the resultant feedhack is positive. This
oscillator is p,lrticularty useful for generating sine
waves of I1xed frequency.
Fig. S.2 mock diagram of a typical oscillator
and C, and we shall consider some of the more basic

types first.
A good example to start with is the TUNED
CO LLECTOR oscillator (Ag. ,5.3), a circuit for
producing sinc·wave oscillations at frequencies from
a few kilohertz up to 1 MHz. TIle amplification is
provided by the transistor which is connected in
common emitter mode. Bias components R" R2
and R3 cause the transistor to conduct, which then
forces the parallel tuned circuit of Lt C1 into oscilla
tion. ll1ese two components determine the operating
frequency, which is given by the familiar formula . Fig. S.4 Phase shift oscillator
fo = 21f..;r:c
Another circuit which depends upon phase shift
is the Wien bridge oscillator 5hovo/n in schematic form
In order to maintain oscillations, positive feed in Fig. 5.5. TIlis is the standard circuit used ill sine
back is provided by the secondary winding of the wave generators over the range of I Hz up to I MHz
0
transformer. This must be connected to give 180 since it can easily be made continuously variable
phase shift. over that range. It has excellent stability and pro·
duces an output with low harmonic distortion. An
Posilive feedback . example of this circuit is discussed later.
~----------r-----~r-------o +Vcc
,
~ Frequency
_C, -<~~ de1ermlning
R, : ne1work
_ _ _ _ _ _ _ ..,1
/\M. Positive
feedback
Amplifier NIH}olive
feedback
Fig. 5.3 Tuned coliector oscillator Fig. 5.5 Block diagnam of Wien bridge oscillator
I
Square, pulse and sawtooth oscillators usually

have an operating frequency that is determined by
the charge and discharge times of a capacitor. Such
circuits are used as clock pulse gcnerators in digital
circuits, time base generators in oscilloscopes, <Ind
pulse generators in radar.
It would not scrve a useful purpose to detail
further ali the many forms of oscillators that are in
current use. Other texts should be consulted for
Rooq.
circuits of say the Hartley, Colpitts, Clapp, and so 011. swilch
What we are concerned with are the important

features, or characteristics, of the circuit, and possible Fig. 5.6 Block diagram of one fonn of frequency cuunter
fault conditions. TIuee of the main characteristiCs
are highly stable and accurate osciilator, d:.spJay the
(1) Operating frequency unknown frequency 011 a five digit or more in-line
(2) Output amplitude display. TIle accuracy of such instrumE'nts, dependent
(3) Frequency stability.
upon the accuracy of the internal oscillator, can be
Depending on the type of circuit however, other
±0·01 % or better. TIle instrument couuts the number
important characteristics will be of oscillations of the input frequency that OCcur
For sine wave oscillators within a time period determined by th.~ internal .
(l) The harnlOnic distortion (the purity of the oscillator and the display switch. For example
sine wave).
suppose a gating time period of 1 milli:;ec is selected,
For square wave and pulse oscillators
and the total count displayed is 1000, then the
(1) The mark-to-space ratio, and width. frequency of the input 1s 1 MHz.
(2) The rise and fall times.
(3) Any overshoot or sag.
5.3 Frequency Stability

For sawtooth oscillators
(1) TIle linearity. TIle stability of the output frequency is very im!.Jort
(2) The flyback time. and in many applications. Various factors can cause
the frequency of an oscillator to drifl from the preset
value. l1lese include
5.2 Measurement of Frequency
(a) Changes in power supply voltage levels.
The measurement of frequency and output amplitUde (b) Qlanges in active component plrameter
can be made using any standard method depending tramistor current gains, etc.
upon the required accuracy. An oscilloscope is per (c) Changes in load.
haps one obvious choice, but even when properly (d) Variations in components that determine the
calibrated this only gives an accuracy of about ±3% frequency.
for both time and amplitUde. This may be adequate The effect of the first three factors can be
for a large majority of cases, such as for example minimized by using stabilized power u aits, and a
the measurement of the frequency of the bias buffer amplifier between the oscillator and load.
oscillator in an audio tape recorder. When greater TIle biggest calise of instability wiU come from
accuracy is required the unknown frequency must changes in the components that make u.p the
be measured by comparing it with a standard frequcncy·.-I~termining network. Obvic'llsly com
oscillator of known frequency. Using a eRO to ponents with good long·term stability, and very low
obtain Lissajous figures is one well known method. temperature coefficient, should be usc:!, and quite
More commollly the preferred method now is to often these components are housed in a temperature
measure tile frequency by a digit.al frequency meter controlled enclosure. To achieve the highest possible
(Fig. 5.6). These instruments, fitted with an internal, stability, the designer must resort to w:ing a piezo·
55 OSCILUHOR AND TIME BASE CIRCUITS
electric crystalt{) determine the frequency. Certain

substances, typically quartz, when speciaJIy cut, will
mechanically resonate with an applied volt:.:tgc.
Stabilities of J part in J 03 arc re:tdity attained. Two
typicat examples of crystal controlled oscillators
are shown in Fig. 5.7 A and B. The first circuit is a
modified C'.olpitts circuit with the crystal in place of
the inductor. 11le second circuit uses the 710 inte
grated circuit which is a high-speed differential
comparator. Positive feedback is provided from the
output to the non-inverting input by the crystal.
The d.c. operating point is fixed by R3 and R2, and '-------~-->--- 0 v
Fig.. 5.7A Crystal-controlled colpitts oscillator
C1 decouples R 2 , thus removing negative feedback
at the oscillation frequency.
5.4 Harmonic Distortion

This results from non-linearity or excessive gain in
the amplifier circuit. With most oscillators the gain
of Ule amplifier must be controlled to a value that
just maintains the losses in tbe rest of the circuit.
In Fig. 5.4 for example, the potentiometer can be
adjusted to give more or less negative feedback. If the
tMHl
amplifier gain is too high then tlle amplitude of the
oscillations builds up and the output distorts.
Distortion is best measured by a distortion Fig.. 5.7B 1 MHz crystal-controUed oscillator using a 710
measuring set, but assuming this is not ava~able, comparator
another method can be to use a narrow, band stop
(0) Single pulse
filter. The oscillator output is passed tluough this
filter, which must have very high attenuation at the
oscillator's frequency, but wllich passes all harmonics_
TIle resulting output from the filter can be measllTed
with a true r.m.s. reading meter, and this can be used
to calculate the percentage harmonic content of the
oscillator's waveform. A twin T-filter is a suitable
type. r
'-<
, .
.............
I, If
5.5 Square and Pulse Waveforms

(b l Pulse train
Square wave and pulse waveform is shown in Fig. 5.8,
and in some cases the same instrument is used to Mork Spoce
t-<-+-f-+---->-'
give either square or pulse output. Botll waveforms '1' I
" r
are called rectangular, a square wave being a special
case of a rectangular wave with a mark-to-space ratio
of olle.
A typical pulse waveform is shown in Hg. 5.8,
in which the various characteristics are defined. Note Fig.. 5.8 Pulse wave form 1
that the rise time of the leading edge is measured as In{b), pulse repetition frequency =T pulses per sec (Hz)
the time taken for the signal level to change from 5.6 Sawtooth and Ramp Circuits
10% to 90% of its full amplitude.
TIle majority of these circuits are those hal produce
Pulse measurements are made Ilsing a wideband
a waveform which rises steadily with tim~ up to a
oscilloscope, and usually an external synchronizing
required amplitude, and thell returns rap;dly to the
signal must be provided in order tlwi the trace
point from which the output can again rise. TIle
displays the leading edge of the wavefonn. With rise
linear rise is usually called the sweep, and the rapid
time measurements, the rise time of the measuring
return the flyback. These circuits find th·!ir main use
instmment itself cannot be ignored, and nor must
in sweeping the beam across the face of a ea thode
the measuring leads present a relatively high stray
ray tube, in other words in oscilloscopes, television
capacitance at the oscillator. An attenuating probe
cameras and receivers, and radar displays. In this
must be used to couple the signal into the V-amplifier
case they are usually referred to as timebase circuits.
of the oscilloscope.
TIle same type of circuit can be found in digital
llle rise time measured on the screen is related
voltmeters where they are referred to liS lamp
to the circuit waveform rise time and the oscillo
scope's rise time by the formula . generators. Ch:cuits like these can be either free
running or triggered from an external oscillator, but
tr =V(t~1 - t}) the basis of all of them is a capacitor which is .
where tr is actual rise time charged and then rapiUly discharged.
tm is measured rise time Apart frol11 the frequency, the important requi-re
ty is rise time of oscilloscope Y -channel.
ment for most circuits is good linearity. 'this means
An example win illustrate the effect of the osciUo tllat the rate of change of the output wit.:t time must
,cope's ris~ time. Spppose tm is 20 nanosecs and be unifoml. In fact, even in general purp.)seoscillo
Ty 15 ns. 111en scopes, the linearity deviation in the tilm: base should
be better than 1%. When a capacitor is charged from
tr = V(400 - 225) =.jf75 :!!: t3 us zero volts towards a voltage V via a resistor R, the
voltage across the capacitor rises·exponentially .
Mlerever tile measured value approaches the oscilla
according to the formula
cope's rise time ti1is formula must be applied.
Square waves can be generated from relaxation Vc = VO - e -ItCR)
lScillators such as the astabie multivibrator or by
TItUS 10 obtain reasonable linearity frc·m a simple
lassing the output of a sine wave oscillator through
discbarger circuit (Fig. 5.10), the maximum change·
squaring circuit such as a Schmitt trigger. lllis
in vcshouh.l not be more than 10% of the total voltage.
ltter method is commonly used in general purpose
lboratory instruments; a typical block diagram is
lown in Fig. 5.9.
;
r-------------7 ;
--t---v
/ / r-""--~-...,
Sine" wove N'v /sine Buffer / Atlenualor

oscillator and R
amplifier
square output stage Output
U
- _. '\
Transistor
switch ,.....---+--0
+~ Ool~ot V\/\
Scnmit1
trigger
Input 0- ~~+) T C
o~~ov
~. 5.9 Block diagram of general p\lrpo~c sine/square Fig. 5. '" Simple disch;lIger circuit for pwducing
lora tory generator S'lwtoofh w~ve forms
57 OSCILLATOR AND TIME BASE CIRCUITS
In this circuit the capacitor charges positively while resistance into the circuit at the operating frequency.
thc input to the transistor switch is zero. When a l11is negative resistance makes up the losses in the
positive pulse is applied to the transistor switch it frequency network. However, it is best to classify
rapidly di::charges the capacitor. 111e linearity from negative resistance oscillators as those that use
a simple circuit such as this is not u~t1ally sufficiently 'devices such as the tunnel diode, tetrode, unijunction
hip)l, unless V is very high. Various mcthods are used transistor, Le. devices that have an effective negative
to overe-orne non·linearity, one being to charge the rcsistance region in their characteristics_ A typical
capacitor from a constant current source. A detailed charac!eristic for a tunnel diode is shown in Fig. 5.11.
example is dealt with later jn this chapter. l11e current first rises wHh forward voltage, then
falls with increasing vol_tage, and finally rises again
as the voltage is further increased. Placing a tunnel
diode across a resonant circuit as shown in Fig. 5.12
provides an effective resistanceless tuned circuit
which will then oscillate continuously. Very high
frequency oscillators, up to as high as thousands of
megahertz, are possible.
l11e unijunction trallsistQT (UJT) is made of a bar
ofn-type material, (sometimes p) with ohmic contacts
5.7 Negative Resistance Oscillators at each end and a p-type emitter junctioll formed
near the centre (Fig. 5.13). 111e resistance of the bar
Strictly speaking, negative -resistance oscillators and is normally around 10 kil, so when connected to a
feedback oscillators are identical, since the latter call supply, with base 2 positive with respect to base 1,
be regarded analytically as having introduced negative the bar acts as a potential divider, and a p.d. of
1'/ V BB appears between the emitter and B1 , where
Forward
Ne~alive V BB is the voltage between B2 and B1 . and 1'/ is called
reSlsla:lce
the intrinsic stand·off ratio (17 is normally between
/j
currenl mA
0·4 and 0·7).
When the emitter voltage is Jess than 1'/ VBB the
emitter junction is reversed. When the applied
emitter voltage exceeds l"/VBB- by about 0·7 V, this
voltage being called the peak point, the emitter
Forward voltoge becomes forward biased and injects holes 1nto the
Fig.. 5.1I Tunnel diode characteristics B1 region. Once this happens, the resistance between
the emitter and BJ falls to a low value. 111e action is
regenerative.
Bose 2
n
type
silicon
Emi1fe(
Sine wOlle
output
L /
p'lype
Symbol
Bose 1
ov
Fig. 5.12 Typical tunnel diode oscillator Fig. 5.13 Construction of an n-bar unijunction transistor
A typical U1T relaxation oscillator circuit is

shown ill Fig. 5.14. Capacitor C charges via R[ and
eventually the emitter voltage exceeds the pe.1k
point of the U1T. TIle UJT conducts and discharges
C rapidly via R3 . As C is discharged the current
through the emitter falls. and wilen it falls below
the minimum holding current then the U1T turns
off. The capacitor can then charge again to repeat c =
the process. In designing such circuits note that the
value of RJ should not be either too low. otherwise
the UJT will not be able to switch off, or too high, ~--------~----ov
otherwise the UJT may not receive sufficient

emitter cur"rent to turn on. Typical values lie be Fig. 5.14 Simple UJT relaxation osci!laloT
tween 10 kU and 1 MU· U1T oscillators such as
this are commonly used to trigger thyristor and
TABLE 5.1 Faults on an Oscillator Circuit
triac circuits.
FAULT SYMPTOMS
5.8 Fault Finding on Oscillators Po!;itive feedback: loop open No output. D.C. bias levels
circuit conecL
Again because the types of circuit in common use
are so various the fault Imding procedure must be Negative feedback loop open Output amplitude increased
circuit and wave shap" distorted.
adjusted for the particular circuit and application.
As with other parts of an instrument. a good under Open circuit component on No output on ':hat range
one switch range only.
standing of the purpose and operation of tbe unit
is essential. Wherever possible consult the maintenance Bias component open No output all !allges. D.C.
circllit levels incorrecl.
manual before attempting measurements or adjust
ments in order to find if any special precautions or Switching tran~istor in ramp Output from ramp perma
generator open circuit nently high.
test instruments are necessary.
A number of oscillator units contain several blocks, Emitter follower with low or Output from ramp reduced
short circuit input impedance and nOIl·linear.
such as attenuator, buffer 3..t"'llplifiers, modulators, etc., in ramp generator
so a logical approach is essential to locate which
Crystal open circuit in crystal Circuit will prc·bably still
block is non-functioning. For example, consider the controlled oscillator oscillate but at a different
block diagram of an amplitude modulated RF signal frequency witl: poor stability"
generator (Fig. 5.15) and suppose a fault exists such
that there is
(0) An output from socket A, which What would be the symptoms for
is continuous wave with SW[ on position 1, (a) An AF oscillator failure, or
and modulated RF with SWI on position 2. (b) A modulator failure?
(b) No output from socket B. For individual oscillator circuits various faults
The fault can only lie in the AF attenuator block. can occur. Table 5.1 lists a few commor, symptoms.
RF
RF output
oscillator Allenuolor
variable A
AF oulput B
Fig. 5. t 5 Block diagram of an Rf ~igl1al gcncr:\tor

59 OSCILLATOR ANO TIME BASE CIRCUITS
R2
470n
RVZ
...oof--()
5k OIP
Fig. 5.16 \'lien 'bridge QscillatQf
5.9 Exercise: Wien Bridge Oscillator (Fig. 5.16) back is from Tr3 collector via thermistor to Tr
. h 2
emlt~er. A termistor is used to stabilize the output
. TIle Wien bridge oscillator is a very popular circuit
3mpl~t,!de. Its resistance will faU if tile ou.tput
for the generation of sine waves in the frequency
amphtude increases, and this increases the feedback
range 1 Hz to 1 MJ fz. TIle main reasons for this are
vo1t3~e, thus reducing the gain and automatically
(a) TIle frequency depends upon the value of R
reducmg the output amplitude. .
and C elements only, and high-grade Rand C
TIle frequency ofWien bridge oscillators can be
components are more readily available than
made variable over a wide range. In this case three
inductors.
switched positions are shown which change the fre
(b) The circuit has good stability and low dis
quen~y by changing the value of the series and parallel
tortion.
capacitors. TIle frequency is made continuous variable
(c) 111e frequency can be made continuously
by using a ganged potentiometer RVI A and RVI B
variable.
for the resistors. A small-value resistor (470 £2) is left
TIlcre is little to be gained from a detailed analysis of
in circuit when the potentiometer is at zero.
the operation of the circuit as it is given in so many
In the amplifier, T r I is an emitter follower feeding
textbooks. Briefly, it consists of a two-stage amplifier
a common emitter amplifier Tr:z. In this way a high
",-jth a positive feedback loop consisti~g of a series
input impedance is maintained. l11e output from
RC network which develops a feedback signal across
Tf2 collector is directly coupled to Trj, another
a parallel RC network. TJ~ capacitors and resistors
are made equal in value, aild the circuit oscillates at common emitter amplifier.
a frequency when the phase shift from output to TIle specification for the circuit is
input is zero degrees. Frequency ranges (1) 100 Hz-3·3 kHz
(2) 1 kHz-33 kHz
1 (3) 10 kHz-330 kHz
fo =:: 2nRC Outpllt amplitude approx. 1 V r.m.s.
,.
The attenuation in the positive feedback loop is Total harmonic distortion less than 0·2% at 1 kHz.
only a [;letor of J, so the overall gain of the two-stage TI\e voltages measured with a standard ll1ultimeter
:It the v:nious test points arc
amplifier has only to just exceed 3 to maintain
Il'~
oscillatiollS. A llcgative feedback loop is used to hold Tf'
the gain at this figure. In this case the negative feed- MR
Questions
(J) If the unit has failed such that no output can be (5) TIle unit fails to give an oulput on range (2) and
obtained on switch position (3), which components the voltages measured with s\\qtch (2) ma :ie, and
are suspect, and with what type of fault? RV.I at maximum, are
(2) If the unit has failed such that no output can be
obtained on any switch position, but the d.c. voltages
appear normal, which components would be suspect?
TP
MR 1·9
2
o
3
16 I l~'2 I 5
153
Give reasons. State with reasons the possible component fault.
(3) What would be the likely effect of (a) the ther· (6) Write down the voltages you would e}:pect to
mistor going open circuit, (b) C4 becoming open measure if Cs became short circuit.
circuit'? (7) The Qscillator fails to give an output ell any
(4) TIle unit fails to give an output on any range, the range. After studying the d.c. voltages, st,.te which
d.c. readings are as given below. State witIl reasons component is faulty and its type of fatIlt.
the faulty component.
Il~.5 I 24
TP 1 2 3 5
TP
MR o
1 2
o 3
16
I I 4
15·2
5
15·3
MR 1·9 0·8 13
5.10 Exercise: Blocking Oscillator Sawtooth . . - - - - - - - - -....---<'}.+ 12 V

Generator (Fig. 5.17)
TItis circ'uit is indu ded because the blocking oscillator
is,;uch a useful circuit, and is quite often used in field
®15r--- 0_,,,
time base oscillators in tv receivers. 1",9t4
TIle operation is as follows. As power is applied, Sowlooth

Cl charges via Rl until the voltage at the transistor
base reaches approximately 0·7 V. 111is will cause the
-g-~_3.
transistor to conduct. Collector current flows -and the
collector voltage falls. The transformer coupling
from collector to base is connected to give positive 0', ] _______4-_____ ,F
feedback; in other words the rise in collector current - --<) OV
in the primary winding induces a voltage at the base
that causes the transistor to conduct more. TIle
transistor collector voltage rapidly faUs to nearly Ou1pul waveform
zero and remains there while the c{lllector current
+6V
continues to increase. At some point, however, the
collector current reaches a limiting value, usually
/
because of the finite current gain of the transistor.
When the primary current in the transformer stops
changing there can no longer be an induced secondary
voltage. TIle base voltage falls and turns off the
transistor. Because of the positive feedback the
transistor switches off rapidly and the base
voltage goes negative. The negative step at the base
.
OV-----------------
-
lime
Fig. 5.17 Blocking oscillator sawtooth generator

equals the change in collector voltage divided by the
turns ratio of the transformer. The transistor is held
off while C} charges via Rl towards +12 V. When
the voltage across C 1 reaches 0·7 V the transistor Questions

again turns on. Without Cz in circuit the output
(1) The sawtooth generator fails and the voltage
waveform would be a series of short-duration negative
levels are
pulses at !l frequency mainly determined by RI and
Ct· TP
The blocking oscillator is particularly useful in MR
generating pulses of relatively high peak power in a
State, with reasons, the component Or components
train of low mean power. For example a tr;,msistor
that are faulty.
with a mean current rating of 100 mA can be used
to give a IJeak pulse of 1 A. TIle average power dis
(2) TIle oscillator fails such that the output is as
shown in Fig. 5.18. State which component is faulty
sipation is kept within the transistor's rating because
and the type of fault.
the duty cyCle is low_ It is this property that is used
to provide a sawtooth waveform by rapidly dis
(3) TIlere are two component failures that could
charging the capacitor C2 when the transistor conducts. cause the oscillator to fail and give the following
readings. State both.
While tile transistor is off, C2 charges via R2 towards
+12 V. TIlUS a sawtooth wavefonu is developed across TP 1 3
Cz; the waveforms are shown in Fig. 5.17. Note MR 0·7 0·1
tI\8t tile linearity is not good.
Wh.en you build the circuit experiment with (~) ~rite down the effect of the collector going open
different values for C 1 and C2 , and note the effect ClrcUlt.
on frequency and waveshape. TIle circuit should give (5) If R2 became high in value, to say 4 kG, what
a sawtOOtIl of-t6 V amplitude at a frequency of about effect would this have on the circuit and output
1·5 kHz. A small pulse transfonner of turns ratio wavefonn?
3:1 was used, but ifnot available, a small audio (6) Which component is faulty if the oscillator fails
driver transformer also works in the circuit. and voltage readings are
~R I ~.7 I ~ I +:2 I I +:2
Fig. 5.18 Output wa....e form under a fault condition

r---------'r-- ----+-----_+----0 + 12 V
R~
Ik
Tr4
Deloe
-~~~--;o- Sawtooth
output
Tr,
BClO6
c, [ R2
0-33p.F 15k
L-----~---_4-~----- __ ~---~----~ov
Fig. 5.19 Free-running sawtooth oscillator with good CV

linearity dt=-.
l
where i =constant current

5.11 Exercise: Free Running Sawtooth Oscillator C = value of capacitor in farads
(Fig. 5.19)
V = rllmp amplitude
TIle previous circuit gives poor linearity. the reason dt = ramp time.
being that the output capacitor is charged exponen With i = I rnA. C= 0-33 /IF, and V = 6 V, then
tially. A marked improvement in linearity results if
0·33 x 10-6 x 6
the capacitor is charged from a constant cun:ent dt - I x lO-a = 1·98 x 1(I-a sec
source. The circuit, which oscillates at a frequency of
500 Hz and gives an output of +6 V, can be considered In other words for a 6 V amplitude the ramp
as three blocks. Tra is the constant current source, time will be approximately 2 ms.
feeding a constant charging current of about I rnA The emitter follower Tr4 buffers the ramp signal
to C l ; Trl and TTl are connected as a triggered to the output so that as little current as possible is
switch to rapidly discharge C I ; and Tr4 is an emitter diverted from the capacitor, otherwise non-linearity
follower providing a low impedance output. To under· will result.
stand any circuit it is always best to consider it in When the voltage acroSs C1 exceeds the base
blocks. voltage of Tr2 by 0·7 ", Tr2 conducts 111is in turn
Taking first the constant current source; the causes Trl to conduct. Since Ulere is positive feed
base voltage ofTra is fD!.ed at a voltage of approxi· back from TTl collector to Trl base, both transistors
mately +9 V by the potential divider Rl and Rz. l]le rapidly turn on_ C, is discharged to a yoltage level
emitter voltage will be at +9·7 V, which means that a of nearly +0-7 V (equal to Trl base emitter voltage),
voltage of2·3 V is set up across R3 and RV 1 _ If tbe When Trl coUector voltage falls, DI conducts tiluS
total resistance of R3 in series with RV I is set to 2k3, cutting off the constant current souro:. As soon as
then the current flowing through the transistor ",':ill C1 is discharged base currellt to Trl ceases, and
be 1 rnA. This collector current will remain constant Trl and Tr2 rapidly turn off, thus allowing C} to
for a large change in collector emitter voltage. When again chuge linearly.
Cl is just discharged and the switching transistors To change the oscillator's frequeuc:,', switch ill
Trl and Tr2 are off, this constant current will flow -:tifferent values of capacitor. TIle outrut amplitude
to charge C 1 , is controlled by the ratio of resistors RG and R 7 , and
Now Q CV for a capacitor and since Q'" idr, these can be adjusted to change the amplitude.
Questions
(J) Vv11ich component is faulty if the generator
frequency changes to nearly 1 kHz and its amplitude
falls to 5 V? l11e d.c. level at test point I is +9-1 V.
(2) What would he the effect of a base emitter short
011 Tr4?
(3) In each of the following cases the gener3tor fails
to produce 3n output. State which component (or
compouents) is faulty and the type of fault. In each
case give a supporting reason.
:~I
2 3 4 5 6
Fault
A +9-8 +9-7 +9 +11-4 +12
B t 9·1 +9·8 +0·8 +0·1 +0·05 +0-1.
C +9·1 +9·8 +9·7 +9 +6-1 +10-9
D +12 +5·3 +0-8 +0-1 +0-05 +0-1
E +9·1 +5-3 +0-8 +0-1 +0-05 +0-1
(4) The oscillator fails so that its output amplitude falls to less than 1 V. The voltages measured:
TP
V
1
+9-1
2
+9·8
State which component is faulty and the type of fault.
I 3 4
o
5 6
+12
( j I RV, fully onliclockwise

+8V
A
output
ov
L.-L--.. . .i -~I
O-ems 2-2 ms
Fig. 5.22 Output wave fonns fI<;Im gated oscillator
Fig_5.22 applies to the Exercise on the Gated UJT

~I +8 Pulse Generator on page 65. It has been placed on
e this page for convenience of space and layout.
I~
ou1pu1
OV
I ji) RVj fully clockwise
'"~, J
L. ___•.1..-J
U I
2-2 inS O-Sms
e
output
Jl
Trigger lIP
Fig. S.20 Slow-speed trigger<-.d ramp generator L ______________________ -20V
~
-
-
-
-
-
-
_
Q
5.12 Exercise: Slow Speed Ramp Generator (Fig. 5.20) and to prevent Ole load from unduly affecting the
linearity of the ramp. As the output rises so does
lllis circuit is designed to provide a single ramp output
the voltage at Ole junction of RiO and R lJ • \\'hen
signal for one positive input trigger pulse. The ramp
the voltage at this point reaches approximately +}·O V,
duration is approximately 1 sec and the ainplitude is
the diode D2 conducts and switches on T(2 to reset
10 V. As in the previous example tbe timing capacitor
the bistable_ Tr3 is switched on and C4 i~ rapidly dis
C4 is charged from a constant current source Trs. A
charged.
tantalum capacitor should be used.
A unit such as this can be useful for checking the
Trl and Tr2 form a bistable circuit with Trl
characteristics of devices when used in conjunction
normally off and Tr2 on. Under these conditions
the collector voltage of Tr I will be high and this with an XY plotter.
foward biases Tr3 via Rs. Tr3 is the switching Questions
transistor which clamps the voltage across ~ to (1) now could the circuit be modified to be free
nearly zero vol~s_ running-?
When a -positive input pulse is applied to Trl base, (2) What is the purpose of Cs ?
the bistable changes state with Tr,on and Tr2 off. (3) In each of the following fault conditi(lJls the
TIle forward bias to Tr3 is s\vilched off. and C4 can circuit fails to operate correctly w]Jen an input
ll\lW dl:trg<'. Th<' \"\)ll:1g~ a':IO~::' th¢ ':;lp:Kilor ri$~s trig&cr is ~upplied_ State \vhich component (or
positively. 1l1is linearly rising voltage is fed via an components) is at fault and give reasons fm your
emitter follower Tr4 to the output. An emitter choice. TIle voltages were measured with respect to
follower is used to provide a low-output impedance o V using a standard tnuitimeter.
2 3 4 5 6 7 8
Fault
._--
A +0-1 +0·7 +0-05 +19·6 -1-6 +19·5 +18·9 +6·4
B +19·6 0 +0-7 +0-\ -8 +19-5 +18-9 +1-4
t +0·1 +0·7 +005 +19·6 -1-6 (I -D-6 -6·7
D +19-6 +0-2 +0-75 +0-75 +0-7 Q -D-6 -j)- 7
E +19 0 +0-7 +O-J +0·7 +0-05 -0-6 -j)·7
F +19 0 +0-' +0-1 0 +19·6 +189 +\-4
G +0-1 +0-7 +0-05 +19-6 +0-7 +0·7 Q -(i·2
H +0-1 +0-7 +0-05 +19-6 -1-6 +19·6 +J9 [9·6
65 OSCILLATOR AND TIME BASE CIRCUITS'
- - - - - - - 1 - - - - - - 1 1 - - - 0 +8 V
Output
a
Gated
pulses
Square c,
waves 0,15 J'F
Fig. 5.21 Gated pulse genetator
~----------~--------~-----~---~-__oOv
5.13 Exercise: Gated UJT Pulse Generator (Fig. 5.21) Ouestions
As stated previously the unijunctioll transistor makes (I) What would be the effect on the output waveforms
an excellent relaxation oscillator. In this circuit the ifTr3 developed a base emitter open circuit?
emitter supply of the unijunction is fed via Ii tran· (2) 1he circuit develops a fault so that the output
sistor switch from an astable multivibrator. The low waveform at B is the inversion of the waveform at A.
frequency waveform from the astable is used to gate Which component is faulty?
the unijunction, so'that a burst of pulses appear at (3) What would be the effect on the circuit operation
tlle output. if Tr 1 developed a base emitter short?
Trl and TT2 from an astable multivibrator, (4) Whj~h component (or componenfs) is at fault if
oscillating at a frequency of approximately 300Hz. a continuous train of negative pulses of o-} ms
TIle mark-to-space ratio o(the waveform at Tr2 duration appears at output B. The voltages measured
collector can be varied ove'r a fairly wide range by the at the test points were
potentiometer RV 1 • When Tr:2 switches off, Tr3
TP 4
conducts and so its emitter rises to nearly +8 V. C3
then charges rapidly via Rs, the unijunction triggers
MR o
discharging C3 and a positive pulse appears across (5) State, for the following fault conditions, the
R , - This c.auses Trs to conduct, and a pulse is componellt that is faulty and its type of fault. In
generated at the output. While Tr2 remains off, the each case tlle fault symptoms are that tJlere is no
, unijunction circuit continues to oscillate producing output waveform at (B) while a 300 Hz square wave
pulses at the Otltput. TIle Humher of pulses appearing i~ present at (A)_
at the output is determined by the mark/space ratio
6 7 8
of the astable mul1ivibrator. This is shown in Fig. 5.22*
Fault
where it can be seen that few pulses are produced for
1 0 0·3 8
a low mark/space ratio, and ten pulses are produced
2 0-32 0-32 8
for t]le maximum mark/space ratio. The operation
3 5-4 0-3 8
can be modified by changing values of R 3, Rot, or ClI
4 3·6 0 8
5 0-4 0 8
*Fig. 5,22 appears on page 63_

6 Pulse and Waveform Shaping Circuits
6.1 I"traduction takes nearly 4·5 time constants for the vol tage
across the capacitor to equal V.
~{JfJ, "f r",~/#"I:!rJ '*:!"tr;.t!(#nir~'t l~ r;/Jf!(.;:rr'~/1 ,lith

lh:: '!ffs::::.'_ (;f an !!":te-g!"~r.r}r (;~ ~!.!!3es -;;,'hlch h~4e
1*~A: :;.~:~ ~:!. ~ :'".t~~·~i;:, 1: ; ..' ..~~-:. ~ "'. ~ ,,_*:--_ -:.~ l.~r,- . (. :t- 1:~~
~ !...~~:~ "';.: ~~:-; ::; :.."'-",:,:,:;>~::::-,.-"'..~.:. '!."#' "',.:--.?! --
~!L<-h VU~:;""'~.IJ1~!il:)l ~.!";·1 '\:1·<-""';'~"'.J!'~~#~;;·t<-;:-i::'lt -=-~t".
....·ork.:s C<!.11 be grouped into
time constant is 10 depade lhe ri<.< and f ill times .
i ~_ ' :.:::'=:!! ":':lSsi";e. ::~:!!r!: "hose ~:!.de -!p. "':1- R .. <:
IL b::w~·,;e::. the ?'..!!se is ,hart in ccmp::tlison to the

·"'l11r~aloi~- ·imi£ ;:;om.Lam, in::Oll tnt" 'C.Ipa:::':irm Will
and L elements.
(2) Non-linear passive circuits: diode dippers and not have sufficient time to charge completely, and
restorers. the output will appear triangular. Circuits such as
(3) Active circuits: those that use traIlsistor these are often used to provide short time delays.
~itches su;::h as lh~ S-:h..I'.ln trigger :and moMstable. An example is shov."ll in Fig. 6-2_
Some of the more com.1TIOrl. eX3ffiples v.iD t-e The DlFfERE;"'llATOR, b:ask:illYl!Ugh pass
dealt ·with in tllis chapter. Such circuits will be found filter, aIlo'ws rugl1 irequencies to pass, but attenuates.
in colour tv receivers, radar sets, and in fact in nearly the low frequencies. TIle circuit togetller with out
all electronic equipment. puts for various pulse inputs is shown in Fig. 6.3.
When a step waveform is appJied, and ~ssuming that
C ~ uncharged, then fue voltage across the capacitor
cannot change instanta::leously. TIle voltage across
6.2 linear Passive Circuits - the Integrator and
a capacitor can only dlange when it acquires some
Diff:!!'rentlBtor
ch2.Tge. .and tlus D2IUrilly L'?kes tL-ne. S:l the output
:::;-;.!s: :i.,= : ~ :~:" S£:'":'r= ~::= z..t: :.3=.' i.~::~ __~ C d!z;ges,
=~ .:,t-~~ i!~:~""'.i:= :::t::-..;:-"~.a~ l.:':r::
.:z::-:s R -:: - ;:
=.z :~::: 7l.:i~?=
st:-e v.;x;;e i:r:pur sig.n:=.L rr:ey or-Jy produce artcnu3tion

J·r = J-( e -l/CR )
and phase shift for a pure sine wave. Howe"er, they
greatly modify other wave shapes. TIle result is tltat a "spike" equal to the change of
TIle INTEGRATOR, often called a low pass ftlter, state at the input is generated.
is shown in Fig. 6.1 together \\ith output signals for If an input pulse iliat is long compare.-:i to the
typical inputs. Since the reactance of the capacitor differentiiltor's time constant is applied, ~hen the
falls with increasing frequency, this circuit removes output must go negative on the trailing edge. This
tlle high-frequency components from a pulse wave occurs because the capacitor, already charged by the
form. '''hen a step input is applied the voltage across leading edge of the pulse, cannot change its voltage
the capacitor cannot change instantaneously. It rises instantaneously when the trailing edge ar.:1ves. TIle
exponentially according to the formula left-hand plate of the capacitor wiII be at +V and the
Vc = Vel _e- t / CR )
right hand plate at 0 V. When the input changes
abruptly from +V to 0 the output must now change
Now CR, the product of capacitance in farads and from 0 to -v.
·esistance in ohms, is called the TIME CONSTANT Differcntiators are often used to convert pulses of
)f the circuit. In one time constant the voltage acros:; one polarity into "spikes" of the oppositf polarity.
he capacitor changes by about 63%. Note that it TIlese spikes can then be used to trigger otl-·er circuits.
·6
I
67 PULSE AND WAVEFORM SHAPING CIRCUITS
tnpul
Output
(0) Slep :~-- O~

(bl Pulse
(widlh ",reoter
Ihon lime conslo nt )
lei Pulse
( widlh equal 10
time conslonl )
(d) Pulse :X---------~~

(widlh less thon •.
lime constant I
Fig. 6.1 An integrat{)I c
R
" "' J ""''
Inpu! Output
Schmilf
'rigg~r OulpUI
o----~"-~ circuil ( olSlep
Time' relc fed wove forms:

+
Inpul
(b) Pulse
o~
nl__
( widlh less lI10n
+V lime conslonl I
-J- Thresl10ld of
Voltage across Irigger circuit
O~--<'
c
+
OUlpul
(clPulse oj
Schm.l! (wld.h qreoler Ihon
hme conston~ )
Fig. 6.2 Vse of an integrator 10 give a time delay Fig. 6.3 A differentiator
6.3 Diode Waveform Shapers
Diodes, because of their properly or conduction in
one direction only, are widely used to remove por
tions of a signal (clipping) anti to clamp a waveform
to a reference level (restoring). Input Output
CLIPPING CIRCUITS are used where only a o....- - - - - - - o O V
portion of the input signal is required. A diode,
either in series with the signal path or parallel to it,
is used to remove part of the signal that lies either Waveforms: +
above or below a d.c. reference level. A typical series
diode clipper is shown in Fig. 6.4. The diode only Input
conducts when the input Signal exceeds the reference
bias level VB. Naturally when the diode conducts,
its forward slope resistance fOllUS a potential divider
with R to the input signal. TIle output signal then
, will be attenuated slightly, i.e.
1
I • R ) Output +Va
I Vo =Vi (R+rd - --.-------------- OV
I
l
TIlerefore, to minimize errors, R is normally made
much greater than rd. In practice this is readily
achieved since the diode slope resistance rdis )ow,
Fig. 6.4 Series diode clipper
o
typically less than 100 U. By reversing the diode the

output signal would contain only that portion of the Input
input below VB, and the upper portion would be
clipped. o--~--"""---o
Similar effects can be achieved by connecting the Fig. 6.5 Shunt diode clipper
diode in parallel with the signal as in Fig. 6.5. Here C 0
'"~' ~3~-~ ~ ~,: : J- '-

ifthe input exceeds J11 the diode conducts and limits
I the output. A circuit can be made using two diodes
to clip both the positive and negative portions of a

__:"'
waveform. Such circuits are sometimes called
"slicers" and can be used to convert sine waves into
\ semi-square waves.
I~ An example of a diode circuit used with a dif·

ferentiator is shown in Fig. 6.6. Here the diode is
used to remove the positive portion of the waveform
Input
L
so that only a negative "spike" is generated. Waveform
A CLAMPING CIRCUIT is one which fixes one across R

-.
of the peaks of an alternating waveform to a d.c.
reference voltage. Note that a clamper should ideally
not affect the wave shape. A typical circuit of a
damper to 0 V (dc. restorer) is shown in Fig. 6.7. Output
For correct operation the time constant of the CR waveform
network should be much greater than the input

pulse width. The output can only be clamped correctly Fig. 6.6 Use of a diode wlth differentiatof to provide a
when the input consists of a train of pulses as show11. negative spike ffCJm a posith'e pulse input
u::J PUL:)t: ANU VV/-\Vtl-UHM 0HAt'II'Hi UKCUI 10
'","' :JI
'nIe capacitor charges when the input waveform goes c
;
negative since the diode conducts heavily. After a
I
few cycles of the input waveform the capacitor 0.,,",
hecomcs charged to the peak value of the input, and
therefore shifts the mean output level positive. By I
reversing the diodc the positive peaks could be
clamped to 0 V.
A good example of the use of diodes is in a field
Waveforms
Inpul
+V
0 - -- ._ -- -- -- - - -- - - -
I
.1
synchronizing-pulse generator used in a monochrome -v

tv receiver. An example is shown in Fig. 6.8 to
gether with the wavefonns. TIle diode D1 is cut off
by each negative synch pulse and this allows C2 to
discharge through R2 - When the synch pulse returns
positive, Dl conducts to recharge C2 rapidly. The
time constant of C2 R2 is such that only a small
OUlpui
I
negative signal is generated at ~ anode by the narrow
line synch pulses. However, when the broad field Fig.. 6.7 Basic d_c. restorer circuit
synch pulses are present~a large negative signal is
generated. A diode cUpping circuit D1R3~ removes
any line synch information, and the signal is dif
ferentiated by C3RS tu produce a series of sharp
field pulses at the output.
Fig.. 6.8 Synch pulse separator circuit
o-Jl-c---'~-[>J--4~--,p--{>f--+--lI-c-"""-<> Field synch

C, pulses only
Synch 0-1 p.F
pulse Rs
inpul lOOk
o v o--------4---.c>-.---......- - - 4 - - - o 0
Waveforms'
Equolizing Field synch Equalizing
pulses pulses pulses
S ynch pul'ie input
Volloge oerass Cz
Field synch oulpul

6.4 Active Pulse Shaping Circuits
TIlese are circuits that utilize the switching properties

of transistor, tunnel diodes, FETs, etc. to reshape
1.~~
- input signals. Under this general heading we shall be
considering the Schmitt trigger circuit and mono
_stables.
A simple TRANSISTOR SWITCH is shown ill
Fig. 6.9. TIle transistor switch is often operated in
: n l·I
Ule saturated mode. lhis is when the input signal ON OFF
condition condition
overdrives the transistor and therefore switches it
+ \!e 1IUII-]ge Bose emit!er
hard ON. Under these conditions ilie voltage at the applied voltage equals
collector falls to a low value. TIlis voltage, caUed 10 baSil zero
the collector emitter saturation voltage VCE(<;at). may Fig. 6.9 Simple transistor switch
have a value from 0·1 V to 0·6 V depending upon Ute
type of transistor in circuit. The transistor is OfF
when no input signal is present, or when the input
is lower than the required base emitter voltage. +V
Under these conditions the collector voltage is high,
at Vee, and the only current flowing through the
collector load is the transistor's leakage current.
With modern silicon transistors this is very small,
Input
ov I L_
and in most cases can be neglected.
When switched to the ON position there is a finite
time before the output falls to VCE(sat); this is the + Vcc
transition time for charge 'Carriers to move through
I Minority chorge
~t
l slori~ ;''::s~~~n
the transistor. When switched OFF, however, Ulere
is a finite delay time before the collector current
ceases, because minority charge carriers. are stored in
the base region of a saturated transistor. These
charges have to be removed before the transistor
OV ---- -j------T-"'-------':: Vco holl
switches to the OFF position. The waveforms showing Turn on delay " Turn on
Iransi/i.an time
transition and storage delays are shown in Fig. 6.10.
Often switching speeds are improved by the use of Fig. 6.10 Switching wave forms for a simple transistor
(a) Speed-up capacitors: capacitors in parallel s""itch
with drive resistors.
(b) An anti-saturation modification.
The latter is a circuit that prevents the transistor
from saturating (Fig. 6.11). TIle diode Dl is used
to hold the collector voltage at a level that just
prevents saturation. When an input is applied the
collect.or voltage falls, but when the collector voltage
is lower than that of the base, the series diode (often
germanium) conducts and diverts the excess input
current into the collector. You may find circuits in
use with R1 replaced by a silicon diode.
Fig. 6.11 Anti-saturation circuit

71 PULSE AND WAVEFORM SHAPING CIRCUiTS
6.5 The Schmitt Trigger Circuit

Tills circuit, shown in Fig. 6.12, is used for level
detection, reshaping pulses with poor edges, ana
squaring sine wave signals. The Schmitt is basiCally
a snap-action switch that changes state at a specific
trip point. Consider conditions when the input to
Trl base is at zero. TT2 is conducting since it has
forward bias provided by the potential divider R 2,
R) and ~. The voltage at Tr2 base is approximately
VCC R 4
,VB:z J'>.Rz + R3 + R4
The voltage at Trr and Trl emitter will be 0·7 V Fig. 6.12 Basic Schmitt trigger circuit
less than f'B2 and this positive voltage reverse biases
Trl, thereby holding it off. The current flowing
through Tr2 is determined by
Icz.h(VBl -O'7)mA
. "Rs(kn)
noise superimposed on the inpu t signal. A detailed
and the collector vol tage of Tr 1 will be' analysis of the circuit is outside the scope ofthis book.
But the reason for hysteresis can be seen by consider- ..
VC2 = VCC-: I C:z R 6
ing that the circuit changes state at the point when the
Uusally the circuit is designed so that Tr:z is not two base voltages are equal. When the threshold level
saturated, thus allowing f~ter switching speed. or upper trip point is passed, Trl switches on and
When the input voltage is increased so that it its collector voltage falls. This means that Trl base
nearly equals the voltage on Trl base, then Tr I starts to voltage also falls, so in order for the circuit to switch
conduct, its· collector voltage falls, and Trl starts to back to its original state the input voltage to Trl base
turn tiff. Because of the positive feedback between must be reduced to a value equal to the lower vol tage
the emitters, Trl rapidly turns on and Trl off. The on Tr2 base. The effect ofthe hysteresis of the circuit
output voltage rises to +Vrr. is shown in Fig. 6.13, where it can be seen that the
A special feature of the Schmitt is that the circuit output switches back only when the input is reduced
does not switch back as soon as the input signal is below the lower trip point. A dual Schmitt trigger
reduced just below the threshold level or trip point, , is available in TTLintegrated circuit (SN 1413), and
but at a much lowerlevel. The circuit possesses hyster this is often used in interface circuits to improve the
esis or backlash, and this is very useful in eliminating noise margin.
I-Vee
Fig. 6.13 Typic;ll wave forms showing tJle usefulness of the

hysteresis in the circuit
6.6 The Monostable by the time constant CrR t and is approximately
TIle monos table multivibrator, sometimes called a td~0·7 CtR t
"one shot", is a circuit that is widely used for
Various modifications can be made to i.uprove
generating an output pulse of fixed width and amp·
the performance of the mOllostable, and circuits will
litude. TIlis output pulse is only produced when the
be found in current lise that use catching diodes,
circuit is triggered into operation by a narrow input
protection diodes, speed·up capacitors, etc. Most of
pulse. The monos table can be made using discrete
these modifications are used to improve the operating
components or is available in an integrated circuit
speed of the circuit so that an output puls(! with fast
package (SN 74121). The most common fonn for
rise and fall times is achieved. A typical be:;t figure
producing the circuit using discrete components is
for the rise time is 10 ns. The circuit shown in
shown in Fig. 6.14, but it should be noted that
Fig. 6.15 includes catching diodes which operate to
there are several variations of this. 111ese include
limit the olltput amplitude and improve speed. When
emitter coupled and compl.ementary types.
Trl switches off, its collector load has to charge any
TIle basic circuit can be seen to consist of a two
capacitive load which is formed by CL and stray
stage amplifier with resistive coupling from output
to input. As the name suggests. the circuit has one circuit capacitance. Initially the rate of riS() is fast
fixed stable state. 111is is with no input trigger but, without the diode, the total rise time would be
~elatively slow. With the diode in circuit, the change
pulse, when TIl is ON and TTl is OFF. Trl conducts
because it has forward bias provided by R t • ntis m voltage at the conector is limited and therefore the
resistor has a value low enough to provide sufficient rise time is improved.
base current to just drive Trl into saturation. The
:ollector voltage of Trl will then be approximately
J.1 V, and this ensures that TrJ is held off in a non
;onducting state. The circuit can be switched into a
I::I-~"'L..-+-4_-4-_~ R3 '-____+_] 0:',.

:quasi-stable" condition by applying a positive pulse
:0 Trl base. This need only be of short duration, 3S
ts purpose is merely to trigger the circuit into __
>peration. 111e pulse causes Trl to conduct and its Irigger r
:ollector voltage falls. This change in voltage is o-----------~----------------4-._____o ov

:oupled via CT to Trl base. Remember that the
'oltage across a capacitor cannot change instant
+lV
neously, so a change of voltage on one plate appears Waveforms' ~
Inpul pulse
s an equal change on the opposite plate. The ______...I ' - - - - - - - - - - - - 0 V
oltage on Tr2 base therefore goes negative and this
urns off Tr2' The collector voltage rises towards
'ee and, because of the positive feedback via R 3 ,
"r. is forced to conduct more. Very rapidly the
ircuit s\vitches state so that Trl is ON and Trl is
lFF, but this is not a permanently stable condition.
lle base of TIl is negative, while Trl collector is at
pproximately +0·1 V, so the capacitor CT now
harges via R t and Trl so that its right-hand plate - - - nearly -Vee
lOVes from -Vee towards +Vee' When this voltage
~aches nearly +0-6 v, Tr2 begins to conduct again,
nd the circuit rapidly switches back to its stable
:atr. The output from Trl collector is thus a positive
ulse of amplitude approximately Vee and with a
efined width. The Vvidth of the pulse is determined Fig. 6.14 Basic transistor monostable circuit
1
I
!
73 PULSE AND WAVEFORM SHAPING CIRCUITS
VI'hen testing wave-shaping circuits ensure'that the
bandwidth of the oscilloscope that you use is sufficient
for the measurement. It is good practice to use an
attenuating probe so that the capacitive loading of
the measuring leads are kept to a minimum_
With circuits that contain active switches together
Inpul with linear and non-linear wave-shapers ifs a good
idea to follow a standard sequence in fault location:
(a) Measure the power supply lines with a multi
meter. nle voltages should be near the correct
Vee> Va value and the ripple sr,.ould be low.
(b) Check that the input signal is present. Often
Va ___A'''':::;,.'''_
'Vv'C - -_
-- --
__ the input is generated from a transducer
(photocell, thermistor) or a mechanical switch.
ov l7 .
Oulpul woveform',
Since the input·actuating device is usually
positioned well away from Ule unit, it is
Time l:on5:onl of RL (C#CLI
much more .prone to damage than internal'
components.
Fig. 6.1S Circuit with catclling diode
(c) aleck that input leads and plug and socket
connections are good. Ensure that any screen
leads are properly earthed.
6.7 Fault Finding in Pulse and Waveform Shaping
(d) If the actuating device is OK, cause the input
Circuits .
signal to change state rapidly by operating the
Some of the faults that occur in pulse shaping and input device, or apply a suitable input from
switching circuits are different in form from those in . an oscillator. Then, by following the signal
oUler units, such as amplifiers and power supplies. flow from input to output, check each stage
Quite often Ule signal is degraded in some way, so until ilie signal is either degraded or lost. TItis
iliat the required wave shape is not produced. nlis will locate the faulty stage.
may happen with or without a change in d.c. bias (e) The operation of individual transistor switch·
conditions. Locating such faults requires a good ing circuits can be checked without unsolder·
understanding of the circuit function. Another ing the device. Remember that the circuit
example of the special faults that can occur is in a wiU be causing the transistor to be either ON
_ switching circui t, such as a monos table, which produces or OFF. For a transistor that is ON, momen·
output pulses or changes state when no input is tarily short the base to emitter. TIle transistor
present. TItis is called "spurious triggering" and it should turn OFF and its collector voltage
can be, in some cases, very diffieul t to locate the cause rise to Vee. If the circuit conditions arc hold·
of the problem. TIle fault may lie in the circuit itself, ing tlte transistor in the OFF state, check that
but more usually will be caused try interference. TIlis it can be turned ON by applying a forward
is "noise''; either picked up on the input leads or bias current. Use a resistor of about 10 kG
travening along the mains wires (mains borne). Such temporarily connected from the supply rail
interference is often caused by rotating electrical to the base connection. The transistor should
machines or switclting surges from heavy inductive turn on and its collector emitter voltage should
loads situated near the electronic unit. In a very fall to a low value (typically 100 m V). See
noisy industrial environment special care has to be Fig. 6.16.
taken in the design and installation of electronic .
units. This involves screens, mains mt~rs, screened
leads, and high noise immunity interface circuits. An
example of the latter is discussed later.
(a) Tronsi~lor normolly ON (b) Transi~lor norm<llly OfF
- ......--~t-- Vee - , - - - - - - - - Vee
Tem~;~~~ll~
Fig. 6.16 Rapid checking of
transistor switches
10k
fram'" Vee
10 bose
-'---+-I
Vollmeter
should reod should reod
neorl ..... v._ nearly lera
(+ 0-1 VI
. - - - - 0 -10 V Ouestions
(1) TIle circuit fails so that the output waveform is as
(0) shown in Fig. 6.18A. State with reasons the com·
...
ponentthatis faulty. How could this fadt be verified?
Ct (2) TIle circuit fails so that the output waveform is
o----{I--_--I as shown in Fig. 6.18B. State the two components
O'OII'F that could cause trus fault. One measurement is
Inpul necessary to pinpoint the fault to one (If these com
ponents.State thls measurement.
. (3) Sketch tbe time-related waveforms for input and
o------~--~~------------~-~~~
output, if the input signal were changed to pulses
of 0·1 ms duration and + I 0 V amplitUde at a
(b I Woveforms
frequency of 1 kHz.
+ 10 V r - - - - - - ,
L
(4) If the circuit fails so that no output signal is
250Hl '~"j
present although the input is correct, v.hat is the
square wove~fI~
first check you would make? Give your reasons.
2ms 4ms 6ms (5) There are three components that could fail and
produce the symptoms of fault (4). Which of the
Output 0 ------.V three is the most likely?
(6) The circuit falls so lhat the output is as shown in
-4V
Fig. 6.18C. State the component which will cause
Fig. 6.17 Waverorm shaping circuit
this fault.
6.8 Exercise: Waveform Shaping Circuit (Fig. 6.17) (7) TIle circuit faiis so that the output is as shown.
in Fig. 6.l8D. Which component is faulty?
T'nis circuit is designed to produc~ negative pulses

of 4 V amplitude with a pulse width of about 0·2 ms
from a 250 Hz 10 V amplitude square wave input.
C 1 and R J form a differentiator with a time (0)
constant of 0·1 ms. The waveform at the junction

of RJ and R'2 will therefore consist of positive and
negative pUlses. The series diode Dl removes the
positive Signals, since it 'will only conduct on the
negative spike. Diode D'2 is used as a negative clipper.
It conducts when the signal at R;zDl junction goes
(b)
OV
leI _4V~.-J.-~--
v
more negative than the voltage set up by the potential
(0) OV---------yy--------~
divider R4 and Rs. The time-related waveforms are
shown in Fig.6.17B. Fig. 6.18 F;nllts in wave form shape
/b PULSE AND WAVEFORM SHAPING CIRCUITS
WOlleforms'
Inpul150HX
squore wove
Bose Trl
Oulpul
+0-7 V
Bose Tr3
OV
-43V
Inp,,1
OV
Qulp...1
n
I
I
,~
I
I
1-Sms
+7-5V
Fig;. 6.19 MOllQstable circuit
6.9 Exercise: Monostable Circuit (Fig. 6.19)

This monostable, of standard design. produces positive to effect the input driving device. The waveform
output pulses of about 1·8 ms dUration with an diagrams shown with the circuit show that a suitable
amplitude of 7·5 V. Catching diodes are used (Dt test signal is a 150 Hz 3 V pk-pk square wave. This is
and D2 ) to limit the voltage swing at both collectors. differentiated hard by C,R:l, and the positive "spike"
You will notice that an additional transistor (Trt) is is used to trigger the circuit into operation. The
used to buffer the input, tIlereby preventing any operation of the circuit has already been described
changes of state in the mOJlostable being fed back in a previous section of this chapter.
Ouestions
(1) Which two components determine the width of
the output pulse"?
(2) What is the purpose of Cl '! (6) State the possible effect on the circuit for tIle
(3) In the stable state v.':ith no input applied, what following component faults:
voltages would you expect to measure with a 20 kQ/V (a) Cl open circuit
meter at all test points? (b) Tr I base emitter short circuit
(4) How would you test that Trl operated correctly (c) Trl collector open circuit.
without removing it from the circuit'! (7) The circuit fails so that the output pulse all1ljlitude
(5) In each of the following cases the monostabJe increases to +20 V although the pulse width remains
fails to produce an output when the correct input is at 1·8 ms. State, with reasons, the component fault.
applied. State which component (or components) (8) How could the circuit be modified to produce
is at fault. (+ve means «just +ve".) negative going output pulses?
1 2 3 4 5
A +ve 0 0·7 0·1 6-2

B +ve 7·0 0·7 0·1 6·2
C 0 7·0 0·7 0-1 6·2
D +ve 0·7 0·7 0-1 0
E 0-7 0·13 0 7·0 6·2
. 76 ELECTRONIC FAULT DIAGNOSIS
6.11 Exercise: Logic Interface Circuit (Fig. 6.21)
The description "interface" is usually applied to a

circuit wllich links a digital electronic unit to some.
out'lide switching signal St,Ufce. The signal may be
derived from a transducer or, as in this case, from a
mechanical switch. Strictly speaking, interface units
are only those useu with computers, but ':he teml is
now applied to cover circuits used with most digital
electronic equipment.
The purp<lse of the circuit is to take the incoming
Switching signal and to change it into a form that can'
be used to drive the internal logic. A furthH important
Fig. 6.20 Schmitt trigger circuit requirement is that the inte.rface must possess good
I
noise immunity. This is so that noise Signals picked
up on the input lead can be rejected. For logic
systems in very noisy electrical sitUations the inter·
6.10 Exercise: Schmitt Trigger Cj~cuit (Fig. 6.20)
face unit is usually a differential receiver. The
The circuit is designed to produce square waves from switching signal is generated by a differelltial driver,
a 1 kHz 1 V pk-pk sine wave input. TIle potential and then carried along a twisted pair of ","ires to the
dividerRt. R2 sets up a voltage on the base ofTr I interface receiver {Fig. 6.22). Any noise signal coupled
: that just exceeds the threshold of the circuit. Thus on to the input wires will be ill common mode and
in the stable state Trt conducts and Tr2 is held off. . will, therefore, be rejected by the differential receiver
Only when the sine wave input actually falls below amplifier.
about -0·2 V will the circuit change state with Trl Where a single signal wire has to be us(,d, the
OFF and Tr2 ON. The circuit continuously switches circuit in this exercise does provide a high degree of
back and forth as the sine wave input changes, but noise immunity, and also has mter circuits to remove
since the switching is so rapid a square wave output very high frequency noise. Lh C1 proviue the first
is produced. The amplitude is about 5 V. low pass mter; additional filtering comes with R I C2 •
The d.c. level from the switch is used to {,rive a
transistor inverter, and the output from t.,is inverter
Ouestions triggers a Schmitt to provide a fast logic edge to the
(1) Calculate the voltages you would expect to internal circuits. TIle threshold at the intt:,rface input
measure with a 20 kll/V meter on the 10 V range is arranged to be +15 V, This is tJle d.c. level at
at each of the test points. Assume no input is applied. which the output changes state. You can readily see
(2) What effect on d.c. voltages and output waveform that the transistor \vill start to conduct when the input
would result if RI went open circuit? is about +15 V, because then the base voltage will
(3) In each of the following faults the circuit fails to just start to go positive. The switch voltage is +30 V
produce an output signal when the correct input is for logic" J" alia z.ero volts for logic "0". A noise
applied. State, with reasons, the faulty component. pulse must be equal or greater than 15 V to cause the
transistor to change state.
1 2 3 4 5
Diode Dl is included to protect the base emitter
Fault
junction against excessive reverse voltage. Thls may
A 1·5' 8·1 1·8 1·15 6·4
be caused by a large negative noise pulse or if the
B 1·45 0·95 0·4 0·85 12
s\vitch sigllallille became open circuit.
C 2 8·15 1·7 1· 7 12
D lA 0·85 0·8 0-8 12
E }·6 1·6 0·2 0·9 12
F 1-45 0·95 0 0·85 12
I
77 PULSE: ANLJ WAV!:.I-UHM SI-iAPING ClHCUIIS
-1--+5V
I
~-30V
1 A 5i9nol
~) wire
o YSWilC h
17k
-15V
Fig. 6.21 Logic interface unit
Dolo jW;Sled poir
Jl
line line
receiver
driver
Fig. 6.22 Typical arrangement of a differential line driver

and receiver for use in very noisY electrical environments
Ouestions
(1) How could you quickly check that the transistor (4) Write down the voltages you would expect to
switch is operating correctly without operating the measure usingastanqard multimeter if the input signal
external switch? . lead became open circuit. .
(2) What would be the effect on the circuit if R3 (5) Write down the voltages you would expect to
became open circuit? measure if the base collector junction became short
(3) In each of the following cases the interface fails circuit.
to produce an output change of state when the external
switch is operated_ State which component is faulty.
1 2 3
Logic 1 30 -0-6 5
A
Logic 0 0 -0·6 5
Logic 1 30 0-7 0
B
Logic 0 0 -0·6 0
Logic 1 30 0 5
c Logic 0 0 0 .5
Logic 1 30 5·8 5
D
Logic 0 0 -0·6 5 II

,-------<1> -----'-'1>-----_-- -1-10 V
Output
8 A
ONo OFF
Switch
Fig. 0.23 FET input Schmitt t{igger circuit
6.12 Exercise: FET Input Schmitt Trigger Circuit 10 - 3 )

!Fig. 6.231 V2 = ( 12.2 kul x 10 kU = 5·8 V
.One of the major disadvantages of the bipolar tran Here the zener is assumed to have a 3 V hreak
isistor version of the Schmitt trigger is that, when the down. In practice there is a wide tolerance (·n zener
input transistor conducts, the input impedance of breakdown voltages.
the circuit falls to a relatively low value. This is Voltage at TPI (drain of Trl) is
because a small base current is required for the input
transistor. By using a field effect transistor a high VI = V2 + VDZ =8·8V
value of input impedance can be achieved for both
states of the circuit. Voltage at TP3 (Trl source and TI2 emitter}will
In the example an n-channel junction FET (type be 0·7 V less than the base voltage ofTr2'
2N3819 or equivalent) is used. With the switch
10sed C1 is discharged and the gate potential will
Ie zero. At the same time the source is held positive This voltage is sufficient to ensure that the FET is
lY a voltage greater than the cut-off voltage of the fuHy cut off.
~ET, thus ensuring that the FET is non-conducting. Tr2 is conducting, so its collector vol tage f.111s and
:he circuit is wired conventionally with the source of turns on TT3' Note that Tr3is connected as a uturated
~r] connected to the emitter of Trl, and this tran switch so the voltages atTP4 and TP5 should be9·3 V
istor is forward biased by Rl and DZ 1 • Under these and 9·9 V respectively.
onditions the voltages at all the test points can When the input switch is opened, C 1 can charge via
~diIy be estima ted using the following formulae R) ,so the gate voltage of the FET rises positive. At
nd at the same time making any reasonable
some point the bias voltage between gate and source
iSumptions.
has fallen to a value that allows a small drain current
TIle voltage at TP2 is given by
to start flowing. 7he voltage at the drain fars and so
does the voltage at Tr2 base. TIle positive feedback
v:!!:(VCC-VZ)R in the circuit ensures that Trl rapidly turns ON and
1 R2 +R4 4
TTl OFF. Consequently Tr3 also turns OFF since it
Here we have neglected the effect of Trl base no longer has forward bias sUfplied from TT2 :::oUector.
ment and also the slope resistance of the zener TIle output faUs to zero volts.
ode. Both these would only have a marginal effect Because of the very high input impedanc-~ this
1 the calculated value, but would make the calcula circuit can be llsed to pmvide a long time dday
m much more complicated. between the operation of the input switch and the
79 PULSE AND WAVEfORM SHAPING CIRCUITS
resulting change of state al the output. With the

values of R! and C1 as specified the time delay is
about 5 sec. Delays of up to several minutes are
possihle by increasing the values of RI and C1 • Note
that C 1 should he a type of capacitor that has low
leakage such as a tantalum_
One of lhe drawbacks of this circuit is that the
value (If cut-off voltage for the FET has a wide
tolerance and this means that the threshold level is
difficult to define accurately.
TIle voltages at the test points were measured
using a standard multimeter for both positions of the
switch. In the latter case the readings were taken after
a 10 sec deJay_
Nonnai readings
1 2
Switch pos. A 8·5 6·1

Switch pos. B 3 1·2
Questions
(I) lfit is required to measure the ga te-to-source (5) ll1e circuit operation changes so that the delay
bias voltage, what type ofvoltilleter should be used? time increases and is erratic. The readings are as given
(2) The circuit fails to change state when the switch below. Which component has failed? 'Give reasons.
is made to B, even after several minutes. The voltages
1 2 3 4 5
remain as for the normal readings in switch position
A. Which components could cause this fault? In each Switch pos. A 9-8 8·2 7·7 9·2 9·9
case describe the type of fault and a measuring S\Vitch pos. B 3 1·8 2·5 10 0
metllod that would confirm it. (6) IJ the switch was situated remote from the cir
(3) Describe the effect on the circuit operation and cuit, describe a simple method for checking all the
on the bias voltages of a base emitter short in Tr3 .
active components without unsoldering any leads.
(4) In each of the following cases the circuit fails to
change state after the operation of the switch_ From
the readings detenlline which component (or com·
ponents) is faulty and the type of fault. In each case
give a supporting reason.
1 2 3 4 5
----
Fault X Switch pos_ A 5 0 l·g 10 0
Switch pos. B 3·1 0 2·6 10 0
Fault Y Switch pos. A 9-8 9-2 8-5 9-2 9·9
Switch pos_ B 8-6 9·2 8·5 9·2 9·9
Fault Z Switch pos. A 2·5 0-8 0 10 0
Switch pus. B 3·1 1·2 2-5 10 0
I
7 Thyristor and Triac Circuits
TIlyristors, formerly called silicon controlled recti One of the aids in understanding the operation is
fiers (SCRs), and triacs are semiconductor devices the two transistor equivalent circuit (Fig. 7.2). By
tJlat act as high-speed power switches. Devices are dividing the thyristor diagonally it can be seen that
ayailable that can operate a1 potentials of several a p-n-p transistor structure exists between anode and
l:w,ndred volts and which will carry currents gfup to gate, and an n-p-Il transistor in the gate cathode
hundreds of amperes. TIlcse solid state units are now region.
being increasingly used to replace conventional
mechanical switches and relays since they offer faster
Switching and greater reliability. lltis is particularly p
the case in continuous a.c. power control systems n -t...."

, #
n
G '-"
such as lamp dimmers, heater control, motor speed p 1==:' P
control, etc. This is a growing and important branch
n
of electronics, so tJle understanding of tJle opera tion
of the devices, their use, and fault diagnosis is also K
important. We shall start by outlining.the basic
Fig. 7.2 Two tran~istor equivalent circuit of-a thyristor
operation.
'DIe operation caP hI( divided under t!-e fqllowigg
7.1. Principle of Operation of the Thyristor ~p-hl(adings:
TIle thyristor has a structure (Fig. 7.1) tJIat consists (a) Reverse kips angde pegative with respect 19
of a four-layer p-n/p·n silicon sandwich. The symbol £Mhgd¥o TIle thyristor is in reverse blockiug sta te and·
is that of a rectifier with an additional terminal called only a low-value leakage current flows. Lllder these
the GATE. 1t is this gate which enables the action of conditions both junctions 1 I and 13 are reverse
the rectificr to be controlled. TIle device can be made biased.
to act as an open circuit (forward blocking) or it can (b) Forward bjm, aDRde pmjtjye wjth respect loU
be triggered into a low resistance forward conducting calhode but no gate sirma!. Thyristor is s:lid to be
stat~ by applying a short pulse of relatively low
forward blocking since it acts as a high resistance.
power to the gate terminal. Only a small leakage current nows. It can be seen
Anode lead
that although J I and 13 are forward bias.~d. the
centre junction 12 is reversed. By referring to the
Anode equivalent circuit one can explain forward blocking
by the fact that, since the gale has no signal applied
to it, Tr2 is cut off. Only a small leakage current
can flow.
Gale
Colt>od~ Gate (c) forward hias with gate Signal applied. I r a pulse
Gale of forward bias is applied between oate and cathode
lead while the al10deis positive with respect to the cathode
lhen the thyristor ,vill be forced into Cll!' duction.
Fig. 7.1 Construction of a thyrislor 111C switch-on time is rapid (microsecs) 3nd a large
80
81 THynlSTOR AND TrilAC CIRCUITS
Forward cUr(en~
current C;1I1 be passed by the device, limited only hy
the external resistaHce. The anode to cathode voltage Amps
(;.llIs 10 a low v;:iuc, typically I V. 'nlis action can be
expbined usilli: the equivalent cilcuit by noting thai
,..,/' Forword conduclinq
;1 pulse of forward bias turns on Tr2' TIlis transistor
Forward breok-ove;
starts to conduct and therefore swi tches on Trl . 'llle voltoqe
Raiding curren!
two transistor circuit has a positive fcedl><lck loop, ______________ _il
Reverse vaUage
since each has its collector 'l.vired to the other base.
~--~~~===-~-r-===r===~==~~
111erefore the two tr<lllsistors rapidly switch on, and Forword voUoge
...., Forward blOCking
will remain on even when the gate signal is removed. 'Reverse breakdown
region
TIle device call only be turned off by reducing the

anode current below a value known as tlle '<holding
current".
Since the two transistors are cOllnected as a posi Re~r:se currenl
tive feedback pair there is a large current gain between
Fig. 7.3 Typical static characteristics of a thyristor
the gate (TTl base) and the anode (Trl collector and
Tr2 emitter c~rrel1t). The voltage gain is also high
since only about I V is required between gate and Another point worth noting concerning circuits
cathode to trigger the thyristor on. l1le amount of using high-speed power switches such as thyristors
gate power required to turn on a typical thyristor is is that, since the s\vitch-on is so rapid, switching
thus relatively small. In fact, milliwatts of gate power "spikes" can easily be generated. Therefore most
can be used to control hundreds of watts ofload circuits, especially a.c. controllers, include RF filters
power in lhe anode circuit. to limit the amount of RF interference on the a.c.
Mention has been made of the important fact power line .
. that once triggered the thyristor remains conducting
even when the gate signal is removed. It can only be
switched off by reducing the current Howing through
it to a value less than the holding current. 111is is the
specified minimum current that will ensure continued
conduction, and it is usually a few per-cent of the
maximum forward current. In a_c. circuits tile
thyristor turns orf every half cycle when the supply
voltage p<lsses through ?em and goes negative; this
gives automatic turn off. In d.c. circuits special
techniques are used to reduce the anode current to
z.ero 1.1..1 switch orf the device.
There ,He two other conditions, apart from gate
Signal, that will switch a forward blocking thyristor
into conduction:
(a) Exceeding the' forward brea.kover voltage.

(b) By applying a fast rising voltage wave form
between anode and cathode, typically greater
than 50 V per microsec. l1~is «ratc-effect" is
explained by the f,lct tha tin ternal capacitance
(J 7. JUTlction) can feed a part of a sharply
rising anode voltage through to the gate. This
turns the thyristor on.
Both these erfect., are undesirable.
7.2. Application of the Thyristor
in
A simple d c power cQntrol circuit is shown Fi~
7.4. Dy operating switch 1 the thyristor is turned on
and power is dissipated in the load. TIle capacitor
R,~ad
1 --RL
C, will be charged via Rii and ths thyristor, so that sw, -I . A

its right-hand plate becomes positive. Dy moment ON
arily operating switch 2 the capacit':;r will be con·
G
nected across the thyristor, thereby reverse biasing
it and causing it to turn off. This is called capacitor
commutation.
Alternative arrangements for switch off are:
(0) To wire a push·to-break switch in the supply ~------~--------~--------o-
line. Fig. 1.4 A simple d.c. power control circuit usiltg a thyristor
(b) To replace switch 2 in Fig. 7.4 by another
thyristor. In this case a bistable circuit is
formed. Such a circuit (Fig. 7.5) can be used
as a high-speed trip.
One of the important uses of thyristors as in the
smooth control of a.c. power (Fig. 7.6). The average
power in the load can be varied by adjusting the
time position of the gate triggering pulse relative --0
to the supply waveform. Conduction angles from La" fevellrip inpul

0 0 will turn SCR, OFF
10 to 170 are possible. As the supply voltage goes
positive C 1 is charged via R J and the variable resistor
RV:2. When the potential across C} exceeds the break·
....·over voltage of the trigger diode (diac) then the diac ~----~------------~~---~
conducts and supplies a gate pulse to the thyristor. Fig. 1.5 Using a slave thyristor as the tum-off device in a d.c.
circuit
TIle thyristor switches on, and nearly the whole of the
supply voltage is applied across the lO<ld. Dyadjust
ing RV:2 the charging time of C1 can be varied and L
this in turn varies the conduction angle. The diode
DJ is included to prevent reverse bias being applied
to the thyristor gate, and at the same time it ensures SCR,
stable triggering by discharging CIon each negative
half cycle.
Decause of its low efficiency this simple half
n
wave circuit is not suitable for most application.
Full wave controllers using a bridge rectifier (Fig. 7.7)
Waveforms
with two thyristors, or better still one triac, being
preferred. Mains supply
Dioe CUTfent
+RJL
Lood voltage
f
SCR,
'F
SCRI
Conducti(ln angle
ON OFF
Fig. 1.6 Half wave a.c. power control using a til yristor
t5J I HYHI~ 1UI, l\NU I HIAC UtlCUII::'
SCRI
n o----+-----'IF'
Fig. 7.7 Full wave a.c. controller using a thyristor
I
~~~:~~ ~~~~~~inq
_!:~~,:~_b_,,:,:king
~:~R~ev~e~rs;e~b~IOC~kl~ng~r-__~~~~~
~:;;;;;: -----------,
~r,::~d
Reverse
conducting
Reverse
current
Fig. 7.8 Basic construction and static characteristics of a

triac .
7.3 Basic Operating Principle of the Triac 1+ mode == MT2 current +ve Gate current +ve
(- mode == +ve -ve
The constrtlction of this device is shown in Fig. 7.8
m+mode = -ve +ve
where it can be seen that it may be regarded as two
nr mode = -ve -ve
thyristors connected in inverse parallel, with a com
mon gate terminal. TIle operation is similar to that TIle highest sensitivity is achieved in the r- and
of the thyristor. nr modes, these being about equal and twice that
TIle triac will therefore pass or block current in of the other two modes.
both directions and it can be triggered into conduc As for thyristors only a small amount of gate
tion in either direction by positive or negative gate power is required to turn the device on which will
sign:tls. The characteristics are also shown in Fig. 7.8. then control large amounts of power in its MT2
11\e four triggering modes are signified as follows: circuit.
84 ELECTRONIC FAULT OIAGNOSIS
Lo----£:?TL~
rv [j--' flV\
- DICe
C
N o~__________~~________2_-I.-____________~
Waveforms:
- Mains
- supply
Diac. current
Phase shift
from RC networ~
Load volloge
Conduction
angle
Fig. 7.9 Typical triac full wave ac. power controller
7.5 Fault Conditions and Fault Finding in Thyristor

1.4 Applications of the Triac and Triac Circuits
Triacs are mostly used in fun wave a.c. control As stated previously the thyristor and triac are
circuits since that is the area in which they offer two basically four-layer silicon devices used to control
distinct advantages over two thyristors: large amounts of power. As with most otl'er elec
(a) Simpler heat sink design. tronic devices, controlled rectifiers faillarge1y for
(b) Relatively economical trigger circuits. thermal reasons. High temperatures in the re.)atively
A typical triac a.c. controller with diac triggering smal.l volume of the device, or a high rate of tempera·
is shown in Fig. 7.9. TIle conduction angle is controlled ture cycling, causes slow deterioration which
by the potentiometer, and the gate recei.ves positive ultimately leads to failure.
current on the positive half cycle of the supply and Controlled rectifiers can also be destroyed like
negative current on the negative half of the supply fuses if they are subjected to overload current surges.
(t and III-modes). The waveforms show that full Naturally it is the designer's job to ensure that the
- wave control is achieved. device is mounted on to an efficient heat 'link and
that overload current surges do not occur.
The faults that can occur are low volta~e forward
breakover, loss of gate control, open or short circuit
anode to cathode, or opell or short circuit gate to
cathode.
It is also possible for thyristors and tri;,cs to fail
by wh.at is called di/dt failure. This is explained IJY
considering the instant at which the device is trig·
85 THYRISTOR AND TRiAC CIRCUITS
gered. Gate current is concentrated in a very small TABLE 7.1 Typical Faults Qn Thyristor Circuits
area {If the gate region. As a result the initial flow FAULT RESULT AND SYMPTOMS
of ::mode current is constrained into a small area, and Gate to cathode. open 111yristor OFF and cannot
if this anodc current has a rate of change (di/dt) that ~cuit be triggered into conduction.
exceeds a critical value, then a large amount of heat Measured gate signal high.
will be gencrated in a small area (hot spot) and the Ga!e to cathode short Thyristor OFF and cannot
circuit be triggered into conduction.
device will fail. In most circuits the load inductance
Measured gate signal is zero.
·limits the rate of change of current.
Mention has already been lII;:tde of the d)'!de effect ,,\no~e to cathode short Thyristor conducting in
. , ' ClICUlt both forward and reverse
whereby a thynstor can be tnggered on by a sharply directions.
rising anode voltage. You will find some circuits r " Measured volt drop between
anode and <:athode is zero.
with an RC network wired in par:a1lel with the
thyristor or triac to elelOinat~ this erfect. Anode or cathode open Thyristor OFF.
circuit
11le effects of various faults can be nlQre easily
understood by using the tw(') transislor equivalent
Note that if the actual leakage curre1tt of the thyristor
circuit.
in the forward blocking mode is required then meter
In most cases the symptOJllS and the voJ tages in
2 should be disconnected.
the circuit will indicate the fault. But always remem
Next depress switch 2 for a short time; this
ber that the device may be tised in a d.c. or an a.c.
should trigger the thyristor into conduction. Meter I
circuit. For example a thyristor may be used to con
should indicate approximately 100 rnA and meter 2
trol the power in a load connected to the 240 V
apprOximately 1 V. The latter is the forward voltage
.50 Hz mains with II d,c. signal applied to its gate.
drop.
aleck with the circuit diagram before making any
To obtain a value for tbe minimum holding
measurements.
current gradually increase R2 until the thyristor
Witllsome circuits it is possible to test the thyristor
turns off. The current indicated just before tum off
or triac wllile it remains in circuit. For example it
is the minimum holding Cl:lrrent.
is quite reasonable to inject a suitable trigger Signal
. TIle circuit should be modified for tests on high
into the gate to detemlille if it is tJle tJ\yristor or the
power devices by using lower value resistors.
gate signal source that has failed in a circuit. Also,
with the power Sllpply OFF, measurements witil an
ohmmeter can be made to test for short circuit anode
to cathode or for open or short gate to caUlode. TIle
gate cathode junction has the same characteristics as
a diode. A resistance of about 500 .n should be
indicated in the forward direction (gate +ve \V.r.t.
cathode) and a high resistance (greater titan 100 krl)
in the reverse direction. Remember that, with the
device in circuit, other components in parallel with Fig. 7.10 ~d.c. test circuit for lnedium power thyristor
the gate will affect the readings.
A simple circuit can be used to check the operation
of a thyristor (or triac). TIle test circuit of Fig. 7.10
can indicate gate operation, forward leakage current,
forward voltage drop, and minimum holding current.
With R2 set to minimum S, is closed. TIle indica
tion of meter 1 should he very tow (typicalty less
thall 50 pA) and meter 2 should indicate nearly 12 V.
86 ELECTRON IC FAULT DIAGNOSIS
r-------~----~----~--~------~---< +12V
J
Reser
Fig. 7.1 tAla fm unit
RVI
25k
L-----4---~--~----~--~----------_oOv
7.6 Exercise: Alami Unit-(Fig. 7.11) (A) The light beam is cut off but the alarm fails to
i 11us circuit shows the use of a thyristor in a d.c. operate.
I mode. If the light beam is interrupted. the thyristor TP 3
is switched on to operate the alarm buzzer. !!l.e MR 3·7
~yzze( wuld ngrmally turn the thyristor off since
II switches open every time it operates, but a holdinE; (8) The light t,,,.. 1il is cut off but the alarm fails to
~urept path is provided Via &;. 111e diode D, is operate.
included to protect the thyristpr from the large TP 2 4
back e.m.f. spikes generated as the buzzer switches. MR 1·3 o
Trl and Tra form a differential amplifier which
~lifies the difference signal between IT? and II? (C) The alarm operates continuously and cannot be
~. Tra base is held ata constant voltage by tlJX reset.
potential divider R} and Rs. while the input to Tr. TP 3 4 5
base depends upon the resistance of the photocell
and the setting of RV I. When the light beam is fully
MR 13\ '1 2
3·7 I 3·7 0 I I
0
on the photocell, this has a low resistance so that the (D) The alarm fails to operate when the li t )1t beam
voltage on Trl base is lower than that on Ir., base. - is cut off.
Trl conducts and Tra is held Off. The collector volt
age ofTr2 is at zero volts and therefore no switching
signal is fed to the thyristor gate.
TP
MR I 4~2 I
2
4·8 I ~.5
I I 4
0 I ~2 I
V,'hen the light beam is interrupted the resistance (E) Interrupting the light beam fails to trif;ger the
of the photocell rises, and tllis increases the voltage alarm.
on Trl base. Trl stops conducting and Tr2 conducts TP I 2 3 4
since its base is now at a lower vol tage than the base
of Tr l ' The collector voltage of Tr'2 rises and supplies
MR o I 0·7 3·7 o
a signal to the gate of the thyristor. 111e thyristor (F) The alarm operates continuously and cannot be
turns on and operates the buzzer indicating that'the reset.
. light beam has been momentarily broken. The TP 2
. circuit can be reset by operating switch 1. MR 1·8
(4) If the alarm operates once, then fails to hold,
Questions when the light beam is momentarily cut, which com
(l) State the voltages you would expect to measure
ponents could be considered faulty? What simple
\,~th a 20 kn/V meter with the light beam 011 and off
method coulu be used to verify, and to locate, the
at the various test points.
fault to one component only?
(2) What would be the symptoms for an anode to
(5) Write down the series of simple, rapid tests that
cathode short circuit of the thyristor'?
could be made to verify the operation (If various
(3) In the following faults the voltages were measured cumponents in the circuit when no l11u!timeler is
using a stanuard meIer Siate with a supporting available.
reason the faul ty component or components.
17 THYRISTOR AND TRIAC CIRCUITS
Fig. 7.12 Lamp dimmer
You can readily calculate that when RV 1 is nearing

'.7 Exercise: lamp Dimmin9 Cifcuit (Fig. 7.12) its maximum value that the voltage across C3 just
reaches the voltage required to trigger the diac. This
\5 explained previously in section 7.4-the triac is an
coupled with the near 90° phase shift means that the
xcellent device fN full wave a.c. power control. 0
triac firing signal is delayed by nearly 170 , and the
:bis circuit uses the phase triggering technique, in
lamp therefore has power applied to it for only 10°
vh.ich RV 1 and C2 provide a variable potential
of each half cycle.
livider and variable phas~ shift network. TIus feeds
phase shifted Signal to a "slave" network R2C3 •
Yhen the voltage across C3 exceeds about 35 V the Ouestions
Hac triggers to partially discharge C3 into the triac (I) What is ilie purpose of Ll and C1 ?
:ate. TIle triac then conducts and power is applied (2) If the r.m.s. current rating of the triac used is
o the lamp. The purpose of the «slave" network is 6 A; what is the maximum load power that can be
o prevent any large change of voltage occuring controlled?
cross C2 when Ole diac triggers. (3) What would be the symptoms if RV 1 wiper
TIle conduction of the trinc can be controlled became open circuit?
Iver a wide angle by adjusting RV I . For example, (4) A fault exists such that the lamp is at full bright·
¥ith RV 1 set to minimulll very little phase shift or ness and cannot be controlled by adjusting RV 1. TIle
)otential divider action takes place and the voltage a.c. Voltage between TPI and the neutral line measures
.cross C2 (and C3 ) almost follows that of the a.c. zero. State the component failure.
upp\y raiL TIle diae therefore passes a trigger pulse (5) TIle lamp fails to light. The a.c. voltages, with·
o the triac gate shortly after the start of each half RV 1 set, to m)nimum, are as follows: .
:ycle of the supply. You can see that almost full
I~I
lower is applied to the lamp. TP
When RV I' is adjusted to near maximum value the A.C. T.m.S. voltage 235
.hase shift in the RC network approaches 90°. From
State the component failure. How could this fault
,asic a.c. theory the phase angle for an RC network is be quickly verified?
tan (j = wCR (6) TIle unit fails so that the control over the lamp's
brightness becomes limited. TIle lamp burns at high
)0 with R ~ 250 kn, then tan (I = 21T50 x 0·15 x
briglltness with RV 1 at minimum but only reduces
10-6 x 250 X 10-3 and 8 ~ 85°. slightly in intensity with RV 1 at maximum. \ly1lich
, component has failed? How could this be quickly
:be attenuation in the network with R = 250 kn can checked?
Ie calculated from
(7) What symptoms would indicate a gate to cathode
)J = l's ¥, open circuit'? How would they differ [rom those of a
c Z· ( gate to cathode short circuit?
Nhere impedance Z = yI(R2 + Xi) and Xc = 21T~C' (8) The lamp fails to light at all. The voltage at TPI
is zero. Vt1hich component has failed?
OJ
CO
m
r
m
(')
-i
:0
o
Z
()
11
P
c
r
-i
2·2 }Ar 2'21'r 2'21'F o
c.. Cs Cll
»
G)
z
o
(/)
+12 V (/)
Storr
SW 1
11
Rl~
"'\
R2
11:
cv
°1
~----~----~~~----------------~
Fig. 7.13 Sequential control unit
NB C., C, and C. must be no.n~peiariz.ed types
89 THYRISTOR AND TFlIAC CIRCUITS
7.0 Exercise: Sequential Control Unit (Fig. 7.13) Questions
In many process control situations II circuit is reo (\) What type ofinstrum~llt should be used to measure
quired that can switch Oil various loads in a defined the voltage across Ct. C2 or C3 ?
sequence, each load being Oil for a conI rolled period (2) If the trigger voltage of the unijunctiGil is assumed
of time. For example a ,equcnce of control may he to be apprOXimately 7 V calculate the minimum and
required as foll(lw5: maximum operate time of a load.
(3) The unil fails so that it will neither recycle, nor
Start LQad \ Operllte belt mechanism for
switch off power to load 1.3 at the eild of the sequence.
5 sec to move work into
Pressing the start button will however initiate
position.
operation. State which portioll of the circuit is at
Load 2 Spray work for 2 sec.
fault and list the likely component failures.
Load 3 Heat for 10 sec. Then stop.
(4) Suggest a circuit modification to ensure that
In the circuit heing considered the thyristors are pressing the start bu tton switches SCR t 011 only when
used to switch power to the various loads and ul1i· thyristors SCR2 and SCR3 are OFF.
junctions are used to provide the time delay. The (5) TIle unit fails so that power is applied to load 2
circuit consists of three identical stages. and load 3 as soon as IQad I is off. Loads 2 and 3 also
When power is applied none of the thyristors switch off at the same instant. State, with reasons;
conduct since they do not receive gate signals. By Ihe component fault and the type of failure.
pressing the slart switch a gate signal is provided for (6) 11le unit fa.ils so that it will not recycle. Operat
SCR 1 which switche.<; into forward cQnduction and ing the start button will initiate the sequence and
connects power to load L,. Since the anode vol tage the unit stops correctly with SW 2 in position B.
of SCR 1 falls to about +1 V, Trl • which is a-pilJJ Which component has failed?
transistor, is forward biased via Rl . 1111S tran~istor (7) A fault exists so that load 3 is always energized
::onducts and C 1 is charged via Rl and RV, towards when power is applied. State the probable component
the positive line. When the voltage acrO$S C, equals failures and show how one voltage measurement
the emittel trigger voltage of UJT 1 , the unijullction could be lIsed to locate the fault.
::onducts and a positive pulse is generated across R6 (8) State fully the symptoms for the following faults:
to fire thyristor SCR2 . TIle anode voltage of SCR1 (a) Tr2 base emitter open circuit
falls and Utis feeds a negative edge via ~ to the (b) C2 open circuit
1I10de of SCR 1 to turn off SeR I _ Power is applied (c) C6 open circuit
to load Ll for a time detelmineq by RV 2 and C2 . (d) SCR 1 gate to catllOde short circuit
T1lis part of the circuit is identical to that already (e) SCR 3 anode to c3thode open circuit.
lescribed so that when UJT2 triggers, SCR3 conducts
lOd this switches off SCR z .
At the end of the sequence when UJT 3 triggers
)n, the circuit call be made to automatically recycle
:Jy placing SW2 to position A. Then the pulse from
UJT3 is fed to the gate of SCRI' With the switch
5W1 in position B, SC~ is made to conduct and this
;witches off SCR3 to end the sequence. 111e cycle
:all be restarted by pressing the start button_
90 ElECTROh!IC FAULT DIAGNOSIS
-1-12 V
02 IN914
1N914
OV
Fig. 7.14 Lamp flasher unit
7.9 Exert:ise: lamp Flasher Unit (Fig. 7.14) of the circuit is given by the lamps. TIle lamps them
Olis unit comprises a low-speed unijunction oscillator selves are the components with the highest failure
...hich is used to drive a thyristor bistable circuit. rate, so the service engineer would check 1hese first.
:.ach pulse from the oscillator is steered 'Via a diode Assuming the unit has faiied with both lamps out, a
:ircuitto the gate of the thyristor that is off. first check would be to short the anode tc cathode
Initially when power is applied to the circuit of each thyristor in turn· to test the lamps. A later
.oth thyristors are off and although positive pulses question concerns the next set of checks that would
re generated from the unijunction neither thyristor be made. .
; triggered since diodes Dl and D2 are reverse
·iased. Depressing the start switch causes SCR 1 to Questions
onduct and lamp LPI lights. TIle anode of SCRJ is
In all cases you can assume that both lamps are
len at approximately +l V while that ofSCR2 is at
functional. .
12 V. TIle next positive pulse from BI of the uni
(1) Calculate the approximate ~;ximum and mini
mction is then sufficient to forward bias Dl but not
mum frequency of the unijunctioll oscillator.
)1, and a triggerpulse is supplied to the gate of
(2) A fault exists so that both tamps are Ii! and
CR 2 - This thyristor switches into a forward con
remain on as soon as the start button is pressed. TIle
ucting state to light lamp LP 2 • Ai the same time a
voltages measured with a 20 kil/V rnultitr:der are
egative step is transmitted via the commutating
as follows: .
lpacitor C 3 to reverse bias and tum off SCR1 • TIle
::xt pulse from the oscillator win be steered via DI
I trigger thyristor SCR 1 • This lights [PI and at the
TP
MR
I 0·7
1 4
o
me time··turns off SCRl' The lamps therefore nash
(3) TIle unit fails with both lamps unlit. List in a
1 and off alternately at a frequency deternlined by
logical order the tests that should be mad{! to locate
.e unijunction pulse generator.
the fault.
Fault finding in this type of circuit is u.~ua\ly
latively easy since a visual indication of the state
91 THYRISTOR N·IO TRIAC CIRCUITS
(4) State, with reasons, the symptoms for the foHow Vi) Suggest a circuit modification so that the unit
ing component failllre~_ wotlld self-start from a power line switch_
(a) DJ open circuit. (7) If it was found that the failure rate of the lamps
(b) SCR2 gale to cathode short circuit. was too high for a particular application, are any
(c) C3 open circuit. modifications of a very simple nature possible to
(d) SCRI anode to cath.ode open circuit improve the situation?
(5) For the following fault conditions state, with a
supporting re:'ison, the component or components
tllat are faulty_
I 2 3 4 5 6 7
Fault
A 0-7 0-85 12 0 0-1 11-8 0
n 0 12 0·85 0-7 Varying Varying Varying
C 0 12 12 0 Varying Varying Varying
D 0 12 0-85 0-7 5·2 5-2 5·8
E 0·7 0-85 12 0 Varying Varying Varying
F 0-7 0-85 12 0 0 0·1 I
0-85
f'S1
L opposition to the voltage across the motor, and this
back-c.m.[ has a magnitude that is proportional to
the motors speed. The faster tll.e armature rotates
rv
240'1
the greater the back-e.m.f. and the smaller the
required motor current. TIlis is because the current
50Hz
taken by tlle motor is proportional to the difference
between the applied voltage and tll~ back·e.m.f. When
the motor is first started a large current is taken since
""\f j +-----;I-~--' .the back-e.m.f. is zero_ This mealis that a high torque
is developed and the motor rapidly increases its
speed. As it does so the back-e.m.f. increases, the
2A current falls, and the available torque is reduced. Now
No---......- - - - - - - - - - - - ' if a load is applied to the ll)otor its speed falls .
initially and this lowers the back·e.m_f. Thereft,re the
Tr, 8C212l
Trl! eCI82l
current increases and this tends to restore the motor
to its original speed. TIlis type of motor then has
Fig_ 7.15 Motor speed control unit
inherent self-regulating speed properties.
In this circuit a thyristor is used to control the
7.10 Exercise: Motor Speed Control Circuit (Fig. 7.15)
operating speed of the mot~)f by switching pulses of
11te 1lIotor that is commonly lIsed in appliances such current on positive half cycles of the mains supply.
as electric drills, sanders, food mixers, etc. is the series For high speed the thyristor will be triggered on very
wound "universal" type electric motor. This consists early in each positive half cycle, and by increasing the
Of:l field winding and an armature wired in series phase shift of the gate trigger signal the thyristor will
between the motor tenninals. When a voltage {a.c. or be triggered later and this reduces the motor's speed.
d.c.} is applied across the terminals a current flows Each positive half cycle of the supply causes a
through the field winding and the armature. Oppos current to flow through the potential divider network
ing magnetic fields arc set up between the field and R 1 , RV band D\ . All attenuated positive half cycle
the armature, and this rorces the armature to rotate. of the supply therefore appears on the wiper of RV t
As the am1ature rotates it generates a voltage in and this positi.ve voltage charges C 1 via D2 _In fact
C1 stores a charge that is proportional to the difference Questions

between RV 1 wiper and the voltage on the cathodc
(I) Explain briefly how die transistor trigger circuit
of the thyristor. TIIC latter voltage is of course the
ofTrl and Trl works. (Hint - study the equivalent
speed-dependent back e.m.f. of the motor. When the
circuit of the thyristor.)
voltage across C 1 exceeds about 3 V, Trl and Trl.
(2) TIle conlrol system fails so that the motor operates
which are wired as a regenerative switch, are triggered
at maximum speed and RV I has no contwl. State,
into conduction and supply a pulse of current to the
with a supporting rcason, tlte component failure.
thyristor gate. TIle thyristor conducts to supply
power to the motor. On the negative half cycle of the (3) TIle control system fails so that the motor fails to
mains the thyristor naturally turns off. One of Ole run at all. list, with a supporting reason, the possible
; features of the universal-type motor is that by operat component failures that could cause this hu1t? How
; iog from a half wave rectified supply only about would you test the circuit to locate the fault to one
20% of the power from an equivalent fun wave component?
system is-1ost. This fact enables quite efficient and (4) State the symptoms for the follOwing c:omponent
relatively cheap half wave control systems to be used. failures:
TIle circuit achieves low speed operation when the (a) Dl open circuit.
wiper of RV I is moved towards the anode of Df - 11us (b) Thyristor gate to cathode short circuit.
means 01at a lower fraction of the positive half cycle (S) A modification is included to reduce the effect
is applied across C 1 , and since the other end of C1 is of "skip-cycling" and judder for a particular motor by
connected to the motor's back-e.m.f. the transistor connecting a 2 IlF capacitor across RV l ' Explain
trigger switches at a later stage in the cycle. TIus briefly how tlus modification affects the operation
reduces the power switched to the motor by the of the circuits.
tltyristor and therefore reduces the motor's speed.
At very low speed, "skip-cycling" takes place,
when the thyristor is used to deliver power in a por·
tion of one half cycle out of, say, five. This occurs
because tlte gate signal cannot, in this circuit, be
later than 90" in each positive halfcycle. So when the
thyristor fires, it supplies at least a quarter· cycle of
power to the motor. 111is causes the motor to accele
rate and to increase the back-e.m.f. Consequently the
voltage across Cion the next half cycle will be too
;mall to trigger Tr I and Trl, and tile thyristor will
:emain off. TIle thyristor will only conduct again
Nhen the speed and back-e.m.f. have fallen to the
Jriginal value and this may take several cycles. At
lrst sight this would seem to '''C a disadvantage, but
",hen the motor's speed is averaged out over a period
)f say a quarter of a second one can see that a fairly
:onstant low speed is maintained.
Regulation at high or low speeds is achieved since
ny sudden change ofload must produce a drop in
lOtor speed with a consequent fall in the back-e.m.f.
l1is means that on the next positive half cycle the
hyristor receives a gate pulse earlier in the cycle and
lis switches on the thyristor whieh delivers more
ower to the motor than previously.
95 LINEAR AND D!GITAL !NTEGRATED CIRCUITS
generators, etc., can e;lsiiy be designed. Some typical (0) Pre- amplifier
examples are shown in Fig. 8.3.
DIGITAL lCs, originally designed [or aero-space
and'computer applications because of the high pack
ing density and good reliability, are now being used
in proces!l control instrumentation, electronic
generating and measuring instruments, and in many I\; o - - - i l - - -.......~
other applications. ll1e logic «families" most com lnpul 1 p.F
monly used are: diode-transistor logic (DTL), tran

sistor-transistor logic (TTL), the high speed emitter
coupled logic (ECL), and the complementary metal
oxide field effect logic (C-MOS). Each of these
have their particular features and advantages.
DTL, one of the first types on the market, is (b) Squore wove oscillo!or
relatively slow having a propagation time of about R.
20 nanosecs. Ilowever there is m1.:lch equipment
designed with DTL stiD in use_
TTL, now the most populac logic type, has a speed
of about IOns and has a very extensive range of basic
gate functions and interface circuits. All devices are
JU1
designed for 5 V d.c. operation (as is DTL). A glance

at a component ca talogue will show just how extensive
the range of TTL logic is.
ECL with its operating time of 2 ns is used more
in computer circuiting.' Field effect logic (C-MOS and
LOCMOS) have advantages in packing density, and Fig. 8.3 TyPical examples of circuits using op-amps
In (b),[= 1/{4CtRt + R,(R, + R~)) .
it is these devices that are used for large-scale integra
tion such as a 4096-bit dynamic read/write random
access memory (signetics 2660) housed only in a
16-pin package.
8.3 Servicing Instruments Containing ICs

Since the 1C itself constitutes a functional block, with
a minimum of extemal components being used, it is
obvious that a failure of one part inside the IC wilt for example, has a maximum supply voltage rating
lead to a complete loss of performance and the IC will of 7 V; no fault must be allowed to occur in the
then have to be replaced. One intemal component
power supply which would put an over~vo1tage on
failure renders the whole Ie useless. Naturally ICs
to the ICs. This would cause many of them hi over
.are designed to give very high reliability but failures
heat and bum out.
wilJ and do occur. Some of these failures \-\-ill be When testing equipment containing ICs take care
caused by the natural environmental stresses acting not to short pins by using large test probes, avoid
Oil the lC, weakening perhaps an internal connecting the use of excessive heat when unsoldering a com
lead and finally causing an open circuit. Alternatively ponent, and lleJ,'er remove or plug in an IC to its
"spikes" on the supply fails, or large current surges socket without first switching off the power supply.
at switch-on, can cause failure at semiconductor TIlis is when excessive surge currents can occur, and
jUlictions. It is the designer's job to ensure that the it is possible for a complete batch of les to be des·
power supply is well regulated and filtered. TTLlogic, troyed one after another by anunski11ed operator.
r-------------------~~------------------~----~--~r_----i_O +12V
'---?-":'::';':;--Y~-l
@;j
,~,
: TH\ GM473
~boe® I Tr~;
t----"':::...( I TIP31A
aelos I on heal sink
I ,
T~ ,
I
I
RIo' Enclosure :
680fi :
Il.. __________ ...!I
Ov
"ig. 8.4 Heater control unit using a 741 op-amp

97 LINEAR ANO DIGITAL INTEGRATED CIRCUITS
For fault finding on IC units follow a logical As the temperature in the enclosure rises the.
procedure: resistance of the thermistor falls so that at some point
(a) Check power supply at the Ie pins. Is it within dependent upon the setting of RVl, the output of the
its rated value? Is the ripple level low? If the 741 falls.TIlis cuts offTrl and Tr2 to stabilize the
answer is yes, proceed. temperature inside the enclosure.
(b) Make slIre that the required input is present
at the IC pin indicated on the diagram. Questions
(c) Check for a suitahle output
(d) Check Visually and with a meter for any open (I) State the voltages you would expect to measure
or short circuit$ in the copper track to the Ie. with a standard multimeter at all test points when
Several aids for servi.cing ICs are available, such as the unit has just been switched on. 11\e temperature
IC inserters, test clips (reduce risk of accidental in the enclosure is abounO<>C. Assume RV lis at mid
shorting), logic probes, etc. Use t1lem wherever pOSition.
possible. . (2) 11\e unit fails so tha~ it will not heat. TIle voltages
If an IC has to be removed by unsoldering, always measured were as follows. State, with supporting
use a desoidering tool to remove the solder from each reasons, the possible component failures that could
IC pin in turn until the IC can be lifted out. It's worth cause this fault.
taking time·over this job to avoid damaging the copper
track of an expensive printed circuit board. TP
MR· 2·1
/1\21314j516
8·1 0·26 O' 0 0
8.4 Exercise: Heater Control Unit using a 741 Op·Amp (3) TIle unit fails so that the temperature rises to a
(Fig. 8.4) high value and cannot be controlled. After studying
the following voltages, state in each case, with a
In many circuits IC op-amps are used together with supporting-reason, the possible component failures
other components to give a particular function. The 11\at could cause each fault.
741, because of its high differential gain and excellent
common mode rejection, is ideal for amplifying the 1 2 3 4 5 6
difference signal from a bridge circuit. In this example Fault
a d.c. bridge is used with a thennistor (GM473) as the A 11·] 6·2 12 1·9 1·25 0·53
sensing element. TIle signal from the bridge is fed to B ll·1 6·2 2·3 1·95 1·35 0·56
the inputs of the 741 which is wired as an illverting C 2·1 0 7·6 1·9 1·25 0·53
amplifi~r. Voltage gain is deterri1ined by the ratio of (4) What would be the full symptoms for the
Rs and R.;, in other words a gain of approximately following faults?
30. ·While the temperature inside the enclosure is low, (a) DZ 1 open circuit.
the thermistor has a fairly high resistance (47 kQ at (b) Tr2 base emitter short circuit.
25"C) and the output of the 741 will be high. This (c) 741 open circuit inverting input.
level forward biases TTl via tJle zener diode and R 7 • (d) Rl open circuit.
Tr2 also conducts and takes a current of approxi
mately 0·8 A. The power dissipation orTr2 is then
nearly lOW, and this heats up the small enclosure. A
current limiting circuit is provided by Tr3 since when
the output current increases beyond 0·8 A. The
voltage across RIO rises to about 600 m V and this
causes Tr3· to conduct, thus diverting base current
from Trl.
•... ,~- .......... ,---- :'~'(;"~'f" - .....~ .........- - -........,.......- -......
STC 4208
C, 1MHz x!ol
PI b <0
co
I 01 • • . m
100 pF I MHz r
Olltput
m
()
P, d P, c -l
::rJ
0
Z
100Hz ()
R, Olilput i1
l>
lk2 C
r
-l
0
l>
G)
C3
O<I,u.F
• ,U' .. I U .. I ' ,',' u • til I.' ..C4 E
0 0 V
z
0
(/)
(/)
O<',u.F,
+5V
Gote input
logie' 10 kHz
output
Fig. 8.5 Frequency standard circuit (100 Hz)
PI is an SN 7400N quad 2 input NAND gate
99 LINEAR AND DIGITAL iNTEGRATED CIRCUITS
8.5 Exercise: Frequency Standard Circuit using TTL Questions
Logic (Fig. 8.5)
(1) The unit fails so that tlle 1 MHz output remains
nlis circuit shows the use of standard TIL logic gates at +4·3 V. TIle 100 Hz output is present. State the
to produce ,an accurate stable frequency of 100Hz. portion of the circuit that has failed.
A quad-two input positive NAND gate (SN7400N) (2) What is the purpose of C3 and C4 ? Where should
is used together with a I MHz crystal to provide a they be located?
1 MHz square wave oscillation'Pld and PIC are (3) The unit has a fault such that there is no 100 Hz
operated as linear amplifiers because of the feedback or ],0 kHz Qutput, although the 1 MHz output is
formed by the resist<)IS. R2 , RI and ~, RJ . Since correct. Which ICs could be at fault, and how could
positive feedback is provided by the crystal, the you quickly locate the fault?
circuit oscillates at 1 MHz. PIb is used to buffer the (4) How could the accuracy and stability of the.
1 MHz output signal from the crystal. In the same 100 Hz output be checked in the laboratory?
way PIa will transmit tIle I MHz signal to the decade (5) What would be the result of the follOwing faults?
count.er P2 when the "gate" input is at logic L (a) P4 open circuit track to +5 V line.
Each of the decade counters (SN1490N) divide (b) C1 open circuit.
the input by 10. so that four ICs are required to (c) PIa open circuit output.
divide tlle J ,MHz down to 100 Hz. Since the counters (6) How could the operation of each decade be
operate only on logic level changes the 100Hz will checked, assuming that the oscillator has failed?
have the same accuracy and stability at the 1 MHz
oscillator.
The following definitions apply to' logic measure
ments in TTL logic:
Logical 0 voltage: less than +400 MV (typically
+200 MV)
Logical I voltage: greater than +24 V (typically
+3·3 V)
N.B. 111ese voltages are measured with respect to
the 0 V rail.
As stated previously ,before checking the operation
of a particular IC, always.verify the presence of tlle
+5 V and 0 V supplies. Make the measurement directly ,
at the appropriate IC pins, not between board con

nections or on the printed circuit wiring.
Never check the voltage levels on an IC by con
necting the ~e meter lead to chassis and monitoring
the voltage ,vith the +ve lead, since a break in the
ground line to the IC will not be indicated.
Most logic circuits, especially those of the series
type as in' this example, are relatively easy to fault
find since the symptoms indicate the fault. for
example suppose the unit fails to give a 100 Hz output
After checking the presence of the +5 V line, one
should then check the output from the I MHz
oscillator, then the input and output ofICs P2 to PJ
until the fault is located.
8.6 Exerdse: Boiler Control Unit using TTL logic time a logic 0 frolll P Ib Sqts the ignition bistable so
(Fig. 8.6) that a logic I switches on the ignition. This should
quickly light the pilot, to give a logic 1 fron: the flame
In this circuit several different TIL ICs ,ITe used to
sensor. When this happens the ignition bist;:'ble is reset
control the ignition and operation.of a gas boiler.
and the pilot valve remains open although C1C Q output
This is only intended as an illustrative example and it
of the monostable reverts to Ipgic I.
is not a circuit in current practice. The intentioll is to Anhe start of the sequence the reset is removed
sho\v how logic ICs can be used to provide several from the two 4-bit counters so these change state with
functions. A more elegant circuit could 110 doubt be pulses from the 4 Hz oscillator. TIlis oscill1lor is a
designed. The assumed operating sequence for the standard type used with logic Schmitt triggers and it
boiler is as follows: produces a square wave output. Since there are 8
!.~ •
(l) Press start button to open pilot valve and to stages in the ripple through counter fonned by P6
switch on ignition.. and P, there is a delay of approximately 16 sec
(2) Flame sensor detects the pilot flame. If the before a logic I appears at th"e final outpl.l:. 111is logic
flame is not lit within 5 sec the pilot valve level is g<!ted with a logic 1 from the flame sensor to
closes. Otherwise the pilot remains open and open the main valve. When this occurs a kgic 0 is
ignition is then switched off.. connected to the 4 Hz oscillator to inhibit further
(3) After a delay of 16 sec the main gas valve is operation. 11le main valve therefore remains open.
opened if both the themlostat indicates low When the thermostat operates at the required
temperature and the flame is 011. temperature it generates a logic 1, and thi~ resets the
(4) When the required temperature is reached the counters to close the main valve. TIle counter reset
main valve is closed. Pilot remains on. is held until the thermostat opens. TIl en the counter
(5) When the temperature falls so that the counts up, so that after a further 16 sec delay the
thelmostat switches to logic 0, following a main valve opens again. Tests on the unit would
delay of 16 sec, the main valve is opened again. naturally be made with th~ Plain gas supply off.
(6) In the event of a pilot llame failure both the
main valve and the pilot valve will be closed.
Questions
In the logic circuit the following gates are used to
Ichieve the required operating sequeilce: (I) Explain the operation of the 4 Hz osciJator.
(2) A fault exists such that the pilot is lit, and remains
SN7400N Quad Two input NAND gate ............ .3
lit, but a logic 1 does not appear after 16 sec to operate

SN74132N Quad Schmitt triggeL ..................... 1
the main valve. Which portions of the circuit are

SN7492N A synchronous 4·bit counter ............. 2
faulty,,? How would you check circuit operation?

SN74121 N Monostable ..................................... 1
(3) TIle unit rails so that it does nollight the pilot.

111e pilot valve, ignition, and 1l1ain v;llve are Describe brieny the sequence of checks y(>ll would
perated by a logic 1. The flame sensor gives a logic make to locate the fault.
when the pilot flame is lit. TIle thermostat gives a (4) If the flame sensor fails, so that it constantly gives
19iC I signal when the required temperature is a logic 0 input, what effect would this have 011 the
~ached. circuit opera Ii on?
The circuit operates in the following way. Two (5) The unit developes a fault so that the main valve
lAND gates from PI are wired as a bistable and this tends to open at the same time as the pilot valve when
in be forced to change state by putting a logic 0 the ON button is pressed. nle 4 Hz oscillator output
lomentarily on one input. When the ON button is is correct and the thermostat is open. Stat,! the
res:;ed the output of Pla goes to a logic 1 and this possible component failures and show how the fault
iggers the monostable P3 • TIle Qoutput from the could be located to one Ie.
tonostable goes to logic 0 for 5 sec, the time (6) How could the cirellit be modified to r.ivc an
~ing determined by the RC network. A logic 0 on increased or decreased delay time for the m.lin valve?
la puts a logic 1 to open the pilot valve. At the same Only two logic N AND gates are available.
1 LINEAR AND DIGITAL INTEGRATED CIRCUITS
--1_--.'- + 5 V
Pilol vOlve
R
C
Ignition
Res~t by loqic 0
•
1
I
P6 SN7493 ....
;l. _____________
.,1
RS
390n
Res~j by logiC 0
Inhibit oscillator
4Hz
Oscillator
from pin 1 P3
Fig. 8.6 Gas boiler COil trQller using TTL logic

Each package is supplied with +5 V and decoupled with a
') 'What would be the effect of the following com· 0·1 p.F capacitor
ment failures?
(a) R3 high in value.
(11) C4 short circuit.
(c) p 4C output short circuit 10 +5 V rail.
)6 ELECTRONIC FAULT DIAGNOSIS
XERCISE 8.6 (page 10]) (5) P4C failed with output in\ow state.
:) Check output from 4 HZ oscillator. Then 1\, P7 (6) Remove Pm lead from P4C input, then gate the
and finally P4 b. P4(: and P4D' output of Pm with P7G using one NAND gate,
.) Unit will fail to light to pilot. invert this. and connect to input of P4{' .
..
Inclex
Alarm circuit 86 intermodulation 42

Amplifiers 34 phase shift 42
classification 36 Driftin d.c. amplifiers 48
distortion 41
faults 43 Efficiency of power amplifiers 40
frequency response 36 Environmental stresses 3
measurements 39
Anti-saturation circuit 70 Fault location methods 13
Astable multivibrator 65 Feedback
negative 37
Bandwidth 36 positive 52
Bipolar transistor 4. ~. Field effect transistors 6
Blocking oscillator 60' Forward breakover voltage 81
Bridge rectifier 23 Frequency distortion
in amplifiers 42
Capacitor in oscillators 55
faults 3 Frequency response in amplifiers 36
measurements 10 measurement of 39
Cathode ray oscill9scope 8 Frequency of oscillators
Oamping circuit 68 measurement 54
dillping circuit 68 stability 54
Gosed loop gain 37 Full wave rectifier 23
Common emitter amplifier 14
Common mode rejection ratio 49 Gain 34
Common mode signals 48 Gain measurement 39
Complementary output stage 50
Cross-over distortion 42 Half power points 36
Crowbar overvoltage protection 25 Half-wave rectifier 23
Crystal-controlled oscillator 55 Half-split method 13
Cmrent gains of transistor 5 Harmonic distortion 55
Current limiting circuit 25 Holding current 81
Decibel 34 IGFET 6
Diac 7 Input impedance measurement 39
Differential amplifier 48 Integrated circuit 93
Differential gain 48 Integrator 66
Differentiator circuit 66 Intrinsic stand-off ratio of UJT 57
DigitallCs 95 Inversion of signals 35
Digital multimeters 8
Diodes 3 Lamp dimmer 87
catching 73 Light operated circuit 86
clamping 69 Linear regulator 22
clipping 68
Distortion 41 Mark-to-space ratio 55
amplitude 4.1 Meters 7
crOSS-()Ver 42 Monolithic Ie:; 93
frequency 42 Monostablc 72
harmonic 55 MOSFET 6
107
INDEX
-tor speed control circuit 91

Regulation 24, 26
Reservoir capacitor 23
gative feedback 37
Ripple 23
§:ative resistance oscillators 57
ise factor in amplifiers 47

Schmitt trigger 7 I
Stability
en loop gain 37
of operating point 14
~rational amplifier 96
of oscillators 54
;illators 52
of power supplies 24
Switching mode power unit 24,32

.se distortioll 42
se shift in amplifiers 35
Thyristor 6, 80
.se shift oscillator 53

testing 85
Hive feedback 52
Transistor measurements 11
ver gain 34
Transistor switch 74
,er output measurement 40

Triac 6,83
ler supply faults 28

Tunnel diode 57
neasurements 2!)
·amplifiers 46
Unijunction transistor 57
tection circuits for power supplies 2~
Unipolar transistors (FET) 6
.e waveforms 55
;e measurements 56
Waveform shaping 68
tifiers 22
Wien bridge oscillator 59
Answers to. Exercises
Chapter 2 EXERCISE 4 (page 2])

:-EXERCISE 1 (page 20) (a) C 1 open circuit. TIle RF should be fillered out.
:Estimation of meter indications in circuits given in (b) C2 open circuit.
Fig. 2:1 4. You answer should be withirr ± I 0% of that (c) C l open circuit. RIC I should form an integrator
given below. of I 1118 time constant.
(d) Diode open circuit. The diode should clip the
Fig. 2.14 A TP I 2
positive spike.
MR +3 +2·3
(e) ~ open circuit. Signal present at colkclor shows

Fig.2.14B TP 1 2 that the colpitts oscillator is working (:onectly.
MR -3 -7·8 1-;3 Chapter 3
Fig.2.14C TP
MR
1
-0·3
2
-1·2
A base current of25 IlA will flow through Rl
I EXERCISE 3.7 (page 29)
Fault
A DZ open circuit. Voltage at TP3 has risen. No
current flowing through R z .
causing the base voltage to be approximately -0·7 V.
B C 1 open circuit. D.C:level at TPI bwer
Connecting a meter to the base win lower this voltage.
Fig.2.14D TP
MR
I 1
+18
C R3 open or high or Tr 1 base emitter open
circuit
D Rl open circuit not C2 short, because TP1 is
high
EXERCISE 2 (page 20)
E DZ short circuit
Nonna! voltages TP '1 I 2 3
. . MR 1·5 0·8
F Tr2 base emitter open circuit
4·5
G R4 open or high in value.
Fault H Trl collector emitter short circuit.
A Base emitter short circuit
B R.,q open circuit EXERCISE 3.8 (page 31)
C Collector open circuit A Rl open circuit or DZ short circuit
D Collector emitter short B ~ or RV2 open circu it. Possibl Y C2 sho rt
E C3 open circuit giving negative feedback C Tr3 base emitter open circuit (or Tr4)
F Rl open circuit. D R.,q open circuit
E RV J. or Trs base open circuit or Rs open circuit
EXERCISE 3 (page 20)
Normal voltages TP
MR
I I
2·3
2
+3 +7
F Tr2 base e1l1itter open circuit
G Tr4 short collector to base
H Transformer secondary or primary open circuit
Fault Transformer primary short circuit '.0 screen.
A R2 open or C 3 short circuit
B Base emitter open circuit EXERCISE 3.9 (page 33)
C Collector emitter short (1) A RI7 open circuit B Ll open circuit
D ~ open circuit C Resistance checks are normal. Fault could be open
E Collector base short circuit component in 1'£7, Trg, Tr;l circuit, i.c. R13
F R3 open circuit. open circuit.
102
103 ANSWERS TO EXERCISES
(3) (n) C 1 sIc. Fuse blown. TPI resistance to 0 V, E Trl base emitter open circuit.
zero U: EXERCISE 4.10 (page 51)
(b) No output, although oscillator will start. No (1) Fault
switching signal at Trll collector. A Rs open circuit
(c) Tr2 ON and Trl OFF. Therefore no forward B R4 open circuit
bias for Trs. Trll will be non-conducting and the C R3 open circuit or Trl base emitter open.
output Z.(~ro. D D, short circuit
(d) Tr9 cannot conduct, so output remains zero. E Tr3 base emitter short circuit
However the oscillator will run when the start F R2 open circuit
switch is pressed. (2) (a) Bias voltages almost normal but output signal
(e) Circuit will operate and output will stabilize at would be grossly distorted on +ve half cycles.
20 V only when the start switch is pressed. (b) Bias voltages normal, zero output signal.
(4) Resistance readings normal. Fault is primary Chapter 5

short to screen EXERCISE 5.9 (page 60)
(1) C 1 or C6 open circuit
Chapter4 .
(2) R3 , RV1B, or Cs open circuit. R3 Qr RVtB open
EXERCISE 4.7 (page 46) would prevent positive feedback necessary for
Fault oscillations. Cs open would introduce m6re
A Cll short circuit. TIlis causes Tr2 to conduct negative feedback.
hea,,;ly and cuts off. Trl (3) (a) High gain giving grossly distorted output.
B Rs open circuit. Causes very high gain and (b) No output on switch position]_
change in d.c. bias levels (4) Rl opendrcuit
C Tr 1 collector ,emitter short
(5) C2 short circuit
D C2 short circuit ~.
(6)
E Tr2 collector open circuit
TP 3 4 5
.F C3 open Circuit. Introduces negative feed
MR +0·8 o +0·1.
back
(7) R7 open circuit.
G Rl open circuit.
EXERCISE 5.10 (page 61)

(1) Rl open circuit or C1 short circuit.
Fault
(2) C2 open circuit.
A R,q open circuit. Note that when the meter is
(3) C 1 open circuit or secondary winding short
connected in circuit it presents a current path circllit. ..
for TP3, 4 and 5. (4)
8 C] short circuit. TP3 and 4 are at the same
TP 3 4
voltage.
MR 12 12
C C 1 open circuit, which introduces negative
feedback to reduce gain. (5) The time constant of R2 Cz will increase. This will
D Drain of FETopen circuit. Tr2 cannot conduct. not change frequency, but will reduce sawtooth
E Trl base emitter open circUit. amplitude.
(6) Transformer secondary open cimuit.
Fault EXERCISE 5.11 (page 63)
A Zener diode DZ 1 open circuit (l) Dl short circuit. This increases current through
R RV 1 wiper or Tr3 collector open circuit Tr3 and therefore almost doubles the frequency.
C R7 open circuit. No negative feedback (2) Emitter follower action would cease. linearity of
D Zener DZ 1 short circuit output would be poor.
I tit< t::LECTRON1C FAULT DIAGNOSIS
(3) A R, open circuit EXERCISE 6.9 (page 75)

B DI or R9 open circuit (I) R~" Ct.
C Tr2 base emitter open circuit (2) Speed·up capacitor.
o R:2 open circuit
(3)
E Tr3 collector open circuit.
(4) ~ open circuit. As soon as voltage across C1

TP
MR
I justI +vc I 2
+7
3
+0·7
4
+0·1
5
+6·2
rises more than 0-8 V, Trl and Tr2 switch
(4) Momcnt;lrily short base to emitter. Collector
voltage should rise to +7 V.
. EXERCISE 5.12 (page 64)
(5) A Rl open circuit
~.(1) Modify Trl and Trz circuit into a low speed a
B C1 open circuit (possibly C2 or 0 3 open)
~~. stable multivibrator.
C ~ open circuit or Tr2 ble short
. (2) A.C. feedback to ensure rapid switch off at tlte 0 OZI short circuit
end of the ramp time. E Tr3 base emitter short 01 Rs open.
(3) A O 2 open circuit no feedback to switch bistable. . (6) (a) Bias voltages normal. No output pulse since
111erefore output remains high.
input caunot reach Trl base.
B Rs open circuit
(b) Same conditions as above.
C ~ short circuit
(c) TP4 rises to +7v. This forward biases Tr2'
o Trz base collector short
E 0 1 open circuit. No trigger pulse reading Trl' TP 4 5
F Tr:3 base emitter short 0·7 7 6·2
MR
G Tr:3 base collector short
H Rio open. (7) Dz open circuit.
(8) Take signal from Tr2 collector via a capacitor.
(1) Output A normal, no pulses from B.
(2) C3 open circuit.
(3) Trl would no longer act as a transistor. Tr2 col· (1)
11' 2 3 4 5
lector, voltage would faIl. So output A would be at
zero voits, and output B would remain at +8 V. MR 1·5 0·95 0·2 0·9 12
(4) Tr2 bise emitter short circuit or R:3 open circuit. (2)
(5) A Tr:3 emitter open circuit or Rs open or C3 TP: 2 3 4 5
short circuit
MR 0 8·3 1·8 1·1 6·7
B Tr4 emitter to B t short
C Tr4 emitter open circuit

Output will not be perfect square wave. Mark-to·
D Tr 5 base emitter short circuit
space ratio will he low.
E R6 open circuit.
(3) A Tr I base collector open
B Trl base emitter open or C, open
01apter6 C It; open
EXERCISE 6.8 (page 74) D Tr2 base emitter short
(1) D t short circuit. Measurement with ohmmeter. E Tr I base colI ector short
(2) D2 open or Rs open. Measure d.c. level at RtRs F. Rs open
junction.
(4) Check presence of signal at junction R2 , D t and EXERCISE 6.11 (page 77)
D2 • (1) Temporarily connect a 10 kS"l resistor from +5 V
(5) Db C l or Rl open circuit. TIle diode would have to transistor base. Collector voltage she·uld fall
the higher failure rate. to nearly zero.
(6) Dl short circuit. (2) Very low noise margin, particularly wi1h switch
(7) ~ open circuit. in '0' position
105 ANSWERS TO EXERCISES
\''J)
, A L1 , R J or R2 opetrcircuiL Possibly C 2 short. TIlen t ~ 0·55 CR.
B It,. open With RV I min. t ~ 0·55 sec.
C Trl Base emitter short or DJ short With RV1 max. t ~ II sec.
D Trl base emitter open. (3) Timing circuiHormed by Tr3 and lIJT:;.
(4) 0 V -0·6 V +5 V RV3 open, C:; short, Tr3 or UJT 3 open.
(5) 30 V +0·75 V +0·75 V Switch '1' (4) Use diodes to sense state ofSCR 2 and SCR 3 anode
o V -fO·75 V +0·75 V Switch '0' potentials and to gate signaL to SCR 1 •
(5) Cs short.
(1) Electro~ic voltmeter with very high input
(6) Dl opeq.
impedance. .
(7) SCR2 anode to cathode short or UJT2 fault such
as Bl to B2 short. Measure d.c. voltage at SCR2
.(2) R,l open. C 1 short.
anode. If zero, SCR2 at fault.
(3) Tr4 clamped to +10 V. OutP1,lt always 0 V.
(4) X DZ 1 open
y Tr2 collector base short
(1) Approx 12 Hz max, 1·5 Hz min.
Z Tr I gate to source short
(2) C:; short.
(5) ~ open. (5) A <4 short, RV 1 or ~ open
B Dl or C 1 open·
Cluipter 7'
C Ri open
EXERCISE 7.6 (page 86) . D un B2 to Bl short
(1) Lal~P TP I ~.2 3 4 5 E C2 or D2 open circuit.
ON 3 :3·6 3·7' 0 12 F R7 open
OFF 5·8 4·4 3·7 0·7 0·9 (6) Replace start switch with high value resistor
(68 kfl) bridged by a O·II-1F capacitor.
(2) Alarm operating continuously. Can1lot be reset.
(7) Fit low value resistors in series with lamps.
(3) A SCR I ' open circuit gate
B Rl or RV 1 open circuit
C .SCR J anode to cathode short
(2) SCR1 short anode to cathode.
D Rs open or high
(3) SCR 1 open circuit anode or gate. 'Possibly Rb
E Tr 1 collector base shmt
RV h D 2 , etc. open. Apply gate signal direct to
F Tr2 collector base short
SCR to localize fault.
(4) R<;open or failure of the reset switch. Bridge R<;
Chapter 8
v.1tb lk resistor.
EXERCISE 7.7 (page 87) (1)

(1) RF filter. TP
(2) 1-4kW MR
(3) Lamp almost unlit. No control.

• (2) ~ or 741 output open circuit.
(4) Triac shQrt MTl to MT2.
(3) A 741 output short to +12 V
(5) R2 or R3 open} Bridge suspect component with

(6) C 2 open another of same value.
B Trl short base to collector
C R2 or RV 1 open circuit
(7) Lamp unlit. With gate open, a larger a.c. voltage

will be indicated at TP3 than when gate to EXERCISE 8.5 (page 99)
cathode is short. (1) PJb'
(8) LI open. (2) DecQupling capacitors. Located near IC
(3) Pta. or P 2 , P 3 .
EXERCISE 7.8 (page 89) (4) Measured against a very stable fH~quency standard
(1) Electronic voltmeter or dlart recorder. fm accuracy. For drift use a frequency caunter of
(2) Vse formula Vc V(l - e -t~CR) high resoLution.

Electronic Fault Diagnosis

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Electronic Fault Diagnosis

Enviado por

Direitos autorais:

Formatos disponíveis

Some other ELBS low-priced editions

Benson and ELECfRIC-CIRCUIT THEORY e(h'ard Amn/d

Gosling A FIRST COURSE IN APPLIED ELECTRONICS Macmillan

THE ENGLISH LANGUAGE BOOK SOCIETY

First published in Great Britain 1977

ELBS edition first published 1981

All rights reserved. No part of this publication may be reproduced,

Text set in 10112 pt IBM Press Roman,

ISBN 0 273 01779 9

1 Basics of Fault Diagnosis 2

J.2 Components and common faults 2

2.4 Transistor faults

5.12 Exercise: slow speed lamp gtn...'>Jalor ~,64

3.J.,v Basic principles of d.c. power supplies 22

3':2 The linear stabilized power unit 22

3.3 Switching mode power supplies 24

3.5 Testing power supply circuits 26

3.6 Fault finding techniques and typical

current limit circuit 30

3.9 Exercise: switching mode power supply 32

6.9 Exercise: monoslable circuit .. 75.;' .

4.1 Types and classes of amplifier 34

4.3 Testing amplifiers: basic measurements 39

4.4 Transient testing of amplifiers 40

4.7 Exercise: two·stage pre·amplifier 44

4.8 Exercise: pre·amplifier with FET input 46

4.9 Exercise: d.c. amplifier 48

ment fomls an integral part of some b~dustrjal manu­

Another cause of component fallure is higll

~. volbge pulses or "spikes", generated from switched

TABLE 1.1 and appearing on internal supply wires. These

Compoltent Common type offault

tions in semiconductor devices.

Resistors High in value or open circuit.

Inductors Open circuit. Shorted turns. Short texts should be consulted.

(including circuit coil to frame (iron cored

In order to pass current in the forward direction, Forward mA

and electrons are accelerated to such velocities that breakdown

they dislodge other fixed electrons from the crystal

(2) Voltage Regulator Diodes

TIle transistor is best considered as a device in which

reverse'biased. Electrons Dow across the emitter

TIle base current Ib is much smaller than both

If the transistor fails collector open circuit, there

(4) Unipolar Transistors - the FET Family v.... ~

sufficiently negative, say -3 V. these depletio~

or as a reclifier depending upon how its gate is used.

deflection system and a fluorescent screen. A high T= O---=.2'-x-'--:-10:--""3 - 5 kHz

velocity, finely focussed electron beam is produced

Fig. 1.13 eRO used to measure a sine wave signal (rom an

Fig. 1.14 Double beam oscilloscope used to measure two

time-related wave forms from an astable multivibrator. TIle

y 1 beam is triggered on its positive edge.

10 ELECTRONIC FAULT DIAGNOSIS

example is given in Fig. 1.14 where the signals from

(3) Simple Component Testing capacitor ~C. c.

First it is necessary to detemline how the internal MULTIRANGE METER

1.5 Fault Finding on Electronic Instruments and

111e previous sections dealt with componellt failures

EQUJPMENT. Usually a list of instruments and

any special instructions are in the maintenance

MULTmANGE METER manuaL

1I (3) The engineer then has to DEFINE THE FAULT

It is no good trying to locate a fault that is

ment fomls an integral part of some b~dustrjal manu

vaguely defined. The symptoms must he accu