Television

Principles
and
Practice
J. S. Zarach
Senior Lecturer.
North Staffordshire Polytechnic

Noel M. Morris
Principal Lecturer.
North Staffordshire Polytechnic

M
 © J. S. Zarach and Noel M. Morris 1979
Softcover reprint of the hardcover I st edition 1979
All rights reserved. No part of this pUblication may
be reproduced or transmitted, in any form or by any means,
without permission.

First published 1979 by

THE MACMILLAN PRESS LTD
London and Basingstoke
Associated companies in Delhi Dublin
Hong Kong Johannesburg Lagos Melbourne
New York Singapore and Tokyo

Typeset in 10/12 Times by

Reproduction Drawings Ltd, Sutton, Surrey

British Library Cataloguing in Publication Data

Zarach, J S
Television principles and practice.
1. Television-Repairing
I. Title II. Morris, Noel Malcolm
621.3888'7 TK6642
ISBN 978-0-333-19221-4 ISBN 978-1-349-16124-9 (eBook)
DOI 10.1007/978-1-349-16124-9

This book is sold subject to the standard conditions of the Net Book Agreement.

The paperback edition of this book is sold subject to the condition that is shall
not, by way of trade or otherwise, be lent, re-sold, hired out, or otherwise
circulated without the publisher's prior consent in any form of binding or
cover other than that in which it is published and without a similar condition
including this condition being imposed on the subsequent purchaser.
Contents

Preface xi

1. Principles of Colour and Colour Perception 1

1.1 Colour Perception of the Eye 1
1.2 Additive and Subtractive Mixing of Colours 3
1.3 Hue, Saturation and Brightness 5

2. Transmission of Monochrome and Colour Television Information 6

2.1 The TV Camera 6
2.2 Block Diagram of Monochrome Transmitter 8
2.3 Principles of Colour Transmission 10
2.4 Colour TV Camera Arrangements 11
2.5 TV Signal Bandwidth 12
2.6 Signal Sidebands 15
2.7 Channel Allocation 17
2.8 Modulation of Colour Information 17
2.9 The PAL System 22
2.10 Other Colour Encoding Systems 23

3. The Composite Video Signal 25

3.1 The Video Signal Waveform 25
3.2 The Luminance Waveform of a Colour Bar Signal 26
3.3 Chrominance Waveforms of the Colour Bar Signal 28
3.4 The Composite Video Waveform of the Colour Bar Signal 33
3.5 The Field Synchronising Pulse Sequence 34
3.6 Teletext Information 36
3.7 Intercarrier Sound Transmission 36

4. The Essential Features of a TV Receiver 37

4.1 Introduction 37
4.2 The Tuner 37
vi Contents

4.3 The I.F. Strip 39

4.4 The Vision Detector and Video (Luminance) Amplifiers 39
4.5 The Colour Decoder 40
4.6 Sound Stages 41
4.7 The Synchronising Pulse Separator 41
4.8 The Line Time Base 41
4.9 The Field Time Base 42
4.10 The Picture Tube 42
4.11 Power Supplies 43

5. Tuners and I.F. Amplifiers 45

5.1 High Frequency Performance of Electronic Amplifiers 45
5.2 The U.H.F. Tuner 46
5.3 The Varicap Diode Tuner 48
5.4 V .H.F. and Integrated V.H.F ./U.H.F. Tuners 50
5.5 Touch Tuners 53
5.6 Remote Control 55
5.7 Tuner Servicing 57
5.8 Frequency Response of I.F. Amplifiers 59
5.9 Practical I.F. Amplifier Circuits 60
5.10 Alignment of I.F. Amplifiers 66
5.11 Automatic Frequency Control (A.F.C.) 69

6. Vision Detectors and Video Amplifiers 73

6.1 Principles of Signal Demodulation 73
6.2 Diode Detector Circuits 74
6.3 Synchronous Video Demodulation 75
6.4 Requirements of Video Amplifiers 79
6.5 Amplification of the D.C. Component 79
6.6 Amplification of Middle and High Video Frequencies 83
6.7 Luminance Stages in Colour Receivers 85
6.8 R, G, B Drive Circuits 87
6.9 Colour Difference Drive 89
6.10 Brightness Control 91
6.11 Contrast Control 93
6.12 Flyback Blanking 95
6.13 Beam Limiting 96
6.14 Typical Arrangements of Video and Luminance Amplifier Circuits 99
6.15 Automatic Gain Control (A.G.C.)-Basic Principles 101
6.16 Mean Level A.G.C. 101
6.17 Peak Level A.G.C. 102
6.18 Gated A.G.C. 104
Contents vii

6.19 Forward and Reverse A.G.C. 105

6.20 Direct and Delayed A.G.C. 107
6.21 Fault-finding in Video Amplifiers and A.G.C. Circuits 108

7. Chrominance Signal Stages 110

7.1 Processing the Chroma Signal 110
7.2 Chrominance Take-off Arrangements 112
7.3 Chroma Amplifiers 112
7.4 PAL Delay line 114
7.5 Synchronous Demodulators 119
7.6 (G'-Y') and R, G, B Matrix 121
7.7 Regeneration of the Colour Sub carrier 125
7.8 The Ident and the Colour Killer Circuits 130
7.9 The PAL Switch 133
7.10 Integrated Circuits in Colour Decoders 136
7.11 Fault-finding in Colour Decoders 140

8. TV Sound 144
8.1 Principle of Intercarrier Sound 144
8.2 Intercarrier Amplifiers and Sound Demodulators 145
8.3 Audio Frequency Amplifiers 150

9. Synchronising Pulse Separator and Field Time Base Circuits 154

9.1 Separation of Synchronising Pulses from Composite Video 154
9.2 Processing of Field and line Synchronising Pulses 157
9.3 Receiver Time Base Requirements 161
9.4 S-correction and Scan Linearity 162
9.5 Valve Field Time Base Circuits 164
9.6 Transistorised Field Time Base Circuits 171
9.7 Integrated Circuits in the Field Time Base 176
9.8 Class D (Switched Mode) Field Time Base 178
9.9 Fault-finding in Sync. Separator and Field Time Base Circuits 178

10. Line Time Base Circuits 181

10.1 line Time Base Synchronisation 181
10.2 The Line Oscillator 185
10.3 Transistorised line Driver Circuits 190
10.4 Equivalent Circuit of a line Output Stage 191
10.5 Simple Transistorised line Output Stage 192
10.6 High Power Transistorised Line Output Stage 195
10.7 Valve Line Output Stage 198
viii Contents

10.8 Generation of E.H.T. 201

10.9 Thyristors in Line Output Stage 204
10.10 Faults in Line Time Base Circuits 206
10.11 Safety Aspects of Line Time Base Operation and Servicing 208

11. Picture Tubes and Associated Circuits 210

11.1 Basic Principles of Picture Tube Operation 210
11.2 The Electron Gun 211
11.3 Arrangements for Correct Picture Geometry 212
11.4 The Shadowmask Tube-Delta Gun Version 213
11.5 The Shadowmask Tube-In-line Version 216
11.6 The Trinitron Tube 217
11.7 Dynamic Convergence Circuits-General Requirements 218
11.8 Dynamic Field Convergence Circuits-Delta Gun Tubes 219
11.9 Dynamic Line Convergence Circuits-Delta Gun Tubes 221
11.10 Dynamic Convergence Circuits-In-line Tubes 223
11.11 Pincushion Correction 225
11.12 Grey Scale Tracking 231
11.13 Automatic Degaussing 232
11.14 Tube Safety Features; Precautions when Handling C.R.T. 233

12. Receiver Setting-up Procedure 235

12.1 Fundamentals of Receiver Setting-up Procedure 235
12.2 Installation Checks on a Colour Receiver 235
12.3 Manual Degaussing 237
12.4 Purity Adjustments 237
12.5 Static Convergence Adjustments 239
12.6 Dynamic Convergence Adjustments 240
12.7 Pincushion Distortion Adjustments 242
12.8 Grey Scale Adjustments 243
12.9 Test Cards 245

13. Television Aerials and Systems 249

13.1 Properties ofV.H.F. and U.H.F. Signals 249
13.2 Fundamentals of the Receiving Aerial; the Dipole 250
13.3 Multielement Aerials 254
13.4 Special Aerial Designs 255
13.5 Aerial Feeder Cables 257
13.6 Choice of Aerial Installation 258
13.7 Aerial Attenuators 260
13.8 TV Signal Distribution Systems 261
Contents ix

14. Receiver Power Supplies 263

14.1 Power Supply Requirements 263
14.2 Heater Supplies 264
14.3 Unregulated D.C. Supplies 266
14.4 Simple Stabilised Supply Circuits 268
14.5 Switched-mode Power Supplies 272
14.6 Thyristor Regulator Circuits 273
14.7 Series Transistor Switched-mode Regulator Circuits 277
14.8 Shunt Switched-mode Power Supplies 278
14.9 Overcurrent Protection 283
14.10 Overvoltage Protection 284
14.11 Power Supply Fault-fmding and Safety Considerations 286
14.12 Mains/Battery Operation 288

Index 289
Preface

The aim of this book is to present the reader with a comprehensive description of
the principles involved in the operation and servicing of modern TV receivers. The
treatment of the subject is such that is should appeal to students preparing for
examinations of the City and Guilds Institute for Radio, TV and Electronics
Mechanics. In this connection the syllabuses of the two options, Television (Colour
and Monochrome) and Additional TV, receive extensive coverage.
Students taking CGU Radio, TV and Electronics Technicians course, as well as
the various 'television options' within the Technician Education Council framework,
should also find the text of considerable assistance.
Many service engineers who wish to refresh their knowledge will find the subject
presented in an easy-to-follow manner.
The contents of the book reflect the circuit techniques adopted in modern TV
receivers; these include transistors, I.C.s, thyristors, valves (still used in many time
base circuits) and other devices as appropriate.
The first three chapters cover the important aspects of colour perception and the
formation of monochrome and colour signals (with special reference to the PAL
system). A review of a complete TV receiver in block diagram form is presented in
chapter 4; this provides the basis for detailed circuit descriptions in later chapters.
Chapters 5 to 7, inclusive, cover the main signal path from the tuner to the video
(luminance) amplifiers and decoder. Practical circuits and principles of the inter-
carrier sound reception are fully outlined in chapter 8. The field and line time base
circuits (including e.h.t. generation) are featured in chapters 9 and 10. This is
followed in chapter 11 by a description of picture tubes, including delta gun and
in-line tubes, together with information about convergence and pincushion correc-
tion circuits.
Receiver setting-up procedures are outlined in chapter 12, including details of
test cards F and G. Aerials and practical aspects of their installation are discussed
in chapter 13; finally, in chapter 14 receiver power supplies are described, ranging
from simple rectifier circuits to switched-mode arrangements.
Throughout the book, reference is made to the practical aspects of TV servicing
and to the accompanying safety considerations.
xii Preface

The authors wish to thank the following organisations for permission to use
their technical information in this book: BBC, Decca, GEC, ITT, Mullard, Philips,
Rank Radio International and Thorn.

[Link]
[Link]
1 Principles of Colour and
Colour Perception

1.1 Colour Perception of the Eye

In this chapter we want to concentrate on a subject that may appear at first glance
to have more connection with physics than with TV engineering. To appreciate how
the transmission and reception of colour TV signals has been made possible, we
must examine the whole process of colour perception, or-as one may say-the
illusion of colour.
As can be demonstrated, so-called 'white' light may be split into a large number
of individual colours which form the whole light spectrum. If the source of light is
sunlight, then the spectrum consists of all colours from dark violet to deep red. If
the 'white' light may be split into a number of colours, is the process reversible? Is
it possible to combine a number of colours, and produce 'white' light? This is indeed
easily achieved; it has been found that it is not even necessary to mix all the colours
of the rainbow to produce an impression of 'white' light. The human eye is cOIl-
structed to be principally sensitive to the overall light intensity-that is, its bright-
ness: only a small percentage of the receiving area of the eye is sensitive to the
actual colour of the light (this phenomenon will be recalled later in conjunction
with the composite TV signal).
It has been discovered that the impression of colour seems to be centred around
three actual colours: blue, green and red. The eye, as it were, analyses the image it
receives primarily in terms of these three hues and also in terms of the overall
intensity of the light, and this is precisely what a colour TV camera does. The
sensitivity of the eye to the three colours is not uniform-it is at its greatest to the
yellow-green, followed by red and then blue. Further experiments show that if we
combine red light, blue light and green light in appropriate intensities, all three
being emitted from separate sources, the eye will perceive white light.
Light is a form of electromagnetic radiation, rather like radio waves, but it
differs from the telecommunication signals in the wavelength or frequency spec-
trum involved. The actual colour one sees is only a sensation produced by the eye
when it is struck by radiation of a given frequency. While frequencies of TV
channels extend up to only 850 MHz, the light spectrum occupies a frequency
range from approximately 400 million MHz to 800 million MHz. In radio and TV
the wavelength is expressed in metres or centimetres; when we speak of light
1
2 Television Principles and Practice

frequencies, the wavelength is so short that the unit used is the nanometre (1 nm
= 10- 9 m; therefore there are 10 million nanometres in one centimetre). Figure 1.1
shows the distribution of the different forms of electromagnetic radiation, their
frequencies and wavelength.

radio/TV heat light X-ray frequency

(MHz)

X1014 MHz

700 600 500 400 nm

red green violet

Figure 1.1 The spectrum of electromagnetic radiation

red light

"ore ~

green light ..
source ~

R+G+B= white
blUelight~
source

red light
source

R+G=yeliow

Figure 1.2 Additive mixing of colours

Principles of Colour and Colour Perception 3

white surface -----:~~----

yellow reflected

yellow light 0
(R+G) ~

no light
reflected
blue surface -----'''''r'"T-----
I
I

• +
(R+G) absorbed

(a)

white light
(R+G+B)
0

.,
I
I
I
I

( R+G) absorbed
(b)

Figure 1.3 Subtractive mixing of colours using paints. (a) White paint reflects all
colours; the blue paint can reflect blue light only; the original yellow light reaches
the eye when reflected by the white surface, but light is not reflected off the blue
surface, which now looks black (no blue in the yellow beam). (b) Both surfaces,
when illuminated by white light, retain their intended colours (all three colours
are present in the source of white light).

1.2 Additive and Subtractive Mixing of Colours

The sensation of 'white' can be produced by the eye as a result of combining the
outputs from three separate light sources: red, green and blue. This implies that the
three colours are added and the process is called additive mixing. (See figure 1.2.)
Using mathematical shorthand the above statement can be repeated as follows
white = red + green + blue
It is obvious that only the correct intensities of the three colours must be used,
for too much of anyone of them will 'bias' our 'white' more towards that particular
4 Television Principles and Practice

colour (producing, for example, the effect of 'warm white' or 'cool white').
The three colours which together produce white are called the primary colours.
If only two primary colours are mixed at a time, the resultant hues are called the
complementary colours. Provided the proportions of each primary are otherwise
the same as those needed to produce white, the complementary colours are as
follows
green + red = yellow
green + blue = cyan
red + blue = magenta
As will be shown later, the colour TV system operates on the principle of additive
colour mixing. It is useful, therefore, to remember the above combinations because
of their important practical implications in TV servicing.
Already, out of the three primaries, it is possible to obtain six different colours
plus the white, and, of course, black on a TV screen simply means the absence of
any light output. All other colours can be produced by altering the proportions of
primary colours during the mixing process. For example, orange can be obtained by
increasing the amount of red in the green-red mixture.

yellow filter blue filter

white light
R+G+B
(R+G) out no Iight output
f>-
eye 'sees'
black

blue absorbed ( R+G) absorbed

yellow filter cyan filter

white light
R+G+B
(R+G) out

f>-
eye 'sees'
green

blue absorbed red absorbed

Figure 1.4 Subtractive mixing of colours using fIlters

Principles of Colour and Colour Perception 5

There is another method of colour mixing which is used in connection with

colour paints, colour photography, etc. This is known as subtractive mixing
because the colour of an object is the result of certain colours being absorbed, or
subtracted from the illuminating light, while other hues are then reflected towards
the observer's eye. A number of examples of subtractive colour mixing based on
the properties of different paints are given in figure 1.3. The diagrams in figure 1.4
illustrate this principle by means of colour mters. It must be stressed that the state-
ment 'white = red + green + blue' is still valid. In additive colour mixing, however,
the light reaches the eye directly from its source (lamps, TV screen), and in sub-
tractive mixing the light has been reflected by another object which can alter the
colour content of the original light source.

1.3 Hue, Saturation and Brightness

Later in the book we shall see how the principles of additive mixing are put into
practice in a colour TV system. At this point certain terms have to be introduced
which are used in connection with the TV signal.
Hue is the actual colour that the eye sees-Leo we can have red, magenta, green,
orange, etc., hues. Any of those colours may appear to be strong, rich-or, as it is
called, saturated; alternatively they may appear to be light, pale-or referred to as
desaturated. Desaturation of a colour is caused by the presence of white light in the
original hue.
Saturation is an indication of how free a colour is from any dilution with white
light. For example, in colour TV increasing the setting of the brightness control
adds 'white' light to the display, which results in very pale colours (see section 7.6).
That causes, say, a rich red colour to become pink-or a de saturated red. Of course,
desaturated colours can also be part of the televised scene; for this reason informa-
tion about colour saturation must also be transmitted in the TV signal.
Brightness is a measure of the intensity of light associated with a given colour
(including white). Brightness is linked with the response of the eye to different
hues; for example, yellow appears to be a brighter colour than green or red.
The term 'brightness' will be familiar to the TV engineer, as it is also used in
connection with monochrome television-a black and white receiver displays the
variations in brightness of the scene on its screen. The white or yellow parts of the
scene would appear almost identically bright ('white') on a monochrome TV screen,
while the blue or dark red content might look very dark grey.
In colour TV terms the brightness of the scene is described by its luminance,
which means that some parts of the scene appear to be brighter, or more luminous,
than others. This is a vital piece of information that must be transmitted for two
reasons: firstly, monochrome receivers can only process the luminance content of
the televised scene; secondly, the luminance signal is responsible for the reproduc-
tion of picture detail in colour receivers. It has already been mentioned that the
eye recognises detail principally as a variation in light intensity; this applies
especially to fine detail in the image.
2 Transmission of Monochrome
and Colour Television
Information

2.1 The TV Camera

The TV camera converts the image of the scene to be televised into an electrical
waveform called the video wavefonn. The camera assembly can be considered as
consisting of an optical and an electronic section.
The optical part of the camera follows the established photographic practice,
but the picture is focused on to a 'target' layer of material whose resistance varies
with the amount of light falling upon it.
The electronic section is formed by the camera tube and the associated ampli-
fiers. The tube produces an electron beam by thermionic emission from a heated
cathode. The electrons are accelerated by the anode, the electron beam being used
to scan the target (see figure 2.1) so as to 'read' its illumination and generate a vide
waveform.
The scanning movement of the beam is in two directions-horizontally across
the target and also in the downward direction; the image is being 'traced' as ifby
individual lines which are positioned below one another. Each time the beam
reaches the end of one line, it is returned quickly to the beginning of the next one.
The scanning is performed by the two sets of scan coils placed on the outside of
the tube (figure 2.1). There is also a focusing coil to ensure that the electron beam
is converged when it reaches the target.
The target part of the camera tube consists of a glass faceplate, on which a very
thin, transparent layer of conducting tin oxide (signal plate) is deposited and then
the target layer proper. The tin oxide allows the light to pass through it, but at
the same time it conducts electric current to the target. A metal ring surrounds
the signal plate to act as a seal for the tube and also as a connector to an external
supply. The output signal voltage is developed across resistor R.
The target layer and the signal plate behave like a capacitor which is charged by
the electron beam. When light falls upon the target, the resistance of the target
layer drops and the capacitor loses some of its charge at that particular spot. The
amount of charge lost depends upon the amount of light received. The design of
the target is such that the loss of charge occurs only between the target and the

6
Transmission of Monochrome and Colour Television Information 7

target
layer '---____________---'1 focus coil
grid
target
ring ----....
glass rllo-+-----------------,I------+---.
faceplate

tin oxide
j----------------------
L _______________________ ~

layer

heater

video accelerating
output anode

h.t.
supply

Figure 2.1 Simplified diagram of a camera tube

signal plate. The optical image consists of a large number of dark and light 'spots',
and, therefore, what we have is not one capacitor but thousands of them, the
number depending upon the detailed contents of the picture. As the beam scans
the target, it encounters all these little capacitors in a varying state of discharge
and it recharges them. As a result, we get a current flowing through the resistor R
whose value, at any moment, depends upon the brightness of the scene. The mag-
nitude of the current is only a few microamperes, and a suitable amplifier is needed
to produce a usable signal. The voltage drop across R gives rise to the video wave-
form.
The most commonly used tubes are called the Vidicon and the Plumbicon. Their
principle of operation is similar to the one described, but the target material is
different in each case. Vidicon cameras are very compact and are used in closed
circuit TV, industrial applications, portable outdoor colour TV cameras, etc.;
Plumbicon tubes are mostly employed in studio colour TV cameras.
In the receiver the process is reversed-ultimately, a scanning beam is produced
in the picture tube to 'trace' the same scene on the screen. It is necessary to keep
the beam in synchronism with the camera beam; this is done by means of synchron-
ising pulses which tell both the transmitter and receiver time bases when to start and
when to finish the scan. Because the required deflection is in two directions, we have
line (horizontal) scan and field (frame or vertical) scan. Thus the output from the
camera will have the sync. pulses added to it before it can be modulated upon the
carrier.
8 Television Principles and Practice

2.2 Block Diagram of Monochrome Transmitter

The diagram in figure 2.2 shows a monochrome transmitter arrangement in block
form. A monochrome system has been chosen here since it provides an insight to
the operation of the TV system without the complexity associated with colour
television which will be described later.
The signal from the camera, together with the synchronising pulses [see figure
2.3(a)] ,modulate the vision carrier in the modulated r.f. (radio frllquency) ampli-
fier. Two methods of modulation are possible: positive modulation and negative
modulation.

sound 'II aerial

carrier
oscillator

audio amp. r- modulated

r.t. amp. r- sound
r.f. amp.
mil2:

TV camera

~ video amp. ..... sync.

pulse adder fo-
modulated
r.f. amp. r- vision r.t.
power amp. r-
vestigial
sideband
tilter

-
camera
time base
circuits

+
vision
sync. pulse carrier
generator oscillator

Figure 2.2 Simplified block diagram of a monochrome transmitter

With positive modulation, the amplitude onhe carrier increases with the
brightness of the scene [figure 2 .3(b)] ; maximum brightness, known as peak
white, corresponds to 100 per cent amplitude of vision carrier. When negative
modulation is used, the carrier amplitude diminishes with increasing brightness;
peak white occurs at 20 per cent of the maximum amplitude [figure 2.3(c)].
Negative modulation is the most popular method, and is used in the UK for the
625-line standard. Positive modulation is used in a few countries and is also
Transmission of Monochrome and Colour Television Information 9

bright scene

dark scene

sync.
pulses vision carrier

(a) (b)

bright scene dark scene

/
----J_ 20% limit of video
modulation

(e)

Figure 2.3 Examples of vision carrier modulation: (a) two lines of video waveform
before modulation; (b) two lines of video waveform modulated on a carrier-positive
modulation; (c) two lines of video waveform-negative modulation

employed in the UK for the 405-line standard. The chief advantage of negative
over positive modulation is the improvement in picture quality. Any impulsIve
interference which normally affects the maximum amplitude of the signal will, in
positive modulated systems, correspond to the 'whiter than white' region; this
causes annoying white spots to appear on the screen. With negative modulation,
such interference produces black spots which are less obvious to the viewer-
unfortunately, this affects the synchronising pulses, and more elaborate circuitry
is needed in the receiver line time base to ensure steady synchronisation.
The sound channel uses frequency modulation on the 625 -line standard and
amplitude modulation for the 405-line transmission. Before they are radiated from
the aerial, both the vision modulated and the sound modulated carriers are ampli-
fied by their respective r J. power amplifiers to raise the power levels of the signal.
10 Television Principles and Practice

2.3 Principles of Colour Transmission

As we saw in the previous chapter, it was possible to mix the three primary
colours, red, green and blue, in their specified proportions to produce either white
or other colours as required. The process is reversible-we can split the white and
the various colours into their primary colour contents. Therefore, at the transmitter
the televised scene can be analysed by the cameras into its red, green and blue
content, while the receiver will mix the signal voltages to display the original
colours on the screen. One of the main requirements of a colour transmission system
is that it must be compatible with the existing black and white system. In other
words, a monochrome receiver must reproduce the signal in black and white with-
out any modification or loss of quality. Also, colour receivers must be able to
display a monochrome picture; this is sometimes called reverse compatibility.
To satisfy these requirements, the transmitted information consists of the
luminance signal and two so-called colour difference signals (chrominance or
chroma). Black and white sets respond to the luminance content, since this pro-
vides the variations in brightness; monochrome receivers will be practically
unaffected by the presence of the chrominance signal. Colour receivers require,
apart from a different tube, a separate circuit called the decoder to detect the
colour content of the scene.
The symbol used for the luminance signal is y' (or E~ , and sometimes it is
simply y), and the two transmitted colour difference signals are known as
(R' - V') or (Ek. - E~) or just (R - Y) and (B' - y/) or (EB - Ey) or (B - Y).
To produce a colour difference signal, we must first have a camera output
voltage corresponding to, say, the red content of the scene, which after some
additional processing we shall call R/. We also need a signal voltage which depends
upon the luminance value of the same scene-that is, y/. The latter signal is sub-
tracted from the R' voltage in an electronic circuit to produce (R' - V'). The
circuit which performs the addition (or subtraction) of voltages is called a matrix.
The second transmitted colour difference signal is obtained in a similar manner to
yield (B' - V'), where BI is related to the blue content of the scene.
The next step is to modulate the carrier with the signals ready for transmission.
The process of modulation in a colour transmitter is fairly complex and it will be
discussed separately.
We can now anticipate what will happen in the receiver, which, after detection
and decoding, will have available y', (R' - V') and (B' - V') information. The
original colour signals, namely R' and B' , are then recovered using a matrix circuit
which adds together the detected voltages
(R' - V') + y' = R'
(B' - V') + y' = B'
The reader will have noticed that the third primary colour, green, appears to be
missing from this system; one might expect to fmd (G' - V') somewhere. It can be
shown, however, that the (G' - y/) signal may be obtained by adding a proportion
Transmission of Monochrome and Colour Television Information 11

of the decoded (R' - Y') voltage to a proportion of the (B' - Y'). The derivation
of this relationship will not be carried out in this book, but it can be proved that
(G' - Y') =-0.51 (R' - Y') -0.186 (B' - Y'); the minus signs indicate that the
signals used have to be of the opposite phase from those processed in the remainder
of the circuit. Finally, the third primary colour, G' , is obtained from another
addition
(G' - Y')+Y' = G'
Why is (G' - Y') not transmitted? As shown above, it is only necessary to
transmit two of the three colour difference signals. It is preferable to exclude
(G' - Y') because it has been observed that, of all three, this one has the lowest
magnitude; hence, it could be much more affected by noise, distortion, etc.

2.4 Colour TV Camera Arrangements

In view of the need for the three transmitted signals, it is possible to have a number
of camera arrangements. At least three camera tubes must be used in a combined
unit, all three looking at the scene through the same lens, but the image is first
split into its colour content before it reaches the targets of the respective tubes.
The task of splitting the light is performed by dichroic mirrors or prisms mounted
between the lens and the tubes. The most important property of a dichroic mirror
is that it reflects one colour only and allows the remaining colours to pass through
it. An example of a much simplified three-tube arrangement is shown in figure
2.4(a).
The image from the lens falls on the dividing mirror (this is not dichroic), which
allows some of the light to reach the luminance signal tube, the remainder being
reflected towards the first dichroic mirror. Here the red content of the image is
reflected and other colours pass through to the second dichroic mirror. This mirror
reflects only the blue content and directs it to the tube. After some initial amplifi-
cation and processing, the luminance and the 'colour' signals are fed to their respec-
tive matrix circuits, where the two colour difference signals are formed.
Yet another, and more compact, three-tube camera arrangement is shown in
figure 2.4(b). Now the dichroic prisms supply the tubes with the contents of the
scene corresponding to the three primary colours, red, green and blue. The prism
PI is provided with a blue reflecting dichroic surface so that the blue light is direc-
ted towards the 'blue' tube; similarly, the dichroic surface of P2 reflects the red
contents of the scene towards the 'red' tube. The luminance signal is not available
directly, and must be formed by the addition of the three colours in the Y-matrix.
In other words we have to reverse the action of the dichroic prism by electronic
means. Finally the 'red' and the 'blue' outputs are matrixed with the luminance to
produce the colour difference signals.
Studio cameras sometimes have four tubes, so that the set-up appears to be a
combination of the two just described. In this case three tubes are used to generate
the colour signals; the fourth tube receives the complete image and produces the
12 Television Principles and Practice
dividing
camera ~mirror

--
luminance
lens camera tube

~ amplifier r----1--_---- E~ (Iuma)

I
I red camera
tube
dichroic

(E~ -E~) ~~;~:r:nce

mirror matrix
I }
I blue camera
: tube signals
dichroic
mirror '\-~ H amplifier ~I---""~I matrix t--(E~-E~)
(a)

r - - - - - - i amplifier r--------_-..,
matrix (E~ -E~)

camera
lens
green camera

H
tube
-~ amplifier ~ H-_-E~

- - - - - ' amplifier ~----------'

(b)

Figure 2.4 Three camera tube arrangements required for the production of colour
TV signals: (a) simplified arrangement using dichroic mirrors; (b) practical arrange-
ment with dichroic prisms

luminance signal. The latter output is used as a source of reference which is com-
pared with that obtained by the addition of the red, green and blue camera tube
outputs; any adjustments can then be made to equalise the two signal routes.

2.5 TV Signal Bandwidth

We now turn our attention to the range of frequencies occupied by the complete
television signal, which we call the bandwidt~. This is a very important property
affecting not only the transmitter (it determines the operation of the transmitter,
Transmission of Monochrome and Colour Television Information 13

allocation of channel frequencies, etc.), but also the receiver design (principally its
iJ. and video amplifier stages).
As we noticed from the operation of the TV camera, its output is in the form
of pulses of electric current corresponding to the pattern of the image being
scanned by the electron beam. In order to appreciate the frequencies involved in
reproducing such a pattern, we have to consider the speed with which the beam
traverses the target; as we shall see, the duration of one picture line is 52 JiS. During
that period the beam current has to rise and fall to follow the variations in picture
brightness; the changes in the amplitude of the video waveform represent its fre-
quency. If the image consists of fine detail, then the changes in brightness are very
closely spaced, so that the repetition rate of the camera output voltage is high-it
can be said that the video frequency of fine picture detail is high. For coarse picture
detail the amplitude of the video waveform varies less and the resultant frequency
is lower. The diagrams in figure 2.5 show parts of two picture patterns and the
corresponding video waveforms along the scanning line X-X. In part (a) the display
is of a relatively coarse pattern-when the beam current encounters the black area,
it falls to zero and the output voltage is a maximum. The reader will recall that the
video output from the camera tube circuit in figure 2.l will be equal to the supply
in the absence of a potential drop across the resistor R. When the beam scans the
white area of our pattern, the current rises and the output voltage drops. Similar
reasoning can be applied to the fme pattern in part (b). We have chosen a simple

I I
I I

x-~f1=Tlf
picture I : I I
-x x-

I
picture I
·4
:c-x

pattern pattern

video waveform video waveform

'"
"0 '"
"0
.E .~
a. a.
E E
'" '"

time time

(a) (b)

Figure 2.5 Effect of picture detail on video frequency. The video waveforms
correspond to the portion of the pattern scanned along the line x-x. (a) Coarse
detail-low frequency; (b) fme detail-high frequency
14 Television Principles and Practice

geometrical display, but the electrical effect would be the same if we analysed a
more usual type of televised scene.
So far we have considered only the changes in the video waveform along one
horizontal line; the display pattern also changes in the vertical direction. Here the
complete picture is split into a fixed number of lines; if there are more lines per
picture, the beam has to move much faster to complete the scan, and the time
difference between the video peaks in figure 2.5(a) and (b) is shortened, resulting
in a higher video frequency. The greater the number of lines, the better the quality
of the displayed picture, because the changes in brightness in the downward direc-
tion are then likely to be dictated by the actual picture detail rather than by the
line structure of the display. British readers can easily verify this by comparing a
405-line with a 625-line picture. The 405-line system originated in the days when
electronic circuits could not easily cope with very high frequency signals.
Theoretically, the complete TV system should be able to transmit and receive
satisfactorily a picture which could consist of the chequerboard pattern shown in
figure 2.5. This display is divided into a number of elements, or small squares,
whose sides correspond to the 'thickness' of one line. In a 625-line system the
actual number of lines in one picture is 585 (as will be shown later, the remaining
40 lines are associated with other functions); therefore the dimensions of each
picture element are equal to 1/585 of the overall height.
The total number of picture elements on a square screen would be in this case
equal to 585 x 585. However, the transmitted display is a rectangle whose propor-
tions are given by the aspect ratio of 4:3 (which means that the picture height is!
of its width). Since the number of elements is increased by the aspect ratio, the
frequency of the corresponding video waveform must increase accordingly.
The speed of the downward movement of the beam is fixed by the field fre-
quency. TV systems were initially designed to have the field frequency locked to
the a.c. mains supply. In a colour TV system a very precise relationship exists
between the various frequencies present and a drift in mains frequency would be
unacceptable. Nevertheless, the field frequency is equal to the nominal supply
frequency, which in the UK is 50 Hz.
If all the picture lines were scanned during each field, then the highest video
frequency required to reproduce the fine chequerboard pattern would be given by
video frequency = 1- x maximum number of elements x aspect ratio x field frequenc)
(As the picture elements consist of alternate black and white squares, two such
squares correspond to one cycle of video frequency-hence the division by 2 in the
above expression).
Substituting the values in the formula would give a video frequency of nearly
11.5 MHz. In fact, if the signal were to reproduce such rapid changes in brightness
as the chequerboard pattern implies, the required frequency range would have to
be extended still further. Finally, the process of modulating the video information
upon a carrier gives rise to sidebands, which increase the bandwidth even more.
In order to reduce the overall bandwidth of a TV transmission, interlaced
Transmission of Monochrome and Colour Television Information 15

scanning is universally used. It has been assumed so far that all the 585 active
picture lines are 'drawn' on the screen during each downward movement of the
beam. In fact, only half that number is scanned during one field; the second half-
which is produced during the next field-is placed in the spaces between the first
group of lines. The pri'nciple of interlacing of the lines of two fields is illustrated in
figure 2.6. With this method each picture (frame) consists of two fields. The field
frequency is still 50 Hz, but there are only 25 pictures scanned per second. Despite
such a low picture (frame) frequency, flicker is avoided, partly because of the
persistence of the image on the screen and partly because of the 'inertia' of the eye,
which retains the image for a fraction of a second.
Interlaced scanning halves the value of the highest video frequency as calculated
above, because only 292-!- active picture lines (3121 overall) are associated with
every cycle of the field frequency. The calculations previously quoted were
simplified to show the main factors involved; there are additional considerations
which influence the fmal choice of video bandwidth, which is 5.5 MHz in the
British 625-line standard. Since we have considered only the transmission of a
black and white pattern, this bandwidth applies to the luminance or monochrome
signal. The presence of other information will be discussed later.

line
number line
number
152--------------------------
-- - - - - - - - - - - - - - - ----465
153 _______________ ----466
1~ ____________________ %7
155--------------------------
-- - - - - - - - - - - - - -------·468
156--------------------------

Figure 2.6 Principle of interlaced scanning. Lines 152, 153, etc., are scanned
during one field; the interlaced lines 465, 466, etc., appear during the subsequent
field. (Flyback not shown.)

2.6 Signal Sidebands

The problem of the transmitted signal bandwidth is further complicated by the
process of modulating the carrier with the video information. It can be proved
both experimentally and mathematically that when a carrier wave at a certain
frequency,fc, is amplitude modulated by a signal at another frequency, say Is,
the resultant output has been altered to such an extent that it consists of three
signals at different frequencies, namely
(1) The carrier frequency,fc.
(2) The carrier plus the modulating [Link] + fs; this is called the upper
sideband.
(3) The carrier minus the modulating [Link] - Is, called the lower sideband.
16 Television Principles and Practice

The output from the transmitter now covers a very wide frequency range if it
includes both the lower and the upper sidebands. For example, if the highest
modulating video frequency is 5.5 MHz, then the transmitted signal would extend
from 5.5 MHz below the carrier frequency to 5.5 MHz above the carrier, making
the bandwidth 11 MHz. We can now see how much wider it would have been if we
did not use interlaced scanning!
There would be two major practical consequences of using such a high band-
width in the transmitted signal: firstly, each transmitter would occupy this
frequency range and the available number of channels would have to be reduced
to accommodate these wide spacings between them; secondly, the receiving circuits
would have to cope with large bandwidths, adding to the complexity of the design
(and servicing!).
Fortunately, it is possible to introduce economies which reduce the necessary
bandwidth before transmission. It so happens that all the information required is
available in either sideband, and the information in the upper sideband simply
duplicates the contents of the signal in the lower sideband. If we could remove one
of the sidebands, we would still broadcast and receive the original programme
intelligence. The diago,am in figure 2.7 shows the distribution of video frequencies
and the sidebands associated with the carrier. We note that the output remains
constant to include signals up to 5.5 MHz; afterwards it is attenuated, so that there
ought to be no video information extending beyond 6 MHz. The slope of the
diagram between 5.5 MHz and 6 MHz indicates that it is not possible to reject

..-__-+___ channel width _ _ _ _-1f--!

8 MHz
I

I
,------ ---+""r"--+---------~
I
, I
I I
I I
I :

-----
I I

-6 -1.75 -1.25 fc sideband frequency +5.5 +6

(MHz)
vestige of
lower sideband

~,----------~~,----------~------------~",--------,j
lower sideband upper sideband

Figure 2.7 Diagram of the principle of vestigial double sideband transmission

(transmitted frequency spectrum shown by the solid outline)
Transmission ofMonochrome and Colour Television Information 17

abruptly all frequencies beyond a certain limit; any filter used offers the relatively
gradual reduction in output. For the same reason it is not possible to cut off one of
the sidebands along the vertical line passing through the fe; instead, it would be
necessary to start filtering at some frequency before fe so that the output becomes
zero at the carrier frequency. In TV broadcasting this is unacceptable, because we
would lose the very important lowest video frequencies which lie close to the
carrier. The alternative is to start filtering after fe, which leaves part of the other
rejected sideband intact. In the UK the upper sideband is transmitted together with
part of the lower sideband. This is called vestigial sideband transmission because the
remaining part of the lower sideband is called vestige (meaning a trace). From
figure 2.7 we see that the overall width of one channel is from -1.75 MHz (negative
sign to denote the lower sideband) to 6 MHz, at which the sound carrier is placed.
An additional 0.25 MHz is needed to accommodate the sound sidebands and also
to offer a buffer space before the next channel-the total is now 8 MHz instead of
about 12 MHz if both video sidebands were transmitted.
Incidentally, in sound broadcasting both sidebands are broadcast as they are
narrower; the full audio spectrum extends only up to 20 kHz. It must be pointed
out that frequency modulation produces the two types of sidebands, but the actual
frequency distribution is somewhat different from that outlined here for an AM
process.

2.7 Channel Allocation

From the above discussion the reader will appreciate why transmitting station fre-
quencies have to be suitably spaced to prevent mutual interference which over-
lapping sidebands could produce. As TV standards may be different in other
countries, the actual frequencies in use also tend to differ. At present, TV broad-
casting in the UK is carried in five bands of frequencies. Broadly speaking, band I
is from 40 MHz to 70 MHz and band III from 170 MHz to 220 MHz, and both are
known as v.h.f.; they are still used for the 405-line standard and, to some extent,
for the 625-line system fed from special relay stations and distributors. Band II,
88 MHz to 108 MHz, is reserved for f.m. sound broadcasting. TV transmission at
u.h.f. falls into either band IV, 470-580 MHz, or band V, 620-855 MHz. Each
station, or programme company, is given a channel number (letters are also used
on v.h.f.) which automatically standardises its vision and sound carrier frequencies.
Each channel on the 625-line standard occupies a bandwidth of 8 MHz which in-
cludes vision (luminance), sound and chrominance. In band I the channels are
A, B, C, R (also from 1 to 5 on the 405-line standard); in band III they are from
D to I (also from 6 to 13); in band IV from 21 to 34; in band V from 39 to 68.

2.8 Modulation of Colour Information

The introduction of colour transmission had to fulfil the requirement that the
existing bandwidth would not be extended by the presence of the chrominance
18 Television Principles and Practice

luminance sidebands
...c::
l!l
c::
o
u
E
~
u
"5.

-
l ow video
frequencies
(a)
-
high video
frequencies

vision chroma sound

sideband frequency (MHz) ____

chrominance
sidebands
(b)

Figure 2.8 Principle of frequency interleaving: (a) distribution of signal energy in

the luminance sidebands at high video frequencies; there are also frequency 'gaps'
into which the chroma signal can be fitted; (b) position of the chroma signal within
the existing video (luminance) bandwidth.

signal. At the same time, the colour signal and its inevitable sidebands should be
accommodated without causing undue interference with the vision (luminance)
and sound signals. Detailed study of the actual frequencies present in the video
sidebands disclosed that they do not possess every conceivable value up to 5.s MHz,
as might be suspected at first. Instead, they are associated with multiples of the line
time base frequency, so that there are gaps between those values which could be
filled by another signal. This frequency distribution is given by the fact that the
video waveform is not continuous but occurs at regular intervals caused by the
line scan.
The available gaps, as shown in figure 2.8{a), can be filled by the chrominance
Transmission of Monochrome and Colour Television Information 19

signal, provided that its carrier frequency is very precise. This ensures that the
chroma sidebands, which are also multiples of the line time base frequency, fall
into the vacant spaces. The actual chrominance signal frequencies must be restric-
ted in order to contain the resultant sidebands within the allocated channel band-
width. Such a restriction implies that fme picture detail is not transmitted in
colour. However, it was explained in chapter 1 that the eye notices fme detail only
as variations in brightness; the colour effect is produced by somewhat coarser
structure of the image. The reader will recall that fme picture detail corresponds to
high video frequency; larger elements result in lower video frequency. The outcome
of this reasoning is that the colour difference signals have a bandwidth of approxi-
mately 1 MHz; consequently their two sidebands are about 1 MHz each.
The chrominance signal is modulated on a carrier, usually referred to as a sub-
carrier, of a frequency normally quoted as 4.43 MHz, but whose precise value is
4.43361875 MHz. Such accuracy ensures the correct frequency interleaving of
the luminance and chrominance signals. This value also minimises the creation of
a dot interference pattern caused by the chroma signal on the screens of TV
receivers (monochrome sets in particular). For the explanation of this process the
reader must be referred to the more specialised books on the PAL colour television
system.
As the chrominance information is centred around 4.43 MHz within the
luminance spectrum, the upper sideband extends to (4.43 + 1) MHz, which
approaches 5.5 MHz, while the lower sideband extends down to about 3.4 MHz.
The diagram in figure 2.8(a) shows in a simplified form how the chrominance side-
band frequencies (broken line) can be interleaved with the luminance signal (con-
tinuous line). The diagram also explains why the colour sub carrier lies towards the
higher end of the video frequency spectrum. It appears that the average picture
content is biased towards the low video frequencies, as indicated by the height of
the verticals in the diagram. Therefore, by placing the chrominance in the region of
low luminance content, any interaction between the two is kept to a minimum. In
order to reduce interference between the different carriers comprising the compo-
site TV signal (in particular, between sound and chrominance), and also to save
transmitter power, the actual colour sub carrier is suppressed before transmission
takes place. It was shown earlier that, as a result of the modulating process, there
exist two sidebands and the carrier. The latter acts only as a 'vehicle' which simpli-
fies demodulation in the receiver; if there were a simple alternative to the diode
detector (see chapter 6), then in most applications the carrier would not have to be
transmitted. The diagram in figure 2.9 shows what would happen if the video wave-
form of figure 2.3(a) were modulated and the carrier suppressed. In the receiver, a
simple diode detector would give an output which corresponded to a rectified half
of such a signal. As we see from figure 2.9(b), this would have very little resemblance
to the original video signal because of the missing carrier reference. The reader is
invited to compare the appearance of normal modulated wave in figure 2.3(c) with
that in 2.9(a); the two envelopes, instead of being apart on either side of the zero
line, are now shown crossing over.
20 Television Principles and Practice

(a)

(b)

Figure 2.9 Possible effect on the normal video waveform if the vision carrier was
suppressed: (a) modulated waveform (compare with figure 2.3c); (b) output from
a simple diode detector [compare with the desired output in figure 2.3(a)]

To demodulate a suppressed carrier signal requires reinsertion of the missing

carrier at the receiver itself. For correct demodulation, the regenerated colour sub·
carrier must be in perfect synchronism with that used at the transmitter. This can
only be done if a colour synchronising signal is transmitted as part of the informa-
tion. Such a signal is known as the colour burst, and it is considered in the next
chapter.
There is also another problem in transmitting the chrominance signal. As des-
cribed in section 2.3, there are two separate components, namely (R' - V') and
(B' - V'). Our discussion so far allowed for one signal to be interleaved which we
called chrominance. To accommodate both components without interaction, a
method known as quadrature modulation is used. This means that the same fre-
quency sub carrier is used, but one of the modulators is supplied with the carrier
via a 90° phase shift network-that is, the two modulated signals are in phase
quadrature. As a result of this arrangement, when the sub carrier is passing through
its maximum in one modulator, it is zero in the other, and vice versa during its next
half-cycle. This system produces two separate signals which are added to form the
composite chroma signal as shown in figure 2.10. The demodulating process in the
receiver is the reverse of the modulation, as will be shown in chapter 7. There the
reintroduced sub carrier will have two paths 90° out of phase with each other, so
that the feeds to the demodulators will be in step with the transmitter and they will
'recognise' their respective signals as being the original (R' -" V') and (B' - V').
Transmission ofMonochrome and Colour Television Information 21

..
time base
sync. pulses

~
,
from vision
carrier oscillator

{ ~
.. to transmitter

-
from luminance
camera green vision carrier aerial
tube -
red
matrix
and amp.
r-~
composite
video amp. r-- modulating
amp.
....... amp.

""'-- (B'-Y')
matrix
- U
modulator
and amp.
burst
gating
t •
-
pulses

- (R'-Y')
matrix f---
V
modulator
and amp.
~
composite
chroma
adder

t
half·line sound carrier
:'900
frequency pulses • PAL switch
modulator
and amp.

t t
subcarrier
oscillator sound amp.

! from sound
frequency carrier oscillator
dividing
circuits

to line time base I

to field time base J

Figure 2.10 Simplified block diagram of PAL colour TV coder and transmitter
arrangement

Readers will note that in the block diagram in figure 2.10 we have introduced
new symbols, namely U to replace (B' - y') and V instead of (R' - y'). This is to
indicate the change in the character of the two initial colour difference signals, as
will now be explained.
22 Television Principles and Practice

The composite modulated chroma is added to the luminance signal to produce

composite video. The addition of the two could cause the amplitude of the output
signal to exceed the permitted limits of modulation. It was mentioned in section 2.2
[and in figure 2.3(c)] that the amplitude of the video waveform must not drop
below 20 per cent of the maximum. If chroma were superimposed upon the lumin-
ance signal, it would be possible to drive the composite video considerably below
that limit. Similarly, at the other end of the video waveform the added chrominance
could extend well beyond the synchronising pulse level; this point is discussed again
in chapter 3. To prevent overmodulating, the two colour difference signals have their
amplitudes reduced by what are known as weighting factors. Thus U and V are the
weighted and modulated (B' - y') and (R' - Y'), respectively. As a result of the
compression of their amplitude, the (B' - Y') signal is reduced by a half and the
(R' - Y') by approximately 12 per cent [more exactly, U =0.493 (B' - Y'),
V = 0.877 (R' - Y')] . They are restored to their original proportions in the receiver
amplifiers.
Referring to the block diagram in figure 2.1 0, the reader will observe that the
time base frequencies are also derived from the colour sub carrier frequency to
ensure luminance-chrominance interleaving, and to minimise the effect of sub-
carrier interference pattern on TV screens.

2.9 The PAL System

The colour encoding system used in the UK and in many other countries is known
as the PAL system, and was invented in Germany by the Telefunken Co. The
abbreviation PAL stands for Phase Alternation by Line. Referring to figure 2.10, it
will be seen that the subcarrier feed to the V modulator goes through the PAL
1800 (±90 0) switch. The purpose of the switch is to invert the phase of the subcarrier
during alternate scanning lines, which, when fed to the V modulator, causes the V
signal to reverse its phase every line. The reversal is brought about by the second
input signal to the PAL switch, that is, a square waveform at half the line frequency.
The output from the composite chroma adding circuit is (U + V) during one line
and (U - V) during the next, to revert back to (U + V) again, and so on. The applied
switching waveform is at half the line frequency, because one half-cycle will main-
tain the original phase of the sub carrier while the next half-cycle will cause the
reversal.
This method of colour transmission makes the system relatively free from changes
in actual colour (hue), which could otherwise occur owing to imperfections in the
transmission-reception chain or imperfections in the performance or alignment of
the receiver. Briefly, the above imperfections can cause a change in the mutual phase
relationships between the U signal, the V signal and the sub carrier used in the modu-
lation process (and a sample of which is also sent out as a separate signal in the form
of a colour burst). These changes, unless corrected, would produce 'confusion' in
the receiver demodulator, whose regenerated sub carrier relies on the various phase
relationships being correct in order to 'recognise' the original values of (R' - Y')
Transmission of Monochrome and Colour Television Information 23
and (B' - V'). As will be shown in the next chapter, relative alterations in the values
of the colour difference signals mean changes in the hues reproduced on the screen.
The phase reversal oflhe V signal causes the phase errors to be cancelled out (the
error is positive on one line and negative on the next); however, there is some deter-
ioration in the detected output in the form of reduced saturation. It is considered
to be less important though, because the viewer might notice this as a reduction in
colour intensity, easily counteracted by the colour control in the receiver. An
explanation of the effect of PAL switching will be given in chapter 7.
The reader will recall that there is a colour synchronising signal, the colour burst,
transmitted for the purpose of maintaining the phase of the receiver sub carrier
oscillator with that of the transmitter. PAL switching introduces a new problem,
because it causes the sub carrier phase to alternate through 180 0 every line. To
maintain perfect synchronism, the same process must be repeated in the receiver.
The phase of the burst is changed every line in step with the operation of the PAL
switch, which allows the receiver to identify which line contains a reversed V com-
ponent. The diagram in figure 2.10 shows how the colour burst is formed. During a
short interval of time, no video information is transmitted (details in the next chap-
ter) and a burst gating pulse is fed to the two modulators, which allows the sub-
carrier to pass through. (Normally the type of modulator used in this position auto-
matically suppresses the subcarrier.) The gating pulse is of a very short duration and
only 10 cycles at 4.43 MHz will be included in the composite chroma. From the
block diagram, readers will note that the burst will consist of two components of
the subcarrier: the one fed to the U modulator (U component) and that fed to the
V modulator via the PAL switch (V component). Since the two components of the
burst are fed to the composite chroma adder circuit, the resultant signal also under-
goes a phase change every line. This signal is sometimes called a 'swinging' burst
because of this alternation of its V component.

2.10 Other Colour Encoding Systems

In the United States a system is used known as NTSC (National Television System
Committee, formed by a number of TV manufacturers). Historically this was intro-
duced before the PAL system and, unfortunately, suffers from defects which the
PAL arrangement managed to reduce. Indeed, in many respects the two systems
have much in common-namely there are two chrominance signals quadrature
modulated. However, instead of the two colour difference signals being kept sepa-
rate, the American system uses a combination of (R' - V') and (B' - V') fu each
component. The system is so arranged that each chroma signal is associated with
certain colours according to the ability of the eye to resolve detail in different
colour. Because of this the bandwidth of one signal is almost three times the band-
width of the other. This was the inventors' solution to the problem of accommodat-
ing the required colour information within the available TV bandwidth. American
broadcasting standards use 4.5 MHz video bandwidth, which offers less space for
frequency interleaving when compared with the British 5.5 MHz standard. The
24 Television Principles and Practice

NTSC system also uses suppressed sub carrier modulation and a colour burst, but
there is no equivalent of the 1800 switching. Perhaps the main criticism of the US
colour TV system is that the various imperfections in the transmitting-receiving
chain mentioned in the previous section do cause a change in reproduced colours.
The viewer has to use a tint control (not to be confused with the tint control of
PAL receivers-see chapter 11) to counteract this.
In France and in several other countries a system known as SECAM (Sequential
a Memoire-line sequential transmission with memory) is used. The system uses
separate colour difference signals (R' -' V') and (B' - V'), but, instead of the two
being transmitted simultaneously as quadrature modulated signals, they are broad-
cast sequentially. During one line the composite video consists of the luminance
and modulated (R' - V'); during the next line we have the luminance and modu-
lated (B' -' V'). Since both signals are needed together to produce the colour
picture, the receiver must store (memorise) the alternate colour signal for the
duration of one line until the second chrominance component is available. As we
shall see later, the modem PAL receiver also uses a 'memory' (PAL delay line), but
for different reasons. In the French system the chroma signal is frequency modu-
lated and there is no need for colour synchronising.
In conclusion, it must be mentioned that there are also slight variants of the
PAL system dictated by the existing TV standards in any given country.
3 The Composite Video Signal
3.1 The Video Signal Waveform
In chapter 2 we discussed the formation of the various signals which together
make up the composite signal broadcast by the transmitter. Now we shall look at
these signals in some detail and study their waveforms. The design of a TV receiver
is based on the existence of standardised video signals; knowledge of their wave-
forms helps the servicing technician in fault-finding or setting-up.
The luminance (or monochrome) video waveform is produced by the camera
tube (or tubes) and the associated amplifiers as described in chapter 2. Time base
synchronising pulses are added to the initial waveform in order to synchronise the
scanning of the picture tube screen with the scanning of the camera tube target
layer.
The diagram in figure 3.1 shows in detail the time and amplitude relationship

4.7 fJ.S
line sync.
pulse
line blanking
12 fJ.S front porch

-[Link]
1.5 fJ.S
......-r-___- - - - a c t i v e picture line _ _ _ _
52 fJ.S
0.4
total line period
t---t--+------ 64 fJ.S -------i~'
black level
---- 0

peak white level

- - - - - - - - - - - - - - ----1.0

Figure 3.1 Time and amplitUde relationships in a video waveform

25
26 Television Principles and Practice

between the various parts of a typical video waveform. Two line synchronising
pulses have been added to defme the limits of the actual scan period. The ampli-
tudes are measured with respect to the reference point given by the black level.
The black level is ftxed at the transmitter and, ideally, it should also be maintained
at the receiver, as will be shown in chapter 6. Picture modulation must always start
from and end on 'black',irrespective of the actual content of the scene. To ensure
that this occurs, there are two short 'resting' intervals on either side of the line
synchronising pulse. The one preceding the pulse is called the front porch; this
period allows receiver circuits to change their voltage levels to that corresponding
to the black level. This is particularly important in the circuits leading to the
synchronising pulse separator, as it is necessary to provide a strictly defmed refer-
ence point, from which the separation of the pulses from the video can take place.
It is especially needed when the line ends on white-a large change in signal level is
required and that, in tum, takes time (for example, to charge or discharge various
capacitors). Following the synchronising pulse, the picture modulation is still
maintained at black level for a short period of time known as the back porch; this
allows the receiver line time base to complete the flyback. During a colour trans-
mission, the back porch is also a convenient part of the video waveform on which
to place the colour burst.
The tips of the synchronising pulses associated with the waveform in ftgure 3.1
correspond to the maximum amplitude of the composite video. This is, in fact, how
it appears at the output from the camera tube, where, for a very bright picture
called peak white, the beam current is a maximum hence the output voltage
becomes a minimum (section 2.1). Of course, the presence of various amplifters
can easily invert the appearance of the waveform as it happens, for example, in the
receiver, as will be shown in chapter 6.
One line of the video signal consists of one synchronising pulse, the two porches
and the actual picture information. The time duration of one line is 64 p.s, out of
which only 52 p.s is used to transmit picture detail (this is known as the active line),
while the remainder is occupied by the synchronising pulse and the maintenance of
the black level. Therefore the 12 p.s interval between the beginning of the front
porch and the commencement of the next picture waveform is called the line
blanking period.
The line time base frequency can be calculated from the total number of lines
scanned per second, which is the number of lines per frame (625) multiplied by the
number of frames per second (25); the resultant gives 15 625 Hz. The duration of
one line, or its periodic time, is 1/15 625 =64 p.s. The time division between the
active line and the blanking period is simply an accepted standard.

3.2 The Luminance Waveform of a Colour Bar Signal

The waveform in ftgure 3.1 represented one line of an arbitrary televised scene. If
an oscilloscope were used to display a video waveform which corresponds to an
average TV picture, the only clearly recognisable part of the waveform would be
The Composite Video Signal 27

the synchronising pulses. The picture detail of a large number of lines (there are
15 625 lines per second) would be superimposed to such an extent that the scope
display would become relatively meaningless.
With the advent of colour television, the use of precise waveform examination
during servicing became much more important and the colour bar waveform was
introduced. In a correctly adjusted colour receiver this type of signal produces a
display of eight vertical stripes of equal width in the following order (from left to
right): white, yellow, cyan, green, magenta, red, blue and black. This arrangement
is from the brightest bar (white) to the darkest (black), or, in other words, the bars
are in the order of decreasing luminance (see definitions in chapter 1). The reader
will note that, in addition to the black and white, the display has the three primary
and the three complementary colours. On the screen of a monochrome receiver
this waveform will produce eight bars of decreasing brightness-from white,
through different greys, to black. This type of pattern can be obtained either
from a signal generator or from a colour bar transmission (including the test cards).
In order to produce the actual colour display, the waveform must contain both

1.0
white

0.11
blue

back
porch

line sync.
pulse

Figure 3.2 Video waveform-luminance part of a colour bar signal (in monochrome
this waveform produces 'grey scale' display). Individual amplitUdes are shown with
respect to peak white
28 Television Principles and Practice

the luminance and the chrominance information. The required luminance waveform
is given in figure 3.2.; it resembles a 'staircase', and it is often called the luminance
steps. The amplitude of each step can be worked out from simple calculations,
which we shall now consider. The reader will recall from chapter 2 that the lumin-
ance signal is formed from the three primary colour signals mixed in specified
proportions. These amounts are given by the following equation
y' = 'O.59G' + O.3R' +'O.11B'
That is, the luminance signal voltage is made up from S9 per cent of the green
camera output voltage plus 30 per cent of the red and 11 per cent of the blue. The
amounts used are due to the response of the eye to different colours (chapter 1).
The camera tube amplifier outputs are adjusted so as to make them all equal when
white is viewed (the impression of white is created when all three colours are
suitably mixed). The actual voltages do not really matter at this point, so let us call
that output 1 unit. Therefore, for the white bar Y' = 0.59 x 1 + 0.3 x 1 + 0.11 x 1
= 1.0. This means that the luminance output voltage, either from a matrix or from a
separate luminance camera, would also have to be 1 unit. The next bar is yellow; as
shown in chapter 1, this is produced by mixing red and green; their proportions are
given in the above equation-that is, Y' = 0.59 x 1.0 + 0.3 x 1.0 = 0.89. For
example, for the red bar, only the red camera tube would produce its full output of
one unit (the rest would be off) hence, Y' = 0.3 x 1.0 = 0.3. Finally, for the black
bar the output is zero from all camera tubes, and Y' ='0. The reader is invited to
calculate the individual 'steps' and compare his answers with the values given in
figure 3.2. The polarity of the waveform in figure 3.2 is opposite to that shown in
figure 3.1, but it has been mentioned that the video signal may be inverted several
times during signal processing, both at the transmitter and at the receiver; therefore
either polarity may be encountered.

3.3 Chrominance Waveforms of the Colour Bar Signal

As shown in the previous chapter, the signal used to modulate the colour sub carrier
is made up from two colour difference signals: U(B' -'Y') and V(R' - V'). These
are also found in the decoder section of a colour receiver; the service technician
needs to be able to identify their waveforms in the process of fault-fmding. The
colour bar signal provides a stationary, easy-to-recognise set of waveforms which
can be displayed on the screen of an oscilloscope.
By use of a technique similar to the method employed in calculating the
luminance steps, the shape of the chrominance waveforms can be determined.
Let us consider the (R' -' V') waveform. The expression (R' -' V') means that
from the output of the red camera tube for any given colour we must subtract the
luminance value for that colour. For the white bar, as shown in section 3.2,
R' = 1.0 and Y' = 1.0; hence, (R' - V') == 1.0 - 1.0 = O~ For the yellow bar
R' = 1.0, Y' = 0.89; therefore (R' -' V') == 1.0 - 0.89 = 0.11. For the blue bar
R' ='0 (since there is no red in saturated blue), Y' ='0.11 (see figure 3.2); thus
 +0.70
red

+0.59
magenta

+0.11
0 yellow 0

blanking level

-0.59
green (a)

-0.70 +0.89
cyan blue

+0.59
magenta

+0.30
cyan

0 0
white black
blanking level

-0.30
red

-0.59
green
(b)
-0.89
yellow

0 0
white black
blanking level

-0.41
magenta

(c)

Figure 3.3 Colour difference waveforms of a colour bar signal (100 per cent
saturation): (a)(R' - y');(b)(B' - Y');(c)(G' - V'). Peak amplitudes for each bar
are given in relation to the maximum value of the luminance signal (see also figure
3.2)
30 Television Principles and Practice

(R' -' V') == 0 - 0.11 = -0.11. The reader can check the calculations for the
remaining colours and compare them with figure 3.3(a).
The second colour difference signal, (B' -' V'), is shown in figure 3.3(b). For
the white bar B' = "1.0; hence, (B' -' yl) = 1.0 - 1.0 = O. For the yellow bar B' ="0
(no blue in fully saturated yellow); therefore (B' -' yl) == 0 - 0.89 = -0.11; and so
on for the other bars.
The third colour difference signal, (G' - V'), is not transmitted, but it has to be
formed in the receiver as outlined in section 2.3. It is instructive at this stage to
construct the (G I - ' V') waveform, since the method is similar to that for the other
two. For the white bar G' ='1.0; hence (G ' _'yl) == 1.0 - 1.0 = O. For the yellow
bar G' = "1.0, and (G ' _'yl) == 1.0 - 0.89= 0.11. For the red bar G' ="0; hence
(G' _'yl) == 0 - 0.3 = -0.3. The complete waveform is given in figure 3.3(c).
One very important conclusion from the above calculations is that for both
white and black (or any shade of grey, for that matter) all three colour difference
signals are zero. This means that when a monochrome picture is being transmitted,
the chrorninance signal automatically disappears. This makes the colour decoder in
the receiver temporarily 'redundant'; indeed, that section of the receiver is then
automatically shut down to prevent unwanted coloured pattern appearing on the
screen (see also chapter 7).
It will be noted from the colour difference waveforms that the synchronising
pulses are'absent; these are part of the low frequency luminance information and
they would not appear in the 4.43 MHz chrominance channel.
The values of the waveforms given in figure 3.3 correspond to fully saturated
colours-that is, with no white light added. For example, for the yellow bar it was
assumed that only the red and the green outputs were present. If the colour were
desaturated, then the additional white would consist of red, green and blue; there-
fore desaturation means that there is always an output, however small, from each
camera tube. It has been argued that the use of fully saturated colours in test
patterns is unfair, since the colours in a normal programme material will always be
desaturated to some degree. For this reason the colour bar signal transmitted by the
broadcasting authorities can be specified as being, for example, 95 per cent saturated
The presence of desaturation and the weighting factors complicate the arithmetic
involved in the calculation of the amplitudes of the above waveforms. However,
from the practical point of view the shape of the signal waveforms remains basically
similar to that shown in figure 3.3.
The modulated colour difference waveforms can now be considered-that is,
when the weighted (R' -' yl) and (B' -' yl) signals become V and U, respectively.
If the usual amplitude modulation technique were adopted, each colour difference
signal would form the upper and the lower envelopes of the subcarrier in a manner
similar to the video waveform in figure 2.3. As the colour sub carrier is suppressed,
the two envelopes crossoveLthe zero line [compare the possible effect on the video
waveform in figure 2.9(a)] ,and the resultant waveforms are as shown in figure 3.4.
The 'shading' inside the waveform denotes the high frequency sidebands which are,
of course, present in this signal.
The Composite Video Signal 31
The precise method of mathematical construction of the composite chroma
waveform is somewhat lengthy, and it is left to the reader to obtain the values
shown in figure 3 .5(b) with the help of the following explanation. Waveforms can
be represented by a rotating phasor or, where the resulting wave is the sum of two
other waveforms, by the sum of two phasors. Since the chrominance signal is pro-

0.61 E 0.61
cyan ~ ~ ~ g,
red
cis,oE
0.44 ~ 0.44
e:
yellow ",e: blue
",ill
. "'~
0.15
00>
~
'"ciE'" 0.15
0.10 0.10 cyan red
yellow blue

white black

(a) V signal (b) U signal

Figure 3.4 Modulated and weighted colour difference signals (100 per cent satura-
tion) corresponding to a colour bar display. Amplitudes referred to the maximum
value of the luminance signal (see also figure 3.2)

~
0.63 e: '" ~ 0.63
cyan ~ ~ It) co red
V signal 0.44 0 0-> 0. E 0.44
colour yellow blue

I
+0.52 0.59 composite b urst colo ur
I
:
chroma =
j(U 2 +V2)
o.2 burst

I white black
I ... I
I
I
, I i
I

11
JrIl
I

il
_ _ _~I +0.29 U signal

(a) (b)

Figure 3.5 Composite chroma waveform of a colour bar signal (100 per cent
saturation): (a) phasor addition of quadrature modulated signals (U + V); amplitudes
correspond to the magenta bar (see also figure 3.4); (b) composite chroma waveform
with colour burst
32 Television Principles and Practice

duced by the addition of the U and Y signals, the resultant chroma can be repre-
sented by the U and Y phasors [see figure 3.5(a)] . The process needs to be repeated
for every colour bar except, of course, black and white. The values of U and Yare
obtained from (B' - V') and (R' - V') multiplied by their respective weighting
factors (see section 2.8 and figure 3.3). The resultant chroma can be calculated by
applying Pythagoras' theorem to the U!V phasor diagram, so that
chroma = v'(U2 + y2)
The waveforms shown in this chapter include the colour burst, since at the trans-
mitter both the U and Y signals are accompanied by the burst, as implied in the
block diagram in figure 2.10. In the receiver the colour burst is deleted before the
separation of U and Y takes place; therefore the reader would come across suitably
modified waveforms, as shown in chapter 7. It will be noted that the amplitude of
the burst is 20 per cent of the peak white value of the luminance signal.

(a)

083 0.90
. 0.76

---m--- -----m-
(b)

- - - - - - maximum carrier level

0.33

absolute limits
of chroma
peak white level
"'Willill~----------

- - - - - - - - - - - - - - - - - zero vision carrier level

(c)

Figure 3.6 Formation of the composite video waveform of a colour bar Signal:
(a) luminance waveform; (b) unweighted composite chroma = v'[(R' - y')2 +
(B' - y')2] with burst added; (c) composite video obtained by adding waveforms
(a) and (b); the resultant exceeds the carrier levels (unacceptable overmodulation).
Weighting factors are applied to maintain the composite waveform within the abso-
lute limits of chroma
The Composite Video Signal 33

3.4 The Composite Video Wavefonn of the Colour Bar Signal

Composite video denotes that both the luminance and the composite chroma
signals are present after the fmal addition in the encoder (figure 2.10). At this
stage it is of interest to recall why weighting factors have to be used. Figure 3.6(a)
represents the luminance waveform similar to that in figure 3.2, but in this case the
polarity has been changed to show what happens when negative modulation of the
vision carrier takes place. Figure 3.6(b) shows a composite chroma waveform made
up from the modulated (R' - V') and (B' - V') without the weighting factors. The
addition of the individual values of the composite chroma to the corresponding
values of the luminance gives the composite video in figure 3.6(c). Ifwe now in-
clude the vision carrier reference levels (zero and maximum), we note that the
resultant video will considerably over-modulate the carrier. For the yellow, cyan
and green, the presence of the chroma signal would drive the vision carrier below
its zero level. If this were allowed, the vision detector would clip these excursions,
which would result in distortion.

- - - - maximum carrier level

- - _. black level

absolute limits
of chroma

- - - - - - - - - - - - - peak white level

- - - - - - - - - - - - - - - - - - - - - - - - - - - zero carrier level

Figure 3.7 Composite video waveform of a colour bar signal (100 per cent satura-
tion). Weighted chroma (figure 3.5) added to the luminance waveform (figure 3.6)-
modulation limits no longer exceeded. If colours are less than 100 per cent saturated,
the chroma amplitude is reduced and the resultant is well within the set limits

At the right-hand end of the waveform on the blue bar figure 3 .6(c), the result-
ant would go beyond the peak of the synchronising pulse, which is well above the
vision carrier amplitude; this would again be clipped and further distortion would
occur. The values of the weighting factors were chosen to restrict the composite
video to within 33 per cent of the peak white amplitude beyond either the white
level or the black level. The modified composite video waveform, together with
the 33 per cent limits, are shown in figure 3.7. As can be judged from figure 3.3,
it is the (B' - V') signal which has the highest amplitude; therefore the weighting
factor is almost 0.5, compared with approximately 0.9 for the (R' - V').
34 Television Principles and Practice

3.S The Field Synchronising Pulse Sequence

Field synchronising pulses are part of the composite video signal; they ensure that
the receiver field time base operates in step with the vertical scanning of the camera
tube. tine time base synchronisation is achieved by a single pulse transmitted at
the end of each picture line. Field synchronising information consists of a sequence
of pulses which are transmitted at the end of each vertical scan. There are 15 pulses
associated with this sequence: 5 narrow pulses (2.3 J.J.S each) called equalising pulses
(see figure 3.8), followed by 5 broad pulses (27.3 J.J.S each) and, finally, there is
another set of 5 equalising pulses. Firstly, this combination ensures correctly inter-
laced vertical scanning; secondly, it facilitates the separation of the line and field
synchronising pulses by the respective time base circuits in the receiver; thirdly, it
maintains line time base synchronisation during field flyback.
The entire sequence is shown in figure 3.8, from which it will be seen that

Line
nUmber! 309 ! 310 ! 311 ! 312 ! ! ! ! ! !
313 314 315 316 317 318 ! 319 !
--.-----....----.---
field synch. pulses equalising suppressed
27.3 iJ.S each pulses picture
lines

line 312t
last picture line field pulse
of first field begins

line
number

line 625 field

pulse begins

final half-line
of second field

Figure 3.8 Field synchronising sequence for two successive fields. Vertical arrows
show the instant of line time base synchronisation during the field flyback period.
The complete sequence, together with the suppressed lines (only a few shown),
reduces the number of picture lines to 585
The Composite Video Signal 35

following the last set of equalising pulses there are a number of 'blank' lines
without any picture information. This is done to allow the field flyback to be com-
pleted before the scanning of the actual picture begins. The pulses in the field syn-
chronising sequence occur at twice line frequency; therefore there is (me pulse at what
would have been the end of a line and one in the middle of a line. The necessity for
the 'mid-line' pulses arises from the fact that there are 625 lines which must be
divided between two successive fields. Therefore, the first field ends half-way
through line 313, while the second field of the interlaced pair ends with a complete
line 625. There are also differences at the beginning of each field; the first field
starts at the beginning of line 1, and the second field commences from the middle
ofline 313. The 'mid-line' pulses are needed to initiate the field flyback sequence
after the first 312t lines. Those pulses which coincide with the end of a line are
necessary to maintain horizontal synchronisation. The two sequences shown in
figure 3.8 demonstrate how the different pulses are utilised to synchronise the two
time base circuits at the end of the first and the second field, respectively. The line
time base triggering intervals have been marked by arrows; it can be seen that they
occur at a regular rate irrespective of the nature of the pulse involved. The details of
synchronisation in the receiver will be described in chapters 9 and 10, where it will
be shown that the pulse processing circuits in the line time base respond to pulses
of any width. The field trigger circuit is designed to respond to the broad 27.3 JlS
pulses in the sequence.
The differences between consecutive fields could lead to slight timing differences
in the triggering of the field time base at the end of each vertical scan. Such in-
accuracies would cause poor interlace-that is, the lines associated with the first field
would not be scanned exactly half-way between those of the second field. The tim-
ing error would be too small to cause an obvious loss of synchronisation, but it
would be sufficient to produce a slight displacement of one vertical scan with respect
to the next. Equalising pulses are transmitted in the synchronising sequence to en-
sure perfect interlace in the manner to be described in chapter 9.
It has to be pointed out that while equalising pulses are desirable, they are not
included in some transmitting standards; for example, the British 405-line system
does not use them.
It was mentioned in section 2.5 that the TV picture consists of 585 lines instead
of the accepted standard of 625. The apparent discrepancy can now be explained.
The line time base does indeed provide 625 lines during a complete 'two-field'
period, but the latter must include two synchronising sequences, shown in figure
3.8. Obviously, no picture information can be transmitted in that time, or in the
period immediately after each sequence. Broadcasting standards allow for the field
fly back to be completed during an interval equivalent to 12-!- lines following the
second set of equalising pulses. Therefore the number of lines which are not avail-
able for picture modulation is: 2-!- for the first group of equalising pulses, 21 for
the broad field pulses, 2-!- for the second set of equalising pulses, and, fmally, 121
lines for the completion of field flyback. A total of 20 lines is 'lost' for every field,
so that 40 lines are subtracted from each frame. The 20 line interval associated with
36 Television Principles and Practice

the synchronising sequence is called a field blanking period. Two such periods can
be observed on the screen of a receiver in the form of a broad black horizontal band
whenever correct vertical synchronisation is lost.

3.6 Teletext Information

Teletext is a form of coded transmission of information by the broadcasting organ~
isations which is carried out in addition to the normal picture information already
described. In the UK the BBC use CEEFAX and the IBA system is known as
ORACLE.
Some of the lines during the field blanking intervals are used to transmit binary
coded infonnation, which is in the form of black and white dots modulating the
vision carrier. A special decoder is needed in the receiver to store the incoming
signals and translate them into a display on the screen. The received information is
grouped into sections, known as pages (for example, weather reports, news items,
etc.), which the viewer can recall from .:he electronic store by operating appropriate
controls. The teletext decoder involve~ complex circuitry, which is most conven-
iently provided by integrated circuits. Without a decoder, the information is mean-
ingless, although the modulating dots can be seen if the field blanking bar is dis-
played on the screen.

3.7 Intercarrier Sound Transmission

TV sound is a frequency modulated signal placed on its own carrier whose frequency
is exactly 6 MHz higher than that of the vision carrier. The term intercarrier refers
to the method of separating the sound information from the video in the receiver.
The two carriers, converted into their respective intermediate frequencies in the
tuner, are amplified together in the receiver circuits and then fed into a mixer
circuit to produce a beat frequency. The difference frequency between the two
signals is 6 MHz, which becomes the intercarrier frequency. The actual audio infor-
mation must still be detected from the 6 MHz frequency modulated signal.
The conversion into the intercarrier frequency can only be achieved if the
amplitude of the modulated vision carrier is not allowed to drop to zero. Some
vision carrier must remain to beat with the sound carrier; for this reason the peak
white of the luminance signal does not fall below 20 per cent of the maximum
amplitude of the vision carrier. Further discussion of the intercarrier sound system
is deferred to chapter 8.
4 The Essential Features of a
TV Receiver

4.1 Introduction
The purpose of this chapter is to introduce a complete TV receiver in a block dia-
gram form. The reader will be able to see at a glance how the composite signal is
processed before the individual outputs are obtained. A colour receiver has a num-
ber of features which are common to its monochrome counterpart. The practical
circu~ts of the two types may differ in detail because of the more stringent require-
ments of a colour set, but the basic principles are similar. The block diagram
approach is most helpful when one is dealing with unfamiliar receivers, whose circuit
diagrams can then be divided into their various functions. In modern designs a num-
ber of integrated circuits are used to replace the blocks in our diagram. Alterna-
tively, the layout may be in the form of modules or panels which carry the com-
ponents associated with essential functions so as to simplify initial fault-finding.
In field servicing it is then a simple matter of replacing a faulty module with a new
one, since efficient repairs may require either a high degree of skill or complex test
equipment.
Our discussion follows the diagram shown in figure 4.1, beginning with the signal
input from the aerial. Detailed treatment of each section is given in subsequent
chapters.

4.2 The Tuner

The tuner receives the signal of a selected channel from the aerial; the amplitude
of the signal could be in the order of microvolts at the frequency up to several
hundred MHz. Such a small input could be easily swamped by electrical 'noise'
signals, which are very low voltages generated at random by the various compon-
ents used in the circuit-transistors, resistors, etc. In audio amplifiers noise voltages
cause a 'hissing' noise through the loudspeaker; in video amplifiers noise shows as
a very 'grainy' background to the picture. To prevent the effect of noise destroying
the entertainment value of the programme, it is essential to start with the largest
possible signal very early in the receiving chain. This requires an efficient aerial
installation, which is then followed by the amplifier in the tuner.
37
 l ~:~tor ~I audio amp] c(] loudspeaker

to venical
I---------""..~ deflection coils
~various
r----.. . flyback pulses

(G'- Y') matrix;

(R'- Y') and (B'- Y')
amplifiers;
R', G', B'matrix

colour I•
decoder

I.
I
L I I I
-----------------------.---1
Figure 4.1 Block diagram of a 1V receiver. Parts enclosed within broken line apply to colour receiver.
In a black and white receiver the output from the video amp. feeds a monochrome tube directly
The Essential Features of a TV Receiver 39

Handling very high and ultra high frequency signals is not easy, especially where
a large number of amplifying stages are needed to produce a usable output. In addi-
tion, each stage would have to be retuned whenever a different channel were to be
selected. To overcome these problems the incoming radio frequency (r.f.) signals
are converted in the tuner mixer stage to anew, lower frequency known as the
intermediate frequency (i.f.). The value of the i.f. remains constant irrespective of
the channel selection, so that continuous retuning is not necessary, which simplifies
the design of the succeeding i.f. amplifiers.

4.3 The I.F. Strip

The i.f. strip consists usually of three amplifying stages, which, between them, pro-
vide most of the amplification of the received signal. They are also responsible for
ensuring a good quality picture which is free from mutual interference between the
carriers comprising the composite video signal. In a colour receiver the design of the
i.f. circuit is more critical than in monochrome sets, since it has to provide suffi-
cient amplification to both the chrominance and the luminance signals.
In order to maintain a relatively constant picture quality under varying recep-
tion conditions, automatic gain control (a.g.c.) is used to alter the gain of either one
or two of the i.f. amplifiers.

4.4 The Vision Detector and Video (Luminance) Amplifiers

When the magnitude of the iJ. signal has reached about 2 V, the composite video
can be detected or separated from its intermediate frequency carrier. This is most
frequently done by a diode detector, which rectifies the modulated i.f. to recover
the original envelope. This is followed by a relatively simple filter which removes
the residual i.f. Alternatively, an integrated circuit could be used; this operates as a
synchronous detector which, by sampling the modulated carrier at a very fast rate,
'sees' the envelope voltages during each sampling interval. In this way the original
envelope can be built up from the individual voltage levels.
It is also at the detector where the separation of chrominance and sound inter-
carrier from the composite video is possible so that the colour decoder and the
sound section can be fed with their respective signals. In some receivers these take-
off points can be found later in the video amplifiers.
Monochrome receivers can only make use of the luminance signal; they are
relatively unaffected by the presence of the chrominance information. The signal
is processed in video amplifiers, in which its voltage level is raised from the detected
I V peak-to-peak to perhaps 150 V peak-to-peak; this is required to fully drive the
picture tube from black to peak white. In a colour receiver the arrangement is
similar, but the counterpart of the video amplifier is now called the luminance
amplifier. To obtain a colour picture, the luminance signal has to be added
(matrixed) to the colour difference signals, which are processed separately in the
decoder. This addition either may take place in a circuit prior to feeding the tube
40 Television Principles and Practice

or may be done using the electrodes (cathodes and control grids) of the picture
tube itself; both methods can be found in practice. Our diagram shows a matrix
block (within the colour decoder) external to the cathode ray tube; this is a very
popular arrangement in modern receivers.
line and field synchronising pulses are obtained from the video waveform; hence
the need for the feed to the synchronising pulse separator.
The automatic gain control (a.g.c.) voltage may be derived from the amplitude
of the video signal, and there is an a.g.c. stage which could be switched by line fly-
back pulses in some designs. line flyback pulses are also fed to the luminance
stages (or the matrix) to provide black level clamping. This ensures that every line
of scan begins from the correct black level as determined at the transmitter.
The luminance and chrominance signals in a colour receiver follow separate
paths, at the end of which both signals are combined again. The propagation time of
signals through the colour decoder is longer than that in the luminance amplifier;
a luminance delay line is incorporated to delay the signal, which ensures perfect
registration of the luminance and chrominance information.

4.5 The Colour Decoder

This part of the circuit is applicable to every colour receiver. Its purpose is to de-
code the colour difference signals from the composite chrominance signal whose
carrier has been suppressed. The decoder has a main path which processes the
chroma itself, and a number of auxiliary circuits to provide the services necessary
for the decoding. In many respects the colour decoder resembles that of the encoder
at the transmitter because of the need for the reinsertion of the missing sub carrier;
without it, demodulation of the colour difference signals would not be possible.
In the main path of the chroma signal there are a number of amplifiers which
are tuned to the subcarrier frequency of 4.43 MHz. Here the signal level is raised
from a few tens of millivolts to several volts before being fed to the demodulator.
In modern PAL receivers a delay line is incorporated which stores one line of com-
posite chroma information for the period of one picture line (64 fJ.s). This system
helps to take advantage of the ability of the PAL colour TV system to minimise
the effects of changes in the chroma signal in the transmitter-receiver chain. It is
done by averaging the received signal over a period of two lines: i.e. it averages the
stored or delayed line with the current line. As a by-product of this process, separa-
tion of the U component from the V component takes place, which helps the pro-
cess of demodulation. Following the (R' - y/) and (B' - y/) demodulators, the
(G/ - y/) is produced by adding proportions of (R' - y/) and (B' - V'), and all
three are then amplified and added to the luminallce signal. From the matrix the
outputs are in terms of the primary colours, R/, G/, B/, as generated by the camera
amplifiers at the transmitter.
In the auxiliary decoder circuits the colour subcarrier is regenerated; its frequency
and phase are identical with those at the transmitter. The receiver crystal oscillator
is synchronised with the transmitted colour burst. The extraction of the subcarrier
The Essential Features of a TV Receiver 41

burst from the composite chroma is achieved by a gated amplifier, which is nor-
mally switched on only during the 'back porch' period. The necessary gating is done
by means of suitably delayed line flyback pulses.
The regenerated subcarrier is then fed to the two synchronous demodulators;
the (B' - V') subcarrier is fed via a 90° phase shift network, while the (R' - Y')
feed passes through the PAL switch.
The PAL switch operates line by line, but it must be synchronised with the
transmitter PAL switch. A suitable synchronising signal, called the ident, is derived
from the changing, or swinging, phase of the burst.
The amplitude of the colour burst is also detected, and is used to control the
gain of one of the chroma amplifiers in the manner similar to the a.g.c. in the i.f.
strip; this system is called automatic chroma control, or a.c.c.
During monochrome transmissions the chroma path is shut down; this avoids
spurious colours and 'coloured noise' appearing on the screen. A circuit known as
the colour killer detects, either directly or indirectly, the presence of the burst.
Whenever the burst is absent, the colour decoder is 'killed'.

4.6 Sound Stages

The beat frequency between the vision and sound Lf. gives rise to a 6 MHz sound
signal at intercarrier frequency. The audio information frequency modulates the
sound carrier, but because of being mixed with the video signal, there may addi-
tionally be a high percentage of amplitude modulation. The latter is removed in
limiting amplifiers which are usually associated with the 6 MHz tuned circuits or
ceramic filters. The demodulation of the f.m. signal follows established techniques
of detection-e.g. quadrature detection either inside an I.e. or in a ratio detector in
which two diodes are used. The resultant audio signal drives an amplifier and, in
turn, the loudspeaker. Integrated circuits are very common here, because of the
identical circuit requirements found in both radio and TV receivers.

4.7 The Synchronising Pulse Separator

By detecting and amplifying the various pulses present in the video waveform, this
circuit provides synchronising pulses to the line and the field time bases. In principle,
the circuit is a level detector, which conducts only when the video waveform has
reached a predetermined level, namely that of the pulses themselves.

4.8 The Line Time Base

The chief purpose of this circuit is to produce the necessary horizontal scanning of
the screen by an electron beam. This deflection must be in synchronism with the
transmitter time base, which is governed by the synchronising pulses. In conjunc-
tion with the field time base, the operation of the line scan circuit produces a dis-
play on the screen called raster which gives the picture its line structure.
42 Television Principles and Practice

The frequency of the line time base oscillator is controlled by means of a signal
which is derived from the comparison of the frequency and phase of the synchronis-
ing pulses with those of the flyback pulses.
The output from the oscillator is used to drive the line output stage, which, via a
transformer, provides the current waveform for the scan coils. When the current
changes rapidly, a large back-e.m.f. is induced in the coils and in the associated out-
put transformer, and is used to generate an e.h.t. (extra high tension). Its value
could be 25 kV in colour receivers, and is required for the final anode supply of the
picture tube. In addition, a somewhat lower voltage level, of between 600 V and
5 kV, is also required for other electrodes of the tube.
Modern line output stages can provide a number of power supplies to a TV
receiver, both high voltage (h.t.), say between 100 and 200 V, and low voltage
(Lt.) of between 15 and 40 V.
A variety of flyback pulses are obtained from the output stage and are used to
provIde a number of services in other parts of the receiver, some of which have
already been mentioned.
In colour receivers, correction or convergence circuits are necessary to keep the
three beams close together. Suitable waveforms are derived from the line output
stage and fed to a separate convergence coil assembly mounted on the neck of the
tube.

4.9 The Field Time Base

The field time base provides the vertical deflection for the scanning beam. The
circuit consists of a synchronised oscillator, followed by an amplifier and the out-
put stage feeding the scan coils mounted on the neck of the tube. As the electrical
requirements of both the field and line time bases differ, the practical circuits of
each also differ from one another. Synchronisation of the field time base is
achieved by producing a single trigger signal for the oscillator from the transmitted
field synchronising pulse sequence. The output stage requires a relatively complex
arrangement of linearity correction circuitry.
In colour receivers, convergence coils are also provided which are fed with suit-
ably shaped waveforms derived from the field time base output stage.

4.10 The Picture Tube

The tube provides the display of the transmitted picture. The cathode ray tube
resembles, in many respects, a thermionic valve. Electrons are emitted from a
heated cathode, and, by means of strong electric fields produced by other elec-
trodes, a narrow beam of electrons is focused on to the face of the tube. The inside
of the screen is coated with a material known as a phosphor, which emits light
when the electron beam strikes it. The colour of the emitted light depends upon
the composition of the phosphor; white is needed for monochrome tubes, while
red, green and blue are required for colour TV tubes.
The Essential Features of a TV Receiver 43

The brightness of the display is controlled by the amount of current reaching

the screen, which, in turn, is affected by the voltages applied to the various elec-
trodes of the tube. The potential difference between the cathode and the control
grid is the video signal which modulates the beam current; hence, the picture is
produced by the variation in the brightness level. The voltages applied to the re-
maining electrodes are normally constant, although they may be preset as part of
the initial installation procedure.
The tube in a colour receiver has three sets of electrodes-each associated with
its own colour signal (red, green or blue). The screen has three phosphor materials
deposited in the form of either dots or stripes. The spacing between them is so
close that, from a normal viewing distance, the emitted colours mix into a resultant
depending upon the amounts of current associated with each of the three beams. It
is necessary to ensure that the respective beams strike only their own phosphors.
This is done by means of a special mask close behind the screen which prevents the
'wrong' electron beam reaching any given phosphor. There are small magnets on the
neck of the tube, which also ensure the 'purity' of the displayed colour. Another
problem with colour tubes is a tendency for the three beams to separate as they
scan the screen both in the horizontal and in the vertical directions. If this occurs,
the beams strike phosphors which are so far apart that the impression of colour
merging is destroyed, and very unpleasant coloured fringes result. A convergence
coil and magnet assembly is mounted on the neck of the tube to allow adjustments
which reduce the above-mentioned effect to a minimum.
The overall dimensions of the tube dictate the size of a TV receiver. Monochrome
receivers use a very wide angle of deflection approaching 120°; this makes the flared
part of the tube relatively short.
Some colour tubes use 1 10° deflection, while others use only 90°. The wider
angle requires a higher power consumption from the already hard-working line out-
put stage. Also, the picture shape tends to depart from a rectangle and develop
bowed horizontal and vertical edges. Such non-linearity (pincushion effect) is
relatively easy to correct in monochrome tubes, but correction is more involved in
colour receivers.
A colour tube has a coil mounted round its flared part, which is required to
remove any residual magnetism in the surrounding external and internal steel parts
of the tube. This is called a degaussing coil, and is fed from the mains supply via a
special network which automatically reduces the operating current after a very
short period of time.

4.11 Power Supplies

A TV receiver requires several d.c. power supplies, the value of the voltage depend-
ing on the magnitude of the signal being handled. For example, the tuner and the
Lf. stages need a relatively low voltage, while the video and the line output stages
operate at a relatively high voltage.
In a colour receiver a number of circuits require fully stabilised supplies. Often
44 Television Principles and Practice

elaborate regulator arrangements are employed in addition to the more conventional

rectifier and smoothing circuits.
An all-transistor monochrome receiver also has stabilised power supplies, pri-
marily to feed the line time base circuits. A portable, battery-operated receiver
must have a supply regulator, since the battery voltage varies over a wide range.
There is also a tendency to either eliminate or reduce the size of the mains trans-
former in TV receivers; this is achieved either by using the line output transformer
to supply most of the power requirements of the TV set or by employing a low
frequency to high frequency converter. In the latter the mains 50 Hz supply is
rectified and smoothed to be fed to a switching circuit which can change the d.c.
back into a.c.-for example, at line time base frequency. The resultant is easily
transformed to any required voltage level and is then rectified and smoothed
again.
All receivers whose power supplies are derived from the mains, especially where
there is no isolating transformer, could present a safety risk to both the viewer and
the servicing technician. Therefore the design and the components used must, by
law, comply with a number of standard recommendations.
5 Tuners and I.F. Amplifiers
5.1 High Frequency Performance of Electronic Amplifiers
The frequency of signals processed by the tuner can extend from [Link]. to [Link].-
that is, from a few tens of MHz to several hundred MHz. A high frequency or radio
frequency (r.f.) signal is converted in the tuner unit to an intermediate frequency
(Lf.). The value of the iJ. is also relatively high-in the region of tens of MHz.
Before we consider any detailed circuits, it is necessary to appreciate the effect of
high frequencies on the performance of electronic amplifiers.
It is inevitable that every amplifier has limitations on the highest signal frequency
which it can amplify. These limits are determined by two factors-namely the high
frequency capability of the transistor (or I.C., or valve) and the overall design of the
circuit. Generally, the amplifying device is chosen so that it has a gain which is
maintained throughout the required frequency range. A replacement transistor or
valve must therefore have a similar specification to that of the original device. When
the circuit design is a limiting factor, the values of certain components, their con-
struction and physical layout are usually critical. For example, a pair of conductors
separated by an insulator behave like a capacitor; therefore a resistor or a coil will
have what is known as stray capacitance. This capacitance exists between the com-
ponent in question and adjacent components or wiring, including the chassis
connections. The value of stray capacitance is usually in the region of a few pico-
farads, but its effects at high frequencies can be considerable because of the result-
ant low reactance. This can cause the signal to be shunted out of the required path,
or it may couple the signal to a circuit where the link is not intended.
Similarly, a length of wire or such components as capacitors and resistors will have
inductance in the order of a few microhenrys. This is again important at very high
frequencies, where the 'stray' inductance may, perhaps combined with the stray
capacitance, produce unwanted tuned (resonant) circuits.
The presence of such 'strayst is taken into consideration in the circuit design and
construction. When testing or repairing high frequency circuits, it is important to
appreciate the above factors. The connection of test equipment, probes and leads
can easily upset the performance of the circuit. Component replacements must be
of a similar type and fitted in a similar manner to the original (that is, length of
connections, pOSition with respect to other components, etc.).
It is unlikely that a particular circuit will have to amplify signals over a very
wide frequency range; this applies especially to tuners and Lf. amplifiers. In fact,
4S
46 Television Principles and Practice

the range of signals to be amplified is limited by the bandwidth of the amplifier.

Any frequencies falling outside this range are rejected-that is, the circuit must
offer satisfactory selectivity as well as gain. The basis of such an amplifier is a tuned
circuit, which consists of either a series or a parallel combination of a capacitor
and a coil. Often the stray capacitance and inductance of wiring are incorporated
into the resonant circuit. The response of a simple tuned circuit is centred around
the frequency given by
1
resonant frequency, 10 =- - - -
21Ty(LC)
If the inductance L is in [Link] and the capacitance C is in [Link], then the frequency is in
MHz.

S.2 The U.H.F. Tuner

The tuner is used firstly to select, or tune to, the desired TV channel, and secondly
to convert the r.f. signal to an i.f. Signal. A limited amount of amplification is also
produced in the tuner, although for practical reasons most of the receiver gain is
obtained from the i.f. section. The conversion of the r.f. vision carrier of any
channel into a single frequency simplifies the receiver design; since each i.£. ampli-
fier is tuned to a fixed frequency, no retuning is necessary when channels are
changed by the user. Additionally, since the ultra high frequency carrier is conver-
ted to a much lower frequency signal, the design of subsequent circuits is easier.
The block diagram in figure 5.1 shows the arrangement of circuits in a u.h.£.
tuner. The aerial isolating circuit is an important safety feature found in mains-
operated receivers; it isolates the aerial installation from the receiver chassis, which
could be 'live' (see chapter 14). Practical aerial isolating circuits are shown in figure
5.2; capacitors C 1 and C2 in diagram (a) and C 1 in diagram (b) offer a reactance of

u.h.f.
aerial r- tuning voltage
I
I
I
I~-------.-------~
I I

aerial input oscillator i.f.

isolating tuned r.f. amp. and tuned
circuit circuit circuit to Lt.
amplifiers
a.f.c.

L . - - - - - - p o w e r supply
'--_ _ _ _ _ _ _ _ _ a.g.c.

Figure S.l Block diagram of a u.h.f. tuner (tuning voltage is used in varicap
tuners)
Tuners and IF. Amplifiers 47

several MU at mains frequency. For example, a 470 pF capacitor has a reactance of

6.8 MU at 50 Hz, effectively reducing the danger of electric shock, but at the radio
frequencies involved its reactance is practically zero. The large-value resistor, R 1
[figure 5.2(a)] , prevents the accumulation of a static charge across C1 , which can

to tuner
[Link]
(not electrically

l
----------------------t-~ connected to
- - - - - - - -- any other
part of the

I'
circuit)
C1
~20-470p

", receiver
chassis
(a) (b)

Figure 5.2 Aerial isolating circuits

arise from atmospheric conditions. In many modern receivers the entire circuit is an
encapsulated unit, and should it develop a fault, the whole assembly would have
to be replaced. For safety reasons the type of capacitors used in it have to conform
to strict electrical specifications; therefore the isolating circuit is not a repairable
item.
Tuned circuits used' at u.h.f. are different in appearance from their v.h.f. counter-
parts. At ultra high frequencies it is difficult to accommodate the inevitable stray
capacitance and inductance associated with conventional coils and their inter-
connections. Instead, the coils are replaced by straight lengths of a heavy gauge
conductor, one end of which is usually joined to chassis, the other end connecting
to a tuning capacitor or perhaps to one of the transistors. These conductors are
called transmission lines or Lecher lines, which, in combination with the tuning
capacitor, resonate at the required frequency (their behaviour is somewhat similar
to that of the u.h.f. aerial elements described in chapter 13). The tuner unit con-
tains individual compartments which house the Lecher lines together with the
components associated with a particular section of the arrangement. The walls of
these compartments form a part of the tuned circuit, and they also prevent un-
wanted coupling between individual sections. Where coupling is needed, say between
the r J. amplifier and the mixer, a small slot is cut in the dividing wall to act as a
coupling capacitor. Alternatively, two or more lines, which could be of different
length, are placed side by side and the arrangement behaves like a bandpass filter.
The positioning of components and the lengths of interconnections inside the
tuner are very critical. Decoupling capacitors are of the feed through type in the
walls of the tuner unit, which eliminates undesirable connecting leads necessary
with conventional capacitors (see C6 , Cl2 , etc., in figure 5.4).
The tuning on u.h.f. is continuous over the entire channel range. Since a number
48 Television Principles and Practice

of tuned circuits have to be adjusted together, a capacitor gang is used. The unit can
be pre tuned as required and channel selection is by means of push buttons. The
push buttons actuate a mechanical lever , and the movement of the capacitor rotor
is then governed by the position of adjustable stops which are preset during the
initial tuning procedure. This type of tuner tends to be bulky, and the mechanical
linkage can be troublesome.

5.3 The Varicap Diode Tuner

A varicap diode, also known as a varactor diode, is a silicon diode which is used in
its reverse biased mode-the reverse bias varies its capacitance. A reverse biased p-n
junction develops a depletion layer between the p- and n-regions, which acts as the
dielectric of a capacitor, the p- and n-regions forming the plates of the capacitor.
Since the width of this layer increases with the increasing voltage, the capacitance
decreases, and vice versa. The available capacitance is small, but at v.h.f. and u.h.f.
only a few pF are needed for tuning purposes.
The reverse bias potential to the diodes in the tuner is the tuning voltage; it is
varied between apprOximately 0.5 V (lower channel numbers) and 28 V (upper
channel numbers). The supply rail which provides this potential must be stabilised.

from receiver
power supplies

Figure 5.3 Varicap tuner channel selector

Any voltage fluctuations would cause a severe tuning drift, or even an inability to
tune the channels at the extremes of their entire range. Supply voltage stabilisation
is discussed fully in chapter 14. The arrangement shown in figure 5.3 uses an inte-
grated circuit stabiliser whose output is decoupled by C1 • The resultant stabilised
voltage is applied across the tuning potentiometers R 1 , R z , etc. (depending on the
number of available channel selectors); associated switches S1, S2, etc., are opera-
ted by their respective push buttons. The channel selector circuit is fixed to the
cabinet, the tuner itself being placed at a point close to the Lf. board.
A complete circuit of a u.h.f. varicap diode tuner is shown in figure 5.4. The
radio frequency amplifier uses two transistors-TR1 and TR2; the biasing of TR1
is governed by the a.g.c. control voltage. Other tuner designs use only one r .f.
 3 4 tuning 5 6 Lt.
h. t. voltage h. t. TP 1 1 ~ output
C'2 C'4 and a. f.c. C20
,-r------- --r---II--I P-r---jI ~ 1-------,---111
I \ I I I ,
I
I'
I I
I
I
\
I
I
I
I
: Lg
I .
C27 1

i I I I I
I \
~
iI :I I
I I I I I I
I ~ __ J I I I C I
I I I 26_ R'4j5\ ciT
I I ",C'6 C,a,. ,.
I I,. I ~
I I c28 T
I I
I I
L4 Ra I Rg I

L3 TR2 L5
: , I,
I I
I
,
I
c,5111-----f--1 Ic17 ~
rlt I C~:
I,
X4
X2
'ill: Xl
Xll
:
I
:
)
X5 'X6 X6 X9 R.. tuner I
screening
R3
C23
0 I
I
I
case •

I
I
01 02
I
I
-
~ I
I
03 I . 04
... ,.J. I I
I
.,.... I
I I
I I
._4----+ - - --- ---- -----4-----_._-- _--+- ___ ---.J•

Figure 5.4 U.H.F. varicap diode tuner (Rank)

50 Television Principles and Practice

transistor whose gain may be controlled either automatically or manually (local/

distant reception switch). At the input to TRI a ftlter consisting of L 1 , C2 , C3
rejects frequencies in the v.h.f. bands.
The. transistors in the tuner are connected in common base mode-the input
signal being applied to the emitter and the output taken from the collector, the base
being decoupled to chassis. This method of connection offers a better high frequency
performance than that of a common emitter amplifier. The base of TRI is decoupled
via Cs , while the d.c. connections to the transistor are provided by the emitter resis-
tor R2 , the a.g.c. line via R 3 to the base, and the r .f. choke L2 to the collector. The
signal from the collector is coupled to TR2 via C7 and the bandpass tuned circuit
formed by the Lecher lines Xl, X2, X3 and varicap diode Dl. The d.c. tuning volt-
age from the channel selector (figure 5.3) is fed to the diode via R4 and X2. The
circuit of the second r.f. amplifier is similar to that of TRI ; from TR2 the signal is
coupled (through the tuned circuits X4, D2, X5, X6, X8, D3 and X9) to the mixer
oscillator stage TR3. U.H.F. tuners use one transistor to perform both mixer and
oscillator functions; the oscillator tuned circuit consists of XIO, XII, varicap diode
D4 and trimmer Tel. Feedback from the output is via X9 to the emitter of TR3,
where the oscillator signal is added to the d. signal. The mixer produces a number
of beat frequencies, out of which the desired Lf. is picked out by LlO and external
capacitors, while L9 is a choke rejecting the high oscillator frequency. The auto-
matic frequency control (a./. c.) voltage can be applied to the common tuning line
at a point external to the tuner.
There are tuners which use a diode mixer stage. In these designs a transistor
oscillator circuit is followed by a u.h.f. diode, which is sometimes described as the
first i./. detector. A diode mixer generates less electrical noise than a transistor
stage and it produces a better quality output.
Where a tuning capacitor gang is used instead of a varicap diodes,_ the basic cir-
cuit arrangement is similar to the diagram in figure 5.4. Individual sections of the
capacitor are then connected to the respective Lecher bars. If a.f.c. is used in a
'mechanical' tuner, the control voltage is applied separately to a single varicap
diode incorporated in the oscillator tuned circuit. A more detailp,o discussion of
a.f.c. appears later in this chapter.

S.4 V.H.F. and Integrated V.H.F./U.H.F. Tuners

In the UK the demand for v.h.f. tuners is restricted to the receivers which operate
from special signal distribution networks and to receivers which were designed for
the 405-line standard. In many other countries v.h.f. transmissions are used exten-
sively in addition to u.h.f. programmes. Instead of two separate units, a combined
v.h.f./u.h.f. tuner, called an integrated tuner, is used in modern receivers.
V.H.F. tuners of early design used fixed coils, each being pretuned to a given
channel number. A suitable set of coils for each bandpass circUit, including the
oscillator, was carried on a rotary switch assembly (turret tuners, wafer tuners).
Final trimming (fine tuning) was then performed by adjusting the oscillator fre-
 Tuners and I.F. Amplifiers 51

u.h.f. aerial

u.h.f. u.h.f.
input r.f. amp. bandpass u.h.f.
circuit circuit mixer/oscillator

u.h.f. i.f.
v.h.f. aerial preamp. and
v.h.f. mixer i.f.
L.....---,-r---' output

52
v.h.f. v.h.f.
input r.f. amp. bandpass
circuit circuit

v.h.f.
oscillator
(a)

u.h.f. aerial

u.h.f. u.h.f. u.h. f. mixer/oscillator

input r.f. amp. bandpass
circuit circuit v.h.f. mixer i.f.
' - - - - - - r - - - - - ' output
v.h.f. aerial
system
switch

v.h.f. v.h.f.
input bandpass v.h.f.
circuit circuit oscillator

(b)

Figure S.S Block diagrams of integrated tuners: (a) u.h.f. and v.h.f. signal paths
are separate; (b) compact tuner design-circuit separation kept to a minimum

quency for best reproduction of picture and sound. Such arrangements were bulky
and the large number of moving contacts were a source of trouble.
Modern v.h.f. tuners are continuously tunable over the entire range of channels
in Bands I and III. Because of the frequency gap caused by Band II (f.m. radio) a
selector switch is normally used offering either Band I or Band III.
52 Television Principles and Practice

Integrated tuners follow one of the two basic arrangements shown in figure 5.5.
In block diagram (a) the u.h.f. and v.h.f. signal paths are separated by the function
switches SI and S2 (the system is not applicable to the British 405/625 dual stan-
dard receivers with their different i.f. on each standard). The final transistor in the
tuner acts as an i.f. preamplifier when switched to u.h.f. or as a mixer on v.h.f. The
power supply to the sections not used at any time is also disconnected by a func-
tion switch. Figure 5.5{b) shows a more economical arrangement; the transistors
perform dual functions depending on the position of the system switch. In the
v.h.f. mode an additional oscillator transistor is required, followed by a v.h.f.

to collector
of r.t. amp.

tuning voltage

Open
on
Band I

,a)

to collector
of r.t. amp.

Open on
Band I

~ S
tuning
voltage
from
receiver
power
supplies

(b)

Figure 5.6 Examples of v.h.f. band switching: (a) ordinary shorting switch;
(b) v.h.f. tuned circuit as in (a), but the switch is replaced by diode D2 with its own
control circuit
Tuners and IF. Amplifiers 53

mixer; on u.h.f. the latter serves as a mixer/oscillator in the manner described in

section 5.3.
The tuned circuits used for v.h.f. reception utilise conventional coil design. The
value of inductance needed at frequencies in Band I is higher than that for Band
III, so that the inductors are tapped via the Band I/I11 changeover switch as shown
in figure 5.6(a). Lecher lines are necessary for u.h.f. reception, but when the tuner
is switched to v.h.f., they simply act as interconnectors which couple the signal
from the v.h.f. coils to the amplifying transistors and to the tuning devices. Either
varicap diodes or ganged capacitors can be used, their functions being shared on
both u.h.f. and v.h.f.
Very low amplitude r .f. signals can be seriously affected by bad contacts of the
system switch. To overcome this, diode switching can be employed; mechanical
contacts are then used to apply a control bias voltage to a diode. This type of cir-
cirlt is shown in figure 5.6(b). The two circuits (a) and (b) are identical as far as r.f.
signals are concerned, but in (b) the switching diode 02 together with the switch S
replace Sl shown in diagram (a). When switch S-diagram (b)-is open, 02 is not
forward biased, so that it behaves as an open circuited switch, and Ll is in series
with L2 . (The magnitude of the signal across L2 is by itself insufficient to turn the
diode on.) Closing switch S applies a positive potential to the anode of 02 through
the current limiting resistor R 2 ; 02 is now biased ON, which, in turn, acts as an r.f.
short across L2 (via the d.c. blocking capacitor C2). Similar switching diodes can
be used elsewhere in the tuner to steer the signal between parts of the circuit or to
provide the switching of complete sections of an integrated tuner.

5.5 Touch Tuners

Touch tuners eliminate the need for mechanical methods of channel changing.
Since it is relatively easy to control the tuning voltage of varicap tuners, they lend
themselves to this form of 'automation'. The control circuit may contain either
individual transistors or integrated circuits.

stabilised I.t.

control voltage
pulse latching tuning
to varicap
generator circuit potentiometer
tuner
touch
I

t
disabling pulse
to/from other
channels

Figure 5.7 Principle of touch tuning. When energised, the circuit transfers the
stabilised l.t. to the appropriate channel potentiometer
 r---_-- h.t.

. - -__ other channels

touch
contacts stabilised I.t.

tuning voltage
f-+"-~----to varicap diodes

to/from other channel

selector circuits

(a)
240 V a.c.

voltage stabiliser
I.C.

8_
d
co
other oco other
channels a: } channels
I::::
w other
~--~--~---~~15
channels
r---i-16

I.C.l

tuning
voltage

(b)

Figure 5.8 Examples of touch tuning circuits: (a) using discrete components-
compare with figure 5.7 (Rank); (b) using an I.C. (GEC). Connections marked
'other channels' are made to identical circuits associated with the remaining touch
contacts and tuning potentiometers
Tuners and IF. Amplifiers 55

The principle of a transistorised control circuit of touch tuning is demonstrated

in figure 5.7. The base bias circuit of the sensing transistor TRI is broken by the
touch contacts. When the resistance of the skin completes this circuit, TRI turns
on and its collector voltage falls. The resultant change in output is fed to a pulse
generator; this produces a 'clean' switching pulse, which is used in tum to apply a
d.c. potential to the tuning potentiometer. A latching circuit is needed to ensure
that the tuning voltage remains after the removal of the fmger from the touch con-
tacts. At the same time, a 'disabling' pulse is sent to the other channel control
circuits to ensure that they remain switched off; this action also 'unlatches' the
unwanted channel. Often arrangements are made to 'bias' the selector in favour of
one particular channel whenever the receiver is first switched on.
A similar method of touch sensing and subsequent Switching can be built into
an I.C. Alternatively, the touchplate forms part of a capacitor which couples the
signal induced by the fmger to a suitable integrated circuit.
In many cases lamps or miniature neon tubes are also connected to indicate the
actual channel selection. Where neon tubes are used, they can be part of the switch-
ing circuit, as shown in figure 5.8(a). Touching the contacts biases both TRI and
TR2 ON; this, in tum, connects the left-hand plate of the charged capacitor C2 to
chassis. The resultant momentary high potential difference across the neon causes it
to strike, which, in tum, switches the isolating diode Dl ON. The conducting diode
transfers the stabilised Lt. supply to the channel selector potentiometer Rs. The
tuning voltage is passed on to the tuner via another isolating diode D2 (D3 com-
pensates for the effects of temperature changes on D2). The entire circuit shown
in figure 5.8(a) is repeated for each set of touch contacts, as indicated by the
connections to the identical points in the other channel selectors. When another set
of touch contacts is used, the corresponding neon will strike, momentarily increas-
ing the current through R 7 . Consequently, the voltage across the previously opera-
ted tube falls and the lamp switches off; Dl and D2 remove their tuning voltage
and the new channel is selected.
Figure 5.8(b) shows an arrangement using an l.c. Bridging the touch contacts
connects the neon to a large potential difference-positive h.t. on one side and
negative rectified a.c. on the other. When the tube strikes, the resultant pulse is
communicated to the I.C. (pin 15) whose internal circuitry transfers the 33 V tuning
supply voltage (pin 1) to the required channel selector potentiometer (for example,
R7 at pin 2). The output to the tuner is via the isolating diode D2 in the manner
shown in figure 5.8(a), described above. Similarly, tuning to another channel will
extinguish the previous neon, and a new 'switch-on' pulse will be initiated.

5.6 Remote Control

Remote control is a natural progression from touch tuning-again a suitable pulse is
needed to initiate channel switching. A control pulse is sent by a hand-held trans-
mitter to a 'pick-up' built into the TV receiver. The transmitted signal is in the form
of ultrasonic vibrations which are generated by a special transducer driven by an
56 Television Principles and Practice

electronic oscillator. The ultrasonic frequency lies in the range 30-45 kHz; one
controller can employ a number of frequencies not only for channel switching, but
also to control the brightness, volume, etc. If only one frequency is used, each
successive burst of signal from the transmitter will cause sequential channel change.
The diagram in figure 5.9(a) shows the circuit of a battery-operated hand-held
ultrasonic transmitter. A Hartley oscillator is used, its preset frequency being
governed by Ll , and the circuit is activated by the control switch Sl. The trans-
ducer requires a high amplitude a.c. signal superimposed on a polarising d.c. voltage.
Therefore the oscillator output is stepped up to a voltage in the range 150-300 V;
the a.c. is applied to the transducer via C6 , while the d.c. polarising potential is
derived from the voltage doubler, Dl, D2, C4 , C s .
The 'pick-up' transducer in the receiver is similar to that in the transmitter, but

- 1 51
Cl

~~--------------~----*---------~--4-----~

(a)

ultrasonic
transducer
to sound
pulse pulse
Schmitt section
amplifier detector length steering
trigger
discriminator circuit

varicap diode
tuning voltage

touch tUlling contacts

(b)

Figure 5.9 Remote tuning control (Rank): (a) transmitter unit; (b) block diagram
of receiver unit. The receiver unit provides both channel selection and sound muting
 Tuners and I.F. Amplifiers 57

the subsequent decoding circuit is more involved. A simplified block diagram of a

decoder is shown in figure 5.9(b). A multistage amplifier feeds a detector, where
the high frequency signal bursts are converted into individual pulses. A pulse shap-
ing circuit (Schmitt trigger) applies 'clean' pulses to the shift register I.C. Whenever
the I.C. accepts a pulse, the internal circuitry connects its output pin to the next
tuning potentiometer in the sequence. The remote control system can be switched
out and the tuner reverts to manual operation (touch tuning).
The circuit in figure 5.9(b) has an additional feature-namely that of sound
muting. The viewer can transmit a longer pulse; this is noticed by the pulse length
discriminator and is steered to the audio section. Shorter pulses are, of course, sent
to the channel selector i.c.
The remote control system uses mechanical signals instead of electromagnetic
waves; therefore transmission-reception at these ultrasonic frequencies follows a
relatively straight line. For reliable operation the transmitting transducer should
face the receiving transducer.

5.7 Tuner Servicing

The tuner unit is perhaps the most reliable part of the receiver, since it not only
handles very small signals, but also dissipates very little power. Sometimes the tuner
circuit diagram is not included in the service manual, because the manufacturers
recommend that the unit be returned to them for servicing and realignment. Work-
shop repairs can easily disturb the critical layout of the circuit and, since replace-
ment components can have different characteristics from the original parts,
replacements may result in a degree of misalignment. Accurate tuner realignment
is only possible with the aid of specialised equipment which is outside the scope of
the average workshop.
BefGre the tuner unit itself becomes suspect, a number of points have to be
checked first, because they can produce fault symptoms similar to those of a tuner
failure. The following list of checks can be derived from the block diagram in
figure 4.1: efficient aerial installation (especially plugs and sockets), d.c. power
supplies to the tuner, a.f.c., a.g.c., Lf. amplifiers, etc. In addition, it may be neces-
sary to inspect mechanical linkages in the tuner channel selector and the condition
of contacts and switches.
D.C. readings can help to isolate the faulty stage, but it must be realised that any
instrument connected inside the tuner is likely to disturb the signal conditions.
Examples of possible tuner faults: very noisy picture-suspect r.f. amplifier(s);
no sound or picture, usually with some noise present-oscillator circuit; difficulties
in tuning on all or certain channels only-oscillator circuit, misalignment, tuning
voltage drift in varicap tuners (the latter normally due to circuits external to the
tuner).
Should any components be replaced, then their positioning, the length of
connections, etc., must follow the makers' original design.
58 Television Principles and Practice
S.8 Frequency Response of I.F. Amplifiers
I. F. amplifiers are responsible for most of the signal gain in the receiver and they
also affect the overall quality of the displayed picture (and sound). This is done
by rejecting unwanted channels and maintaining a correct balance between the
signals which make up the composite video waveform.
The performance of an i.f. strip is represented by means of a frequency response
graph. The graph shows the change in the output voltage, expressed in dB, to a
base of the intermediate frequency (see figure 5.10). This range offrequencies
arises from both the video signal bandwidth and the presence of sound and chromin·
anceinformatton.

N
::t: N
~ ::t:
~ N
0 ::t:
~

-6 "'!
:;
"C
-12 c:
'5l"
iD
~ -18 Qj
c:
II! N
c:

..
c: ::t: IV
0 -24 ~ .J:
Co
~
"'! "c:
"C

..,'"> -30 M IV
"C
c: c:
.~ '"5l
IV
~ -36 .;;
"C
:6' c:
.-42
IV
'g"

32 33 34 35 36 37 38 39 40 41 42
intermediate frequency (MHz)

Figure S.10 Typical i.f. response curve of a TV receiver

The recommended standard vision i.f. in the UK is 39.5 MHz; its choice is based
on signal bandwidth, frequency allocation to other channels, etc. The oscillator in
the tuner must operate at a frequency which is 39.5 MHz above the vision carrier
of the required channel. For example, if the receiver is tuned to channel 33, whose
allocated vision carrier is 567.25 MHz, then the oscillator frequency is 567.25 +
39.5 =606.75 MHz. Channel 33 sound carrier is 6 MHz above the vision carrier at
573.25 MHz, and the chrominance information is centred 4.43 MHz above the
vision carrier at 571.68 MHz.
The mixer in the tuner produces a number of signals whose frequencies corres J

pond to the differences between the oscillator frequency and the respective
carriers. The resultant intermediate frequencies are fed to the Lf. strip as follows:
Tuners and LF. Amplifiers 59

vision i.t, 39.5 MHz; sound i.t, 606.75 - 573.25 = 33.5 MHz; chrominance i.f,
606.75 - 571.68 = 35.07 MHz. Although these figures are for channel 33, exactly
the same values would be obtained for any other channel in the UK.
Signal frequencies corresponding to the carriers of the unwanted adjacent channel
numbers can also reach the mixer; this is caused by the lack of perfect selectivity
in the tuner r.f.. amplifiers. Adjacent channel numbers are not used by the same
transmitter in order to prevent interference, but they may be received in an area
which is within the coverage of a number of transmitters.
In the above example adjacent channels are: channel 32 (vision 559.25 MHz,
sound 565.25 MHz) and channel 34 (vision 575.25 MHz, sound 581.25 MHz). The
mixer/oscillator tuned to channel 33 will produce the following additional Lf.s:
channel 32 vision: 606.75 - 559.25 = 47.5 MHz;
channel 32 sound: 606.75 - 565.25 =41.5 MHz;
channel 34 vision: 606.75 - 575.25 =31.5 MHz;
channel 34 sound: 606.75 - 581.25 = 25.5 MHz.
It can be seen from the above results that channel 32 sound and channel 34 vision
Lf.s are separated by only 2 MHz from the required channel vision and sound,
respectively. This situation is made worse by the presence of the sound carrier fre-
quency of VHF channell (405-line system), which is also 41.5 MHz. The unwanted
signals can interfere with the reception of the desired channel either in the form of
patterning on the screen or as 'noisy' sound. The extreme frequencies in the above
list (47.5 MHz and 25.5 MHz) are considered to be too far away from the wanted
spectrum to cause any problems. The same results for adjacent channel iJ.s would
be obtained if other groupings were considered. This is, of course, due to the fixed
spacings between the frequency allocations for the various channels and their
carriers.
The frequency response of the Lf. strip must be arranged so as to suppress the
unwanted signals of the adjacent sound (41.5 MHz) and adjacent vision (31.5 MHz).
A typical Lf. response graph is given in figure 5.10-it will be noted that it has a
'flat' centre portion and very steep sides. The region of the relatively constant out-
put is governed by the frequencies of the video bandwidth; the steep sides offer
the necessary selectivity, or rejection of unwanted signals. Comparison of the Lf.
response graph with that of the transmitted frequency spectrum (figure 2.8)
indicates a reversal in the relative positions of the sound and chroma with respect
to the vision carrier. This is, of course, due to the frequency subtraction in the
mixer stage of the tuner.
Let us consider the Lf. response to the desired frequencies. The output at
39.5 MHz is approximately one-half of the maximum possible value (half voltage
corresponds to -6 dB). The necessary reduction of the vision carrier amplitude is
caused by the vestigial sideband transmission described in chapter 2. As indicated
in figure 2.7, the low video frequencies, up to 1.25 MHz, appear in both the upper
and the lower sidebands. Therefore the energy transmitted in the frequency range
close to the vision carrier is twice the amount corresponding to the higher video
60 Television Principles and Practice

frequencies. The imbalance must be corrected in the receiver-partly in the Lf.

strip and partly in the video amplifiers (to be discussed in chapter 6).
The response to the chrominance information, centred around 35.07 MHz, is also
6 dB below the maximum. This arrangement helps to minimise the effects of the
4.43 MHz interference pattern on the screen. In colour receivers the chroma signal
is amplified separately in the decoder at 4.43 MHz; therefore the reduction at
35.07 MHz can be compensated for.
The response at the sound Lf. of 33.5 MHz is considerably depressed (between
20 and 30 dB) to prevent pattern interference on the picture owing to the vision-
sound and chrominance-sound beat frequencies. Again, the sound information is
amplified separately at 6 MHz, and the present rejection, provided it is not exces-
sive, does not affect the audio quality.

5.9 Practical I.F. Amplifier Circuits

The reader will recall that a resonant circuit can discriminate either in favour of or
against signals of a certain frequency. Therefore, if a tuned circuit is placed in the
collector of a transistor amplifier, a substantial gain will be obtained at the resonant
frequency. Unfortunately, the response of the resultant simple tuned circuit ampli-
fier cannot follow the required Lf. response of a TV receiver. It is not possible to
achieve the necessary bandwidth and good selectivity in the simple amplifier
arrangement.
A number of practical methods can be adopted to modify the circuit basic res-
ponse to suit the requirements. One type of arrangement uses coupled tuned cir-
cuits, shown in figure 5.11(a). Inductors Ll and L2 form a transformer, but the
primary and the secondary windings are independently tuned by capacitors C1
and C2 • The amount of magnetic coupling, or interaction, between the windings is
made variable by means of adjustable cores. The shape of the resultant frequency
response curve depends on the amount of coupling between the two tuned circuits.
(1) A single sharp peak is produced-as for a normal tuned circuit. This arises with
so-called loosely coupled circuits.
(2) Two peaks appear in the response curve with a dip between them (a 'double
hump' response). This is due to the so-called closely coupled circuits.
(3) Level response occurs over a reasonably wide frequency range owing to critically
coupled cii:cuits.
The three possibilities are represented by the response curves in figure 5.11 (b). Res-
ponse (3) above is of most interest here. Unfortunately, practical alignment of tuned
coupled circuits is difficult, because of the very critical interaction between the
circuits.
Another method of increasing the bandwidth is by means of staggered tuning.
This is represented in block diagram form in figure 5.12(a). Three iJ. amplifier
stages are shown to be tuned to slightly different frequencies within the required
passband. The resultant response curve, when expressed in dB, is the sum of the
Tuners and IF. Amplifiers 61

R,

L..-.._ _--+-_---..- to next

stage

signal -4I-~--t-t
input

(al

I~'
I (3) critical coupling -- desired
I
I
I
I
I
I (11 loose coupling --- insufficient
I bandwidth
I
/
/

frequency

(bl

Figure 5.11 Coupled tuned circuit amplifier: (a) circuit diagram; (b) response
curves produced by different tuning of L1 and L2

three individual curves. A relatively flat response with reasonable selectivity can be
achieved, as given in figure 5.12(b).
The most popular method employs broadband or wideband tuning in conjunction
with suitable response shaping wave traps. The Lf. amplifier in this arrangement also
uses a resonant circuit in its collector, often without any provision for tuning
adjustments. A damping resistor is connected across the circuit to damp, or to
62 Television Principles and Practice
i.f. strip

from
tuner
-I stage I

35.5 MHz
H stage II

37.0 MHz
H stagem

38.5 MHz
• to video
detector

(a)

intermediate frequency (M Hz)

(b)

Figure S.12 Principle of staggered tuning: (a) block diagram; (b) individual response
curves of each stage added to produce the resultant of the Lf. strip
flatten, the response by 'spoiling' the Q factor. This arrangement satisfies the need
for a considerable Lf. bandwidth, but its selectivity is very poor-especially where
tuning adjustments are not provided.
The required shaping of the response curve of broadband amplifiers is by means
of separate circuits which are made highly frequency selective. Traps, or fIlters, of
this kind are often grouped at th~ input to the Lf. section. Basically they consist of
Suitably connected series and parallel resonant circuits. A series tuned circuit has a
very low impedance at resonance, while a parallel circuit has a high impedance.
Should a parallel circuit be placed in series with the signal path, it attenuates all
information at its resonant frequency. A series tuned circuit placed in the same
position would enhance the signal. On the other hand, if a series resonant circuit is
connected across the signal path, its low resistance reduces the gain at that frequency;
a parallel circuit in that position tends to boost the resultant output.
Another form of frequency selective network consists of a series-parallel
combination which gives rise to two resonant frequencies-one of them enhances the
response and the other one reduces it.
Finally, a popular rejector circuit is the so-called bridged T network formed by
a combination of capacitors, resistors and inductors. Two types of these compon-
ents (e.g. capacitors and resistors) are connected to form a letter T; the upper arm
of the T is then bridged by the third type of component (in this case it would be
an inductor). This filter design offers a very high degree of attenuation especially
Tuners and I.F. Amplifiers 63

a.g.c.

Lf.
from
tuner

-...- -....-y----
I
low pass 41.5 MHz 39.5 MHz' 36.5 MHz
filter
31.5 MHz

Figure 5.13 Tuned circuit i.f. amplifier with response shaping fIlters (Decca)

useful in suppressing adjacent channel frequencies.

An example of an i.f. amplifier with its input selective network is shown in
figure 5.13; the resonant frequencies of each section are stated below their respec-
tive circuits. R 1 , C1 form a general low pass fIlter at the input; C2 , C3 , L 1 , R2
form a bridged T network (R 2 is the 'bridging' component); Cs , C6 , L6 produce a
series-parallel tuned circuit (series resonance occurs at 33.5 MHz as indicated, and
parallel resonance would then automatically take place at a higher frequency to lift
the response curve up). Inductor Ls is part of a parallel resonant circuit which is
tuned by the inevitable stray capacitance associated with the coil. A similar situa-
tion arises in the collector circuit of TRI, where the inductor L7 is also tuned by
the circuit stray capacitance in addition to the coupling capacitor C8 • The i.f.
amplifier itself, TRI, is tuned to 36.5 MHz, which corresponds nearly to the centre
of the 'flat' portion of the response curve (see figure 5.10).
Resistor R3 in the input circuit and Rs across the collector tuned circuit are
damping resistors; these prevent the possibility of self-oscillations (instability),
which may occur in high gain, frequency selective amplifiers. In some circuits
separate damping components are not needed, since the relatively low input resist-
ance of the following transistor stage provides a sufficient degree of damping.
Frequency selective circuits can also be connected in the feedback path of an
amplifier, either to enhance or to suppress particular frequencies.
64 Television Principles and Practice
- __- - h . t .

~ to next
....---~r---- stage
C6

C,
i.f. ~
input ------.1--+---'"'

Figure 5.14 Cascode Lf. amplifier: TRI (common emitter amplifier) connected
directly to TR2 (common base amp.-base decoupled via C2 ) (Rank)

Certain designs adopt values of trap frequencies slightly different from the
standard values quoted so far. However, some interaction obviously exists between
the fIlters, and it is the overall Lf. response which must be maintained.
The common emitter connection is usually used in i.f. stages (apart from inte-
grated circuits), but occasionally a cascade amplifying circuit is included. An
example of such an arrangement is shown in figure 5.14. Transistor TRI is a
common emitter amplifier whose collector feeds directly to the emitter of TR2.
Since the base of TR2 is fully decoupled via C2 , this transistor operates in the
common base mode; the output from the stage is at the collector of TR2. A cascode
amplifier offers high gain with good stability, and it is particularly useful in high
frequency applications.
A recent introduction which provides the necessary i.f. response shaping is the
surfacing acoustic wave filter. The device consists of two transducers-one at the
input, which converts the i.f. signal into mechanical (acoustic) vibrations, and
another one at the output, where the incoming surface waves are reconverted into
electrical signals. The term 'acoustic' denotes that the signal is propagated along
Tuners and IF. Amplifiers 65

the surface of the device in the form of mechanical waves. The construction of the
fIlter resembles that of an I.e.; the transducers are a comb-like, interlocking struc-
ture deposited on the surface of a small chip or substrate made of a special material.
Filtering action takes place because the amount of energy converted by the trans-
ducers depends on the frequency of the electrical signal. They can be designed
to develop a maximum output at, say, 37 MHz, and a series of minimum outputs at
31.5 MHz, 33.5 MHz and 41.5 MHz. An amplifier is needed to compensate for the
overall signal power losses in the device, but tuned circuits are no longer necessary
and there are no alignment problems.

00- 00-
~ ~
c:
.n; c:
.n;
'">
.,
CIl
. .,'"
II>
.>
'"
~ '"
~

i.f. (MHz) Lt. (MHz)

(a) (b)

-10
00- dB
~
c:
.~

..,>
II>

'"
~

33.5 35 39.5 35 39.5

i.f. (MHz) i.f. (MHz)
(e) (d)

Figure 5.15 Examples of incorrect Lf. response: ( a) insufficient response to the

vision carrier (weak contrast, possibly poor sync.), excessive response to chroma
signal (too much colour); (b) insufficient bandwidth (poor resolution of picture
detail), insufficient chroma (no colour); (c) excessive peak in Lf. response (possible
instability-ringing), excessive sound Lf. (beat patterns on the screen); (d) excessive
response to vision i.f. (strong contrast, smears around coarse picture detail, vision
buzz on sound), insufficient chroma (possibly loss of colour)
66 Television Principles and Practice

5.10 Alignment of I.F. Amplifiers

The purpose of alignment is to ensure that the frequency response of the Lf. strip
conforms to the makers' design. Incorrect alignment can lead to defects in the
picture (and in the sound!), as suggested by the examples of grossly misaligned
response curves sketched in figure 5.15.
I.F. alignment, or more often realignment, may be necessary for anyone of a
number of reasons, including the replacement of components (usually those associa-
ted with the tuned circuits).
The alignment procedure involves the application of a number of signals at fre-
quencies within the required range, say between 30 MHz and 42 MHz, to the input
of the i.f. section. The resultant amplified output is then monitored at some con-
venient point in the receiver. Ideally, a complete response curve should be obtained;
alternatively, the output only at certain specified frequencies can be checked against
makers' recommendations. In any case, manufacturers' recommended procedures
for alignment must be carefully considered.
The point at which the test signal from a signal generator is applied to the
receiver i.f. section may vary between differing designs. Frequently it is injected
at the beginning of the i.f. chain-that is, in the mixer circuit of the tuner (for
example, TPI in figure 5.4). In other cases the signal could be applied to the indivi-
dual i.f. stages in turn. The signal generator must always be properly terminated,
not only to preserve its correct calibration, but also to simulate the impedance levels
found in such circuits. A diagram of a simple iJ. injection probe is shown in figure

0.011' (2p when

coaxial applied to tuner TP)
cable - - - - - - - -
from _____ _ I -- to i.f. injection point
signal C,
generator

from
i.f.

(b)

Figure 5.16 Examples ofLf. alignment probes: (a) i.f. injection probe; (b) iJ.
demodulator probe (voltage doubler)-used when signal is to be taken directly from
one of i.f. stages
Tuners and I.F. Amplifiers 67

5.16(a); the value of the d.c. blocking capacitor C1 depends on the actual point of
application into the circuit.
The output voltage from the Lf. stage can be obtained most conveniently from
the vision detector load resistor or the video emitter follower (see chapter 6). The
magnitude of the rectified signal can be measured using a d.c. voltmeter (preferably
an electronic instrument or an oscilloscope). If necessary, a demodulator probe
can be made up, as shown in figure 5.16(b), and connected to the output of the
appropriate Lf. stage.
There are two basic methods of alignment: the wobbulator method and the
spot frequency method.

Wobbulator method
A wobbulator, also known as a sweep generator, is a signal generator whose output
frequency sweeps back and forth between certain limits. For vision iJ. alignment
these limits are approximately ±5 MHz about a 'centre' frequency. If the wobbulator
carrier frequency is set to 35 MHz, its output will swing from 30 MHz to 40 MHz.
This signal is injected to the Lf. stages and the output from the iJ. strip is displayed
on an oscilloscope ([Link].). The time base of the oscilloscope and the wobbulator
action are synchronised-that is, the minimum frequency (say 30 MHz) of the
generator coincides with the beginning of the [Link]. scan. When the beam reaches
the right-hand side of the screen, the wobbulator output corresponds to its maxi-
mum frequency (say 40 MHz). During the beam flyback period the output fre-
quency is reduced back to its minimum value. The display on the screen will then
be in the form of the Lf. response curve.
Some wobbulators utilise the [Link]. time base waveform (available at the ter-
minals marked-for example, X-OUT) to modulate the sweep. Alternatively, the
wobbulator has a built-in time base oscillator which can be used to control both the
generator frequency sweep and the oscilloscope horizontal deflection. In either case
the sweep rate must be kept relatively low, which means that the [Link]. time base
can be set to a sweep speed of about 10 ms/ cm. If the rate is too high, the results
are likely to be incorrect since the receiver circuits may not be able to respond to
such rapid changes in frequency.
A block diagram of the arrangement of instruments in the wobbulator method is
shown in figure 5.17. It is important to be able to identify the position of the
actual frequencies on the displayed response curve. This is done by injecting a
separate signal at a required frequency together with the wobbulator signal. The two
signals are mixed in the receiver and a marker 'pip' appears on the response curve in
its appropriate position. The separate signal source is called a marker generator,
which may either be built into the wobbulator or be connected externally via a low
value capacitor, C
The accuracy of the alignment depends upon the accuracy of the marker genera-
tor. The tuning of the Lf. coils is adjusted to produce a curve of the desired shape,
as indicated by the positioning of the identifying 'pips'. To prevent overloading the
68 Television Principles and Practice
TV receiver
marker C I
generator
31.5-41.5 MHz
I """--tunerfrom

I
I
wobbulator
30-42 MHz
injection
probe
I receiver
i. f. strip
video
detector
to video
amp.
L ____________ _
(D] C.R.O.

_--------TI y
input '--_ _...JX output
C.R.O. display

Figure 5.17 Block diagram ofLf. alignment setup. C.R.O. display can be either
positive-going (as shown) or negative-going, depending upon polarity of the video
detector

if. amplifiers, both the wobbulator and the marker generator outputs must be kept
to a low amplitude. These should be just sufficient to produce a reasonable display
on the C.R.O. screen. It is of interest to note that the oscilloscope display may
appear 'upside down', depending upon the polarity of the vision detector diode.
The advantages of the wobbulator method are: speed of adjustment, good
accuracy of alignment over the complete Lf. range and the ability to observe the
interaction between the traps, coupled tuned circuits, etc.

Spot frequency method

This method requires an accurate a.m. signal generator and a suitable output meter
connected in the manner explained in the general notes above. The applied signals
should include the trap, vision, chroma and midband frequencies. Adjustments are
then made until the correct response is achieved. Since an automatic frequency
sweep is not provided by this method, it is necessary to sweep manually through
the full frequency range. This provides a check that the response is symmetrical,
without undue dips or peaks, that the bandwidth is sufficient, etc. Because of the
uncertainty of the results obtained by this method, a number of manufacturers do
not recommend its use.

General precautions
Irrespective of the method used, the following should be carried out to ensure
correct alignment results.
 Tuners and iF. Amplifiers 69
(1) The automatic gain control (a.g,c.) must be rendered inoperative. Usually a fIxed
d.c. bias is applied to the a.g.c. line either from a suitable battery or from the
receiver h.t. via a resistor or a test link according to makers' instructions.
(2) Coaxial cable should be used for the test leads, and it should be kept as short as
possible. The input and output test leads must be kept well apart to prevent feed-
back between them.
(3) Special trimming tools should be used to adjust the cores in order to prevent
damage. These tools are made of non-magnetic material-usually plastics.
(4) Allow both the signal generator and the receiver to 'warm up' for at least 15
min before carrying out any adjustments.
(5) Select the correct position of the tuning core, since there are often two different
positions which appear to produce the required result. The accepted setting, unless
one is advised to the contrary, is with the core closer to the top of the coil former;
the alternative position could give incorrect bandwidth.
(6) The injected signal should be maintained at a reasonable level. Badly misaligned
circuits may require a large input initially, but as they are brought into alignment,
the signal strength must be reduced to prevent overloading.
(7) Ferrite cores should be fmally secured with a core locking compound, any
temporary components removed from the circuit and the operation restored to
normal.

Warning The danger of live chassis must not be overlooked, since a number of
instruments with earthed test leads will be used. The reader is advised to study the
discussion on this aspect in chapter 14.

5.11 Automatic Frequency Control (A.F.C.)

AF .C. maintains frequency stability of the oscillator in the tuner which, in turn,
ensures constant vision Lf. This is a highly desirable feature, since in a properly
aligned receiver the quality of reproduction will depend upon the stability of the
intermediate frequency itself. The performance of a colour receiver can be especially
affected by tuner oscillator drift. The chrominance signal is pOSitioned on the slope
of the Lf. response curve (fIgure 5.10) and it is also balanced with respect to the
response to the vision and sound carriers. Should the actual value of Lf. shift, the
respective carriers would effectively change in value. Consequently, this may result
either in the loss of colour or in severe interference patterns.
Automatic frequency control is obtained by the application of a d.c. feedback
signal to the tuner; the voltage fed back is proportional to drift in vision Lf. nearly,
the task of the a.f.c. circuit is similar to that of an f.m. demodulator, since, in
either case, a voltage proportional to frequency deviation is produced. A block
diagram of an a.f.c. arrangement is shown in fIgure 5.18, together with a typical
circuit diagram. Part of the output from the final Lf. stage is applied to the limiting
amplifier (TRl), which removes the amplitude variations from what is basically
an a.m. signal. A limiting amplifIer achieves its high gain usually with the aid of a
 from final
vision i.f.
amplifier

tuned control
limiting
circuit discriminator voltage to
amp.
39.5 MHz tuner

(a)

control
voltage
to tuner

input

~ --.,-- ---.,--"-.,..' ~
limiting amp. discriminator set d.c. filter
and tuned circuit a. f.c. amp.

(b)

39.5 MHz

(c)

Figure S.18 Automatic frequency control system: (a) a.f.c. block diagram;
(b) a.f.c. circuit diagram; (c) discrimiaator response curve
Tuners and I.F. Amplifiers 71

tuned circuit. The carrier is now amplified to such an extent that the modulated
portion cannot be accommodated by the amplifier.
The discriminator uses a Foster-Seeley type of f.m. detector formed by Dl,
D2 and the associated components. The collector circujt of TRI is tuned to the
required vision i.f. of 39.5 MHz. Consequently, when the incoming signal is exactly
at that frequency, the drive to the detector is balanced and the net voltage across
C7 is zero. Should the incoming frequency deviate from 39.5 MHz, the diodes
would conduct by differing amounts, resulting in a potential developing across C7 •
The polarity of this voltage is dependent on whether the frequency drift is above or
below the centre frequency. Therefore the discriminator output will either subtract
from or add to a preset value which is controlled by the set a.f. c. potentiometer,
VRI. The resulting voltage is amplified by the d.c. amplifier, TR2, and after ftlter-
ing (R9 , C9, RIO, C10) the d.c. signal is fed to the tuner.
When a receiver is being tuned to a programme, the a.f.c. is usually disabled. This
is done to ensure that the initial oscillator frequency is as close to the correct value
as pos!.ible, as is evident from the quality of the reception. Any sbbsequent tuning
drift will be corrected by the a.f.c., but it must be realised that the automatic
feature can only cope with a limited frequency error. It is therefore possible to
tune the oscillator to a frequency close to this limit, and the picture quality would
not be immediately affected, owing to the corrective action of the a.f.c. If, however,
further frequency drift occurred in the oscillator, a.f.c. would then lose control and
tuning would be lost. This behaviour is also found in the other frequency control
systems of a TV receiver (line time base oscillator and colour sub carrier oscillator).
An auxiliary contact can be provided on the channel selector buttons which
automatically disables the a.f.c. [for example, by short circuiting terminals TPI
and TP2 together in figure 5 .18(b)] . In other designs, channels are preselected by
the servicing engineer (with the a.f.c. disabled!), and it is assumed that the viewer
will not interfere with the settings.
It may be sometimes necessary to realign the a.f.c. detector circuit. A 39.5 MHz
signal is then applied either to the base of TRI in figure 5.18 or to the main i.f.
injection point (see alignment procedure). A voltmeter is connected between the
test points TPI and TP2 and, with L2 temporarily detuned, L 1 ' is adjusted for maxi-
mum instrument reading. Coil L2 is then retuned so as to give zero voltmeter indi-
cation across C7 • The 'set a.f.c.' control (VRl) is adjusted to produce a specified
value of d.c. as given in makers' instructions.
Ideally, the alignment of the a.f.c. tuned circuit should be carried out using a
wobbulator. In this case L2 would be adjusted to obtain a typica}. S-curve of an f.m.
detector. The expected shape is shown in figure 5.18(c), from which it is evident
that the output must be symmetrical about 39.5 MHz. If an ordinary signal genera-
tor is used and L2 tuned as previously described, the symmetry of the response
should be checked on either side of 39.5 MHz.
An a.f.c. detector can be incorporated into an I.C., usually, as part of the video
synchronous detector circuit, to be discussedjn chapter 6. A few receivers use LC.s
72 Television Principles and Practice

designed for normal f.m. sound detection, but the external circuitry has to be
adapted to the requirements of TV a.f.c.
Fault-finding in a.f.c. circuits is relatively simple. The unit can be easily disabled
to check whether normal tuning can be restored. Similarly, the a.f.c. voltage can be
monitored to observe the effect of channel tuning on the control potential.
6 Vision Detectors and Video
Amplifiers

6.1 Principles of Signal Demodulation

The Lf. signal in the receiver resembles the transmitted signal, except that the carrier
frequency is much lower. The purpose of demodulation, or detection, is to separate
the video information from the amplified Lf. signal. The waveform of the detected
video signal represents an envelope of the amplitude modulated vision carrier. As a
'by·product' of the process of video detection, it is also possible to obtain both the
intercarrier sound and the chrominance signals.
The video demodulator circuit is arranged so that the detected signal has either
positive or negative polarity-that is, the video waveform is either positive- or
negative-going with respect to the sync. pulse level. The required polarity is governed
by the following factors.
(1) The method of applying the video signal to the picture tube-cathode- or grid-
fed.
(2) The number and the type of video amplifiers-each common emitter (or common
cathode) stage will invert the signal, while emitter (or cathode) followers are non-
inverting.
In addition, the video detector has to be designed to provide the correct output
polarity irrespective of the type of modulation-positive modulation (405-line sys-
tem) or negative modulation (625-line system).
Figure 6.1 shows a block diagram of the path of the video signal in a 625-line
receiver. In this case the signal is applied to the cathode of the picture tube via a

i.f. input detector J-_.-jvideo amp.-J-_-.J

circuit driver

Figure 6.1 Block diagram of video path (see text for the waveforms)
73
74 Television Principles and Practice

driver (emitter follower) stage and the output stage (common emitter). Since the
C.R.T. is cathode fed, it requires a negative-going video signal. As the video output
stage is phase inverting, its input must be a positive-going signal. The latter is
supplied by an emitter follower; consequently, the detector must also provide a
positive-going video waveform. From reference to chapter 2 it can be seen that the
positive-going video signal appears on the lower (negative) part of the modulated
waveform.
The above comments apply to both valve and transistor receivers, except that in
the case of valve equipment a single video stage is normally used (monochrome
receivers). Similar reasoning governs the design of colour receivers or where I.C.s
are employed.
There are two types of video detector circuits in use: (1) diode detector and
(2) synchronous demodulator, both of which are described below.

6.2 Diode Detector Circuits

A diode detector is basically a half-wave rectifier whose input signal is the Lf. wave-
form. A closer study of the behaviour of the demodulator reveals, however, that
the output produced consists of the following signals:

(1) The composite video signal (the modulating envelope).

(2) The d.c. level of the video signal.
(3) Remainder of the Lf. carrier (at 39.5 MHz).
(4) Signals at beat frequencies which are due to the non-linear diode characteristics
-the detector behaves like a mixer so that the resultant outputs are
(a) 39.5 MHz - 35.07 MHz = 4.43 MHz-this is the chroma Signal.
(b) 39.5 MHz - 33.5 MHz = 6 MHz-intercarrier sound Signal.
(c) 35.07 MHz - 33.5 MHz = 1.57 MHz-chroma/sound beat signal.

The composite video signal and often its d.c. level are fed to the video amplifiers.
The remaining Lf. carrier is flltered out by a simple circuit, an example of which is
shown in figure 6.2. Capacitors C1 and C2 together with the inductor L 1 form the
Lf. fllter. Since the value of their capacitance is very small, the capacitors may be
replaced by the inevitable stray capacitance of the other components in the circuit.
The design of the detector circuit in general, and of the fllter in particular, is very
critical, to ensure that the desired high video frequencies are not severely attenuated
together with the unwanted Lf. In figure 6.2 the demodulated signal is developed
acroSS the detector load resistor, R 1 , which is in series with a high video frequency
peaking coil L 4 . Since the total impedance of R 1 and L4 rises with frequency, this
arrangement produces an increased signal voltage at high video frequencies.
The 4.43 MHz chrominance signal can be steered to the decoder in a colour
receiver, but it is unwanted in the video amplifiers of a monochrome set or in the
luminance stages of a colour receiver. For that reason a suitable fllter may be in-
corporated in a video amplifier circuit to prevent patterning on the screen.
Vision Detectors and Video Amplifiers 75

~~ to sound circuits

W
L,
.----....I----...--rr""'--+_.......---... ........_ - + _ - _ to video amp.

3.9k

C, 5.6p

final
i. f. transformer

Figure 6.2 Diode video detector circuit

The 6 MHz intercarrier sound signal can be taken off at the detector and fed to
the sound section of the receiver. However, a 6 MHz ftlter is placed after the take-
off point, again to stop pattern interference appearing on the screen. In figure 6.2
the circuit L 3 , C3 forms a trap tuned to 6 MHz, and at the same time L 2 , L3 form a
transformer which feeds the sound section.
Production of the 1.57 MHz beat signal is normally prevented by considerably
depressing the Lf. response at 33.5 MHz, as explained in chapter 5. Otherwise, a
rather objectionable and coarse interference pattern would appear on the screen
(1.57 MHz is a relatively low video frequency). The entire video detector circuit is
housed in a metal can to prevent radiation of beat frequency interference signals to
Aher sections of the receiver.

6.3 Synchronous Video Demodulation

Synchronous demodulation means that the output is obtained by 'measuring' or
sampling the amplitude of the video signal at least once every cycle of the Lf. In
other words, the demodulator 'looks' at the amplitude of the incoming modulated
Lf. in synchronism with the carrier. The sampling circuit behaves as though it were
a switch which closes at the moment when the carrier reaches its maximum value.
The output from the switch follows the modulating envelope, causing the original
video signal to be recovered. The diagram in figure 6.3 shows the principle involved,
the sampling switch being either a diode circuit or a transistor arrangement. The
ftlter at the output smooths the changes between the sampling points to produce a
continuous output waveform.
Because of the complexity of the circuitry, synchronous demodulation of the
video signal is only economical in I.e. form. Figure 6.4 shows a block diagram of
76 Television Principles and Practice

sampling at sampled

..---- .......---.-.
carrier peaks output

4 etc. 1 2 3 4 etc.

JJr
envelope

Lt. input
circuit
filter ~ video
input L--_ _ _...Jloutput

nnn switching
.J U U L pulses

synchronising
link

Figure 6.3 Principle of synchronous vision demodulator; switching pulses are

derived from the vision carrier Lf. For practical reasons, the relationship between
Lf. and picture line frequency is not correctly represented in the diagram

the Mullard TeA 270 I.e. together with some of the external components. The
drive for the synchronous demodulator 'switch' is derived from the unmodulated
part of the iJ. waveform (that is, below 18 per cent maximum amplitude). As
described in the section on a.f.c., it is possible to obtain an unmodulated 39.5 MHz
carrier by means of a limiting amplifier; in fact, this I.e. can also generate an a.f.c.
signal (pin 11, figure 6.4).
The switching action of the synchronous demodulator is accomplished during
both positive and negative half-cycles of the carrier; the negative voltage is then
inverted and added to the positive half of the waveform. Examining the Lf. wave-
form in figure 6.3 shows that this process 'fIlls in' the gaps between the positive
Vision Detectors and Video Amplifiers 77

....---..
a.g.c .

to to a.g.c.
tuner Lt. strip clamp point

a.f.c. output
'----I~ discriminator d.c. amp. buffer
stage

positive
video

Figure 6.4 Simplified diagram of video detector I.C. (Mullard). The device also
includes other functions as shown

sampling points, thus improving the detection of very fme detail. Owing to this
double switching action, the video output now has a r.f. component at twice the
original Lf. The ftltering of the 79 MHz content is simpler and does not cause serious
loss of high video frequencies, because the desired and the unwanted frequencies
are well apart; the necessary ftlter circuit is external to the I.C.
This I.C. has a built-in video preamplifier offering video of either polarity,
together with the intercarrier frequency and a.g.c. The external tuned circuit L 1 ,
C1 is associated with the limiting amplifier; L 2 , C2 is the quadrature detector coil
for the a.f.c. (here the a.f.c. discriminator circuit employs quadrature detection,
which will be described in chapter 8).
 sync. pulse level -..J
00

y' y'
d.c.leveI0

dark picture bright picture zero carrier

level
(a)

black level ®
bioc" ...
'0 ----1-- -
__ .:J____ __ ~
W
to·
(b) ~.

Figure 6.5 Effect of d.C. component in the video waveform (at tube cathode). Note: Shaded
areas above and below the bold broken lines are of equal value. (a) Complete video waveforms-
~
'S.
one line of dark picture (high d.c. level) and one line of bright picture (low d.c. level); the video ~
black level remains constant. (b) Waveforms as for (a), but the [Link] has been removed by
video coupling capacitor; black level applied to the tube is no longer constant (see text for the
~
effects on the picture) ~
!"")

~.
(II
Vision Detectors and Video Amplifiers 79
Synchronous demodulation overcomes some of the disadvantages arising from
the non-linear characteristics of the diode detector. For example, a diode detector
can sometimes give an incorrect contrast ratio-that is, the output at low signal
levels is not in direct proportion to the output at high levels of modulation
(demodulating weak signals is especially prone to distortion). The previously men-
tioned beat frequencies generated by diodes can produce undesirable patterns on
the screen, while the filtering of the residual Lf. after a diode detector is also a
problem if a loss of high video frequencies is to be prevented.
Another form of synchronous demodulator, as used in the colour decoder, is
discussed in chapter 7.

6.4 Requirements of Video Amplifiers

The video amplifying section of the receiver must have a relatively high gain and
a wide frequency range. The amplitude of the detected signal is in the range 1-2 V
peak-to-peak. The necessary signal swing to drive the tube from black to white
varies from about 40 V (small screen) to 150 V (large screen).
Theoretically, the frequency response of video amplifiers should extend from
d.c. to approximately 5.5 MHz. The requirement becomes evident if the effect of
the various frequencies is considered on the appearance of the displayed picture.
(1) D. C level is responsible for the overall illumination level of the picture. The
same scene can convey the impression of either daylight or darkness, depending
on the d.c. content. A similar effect is, of course, obtained by means of the bright-
ness control, which alters the d.c. bias of the picture tube. The d.c. component of
the video signal arises from the asymmetrical shape of the video waveform. As
shown in figure 6.5(a), the d.c. level corresponds to the average value of the wave-
form-the shaded areas above and below the y-y and y' -y' are equal. In the detec-
ted signal the d.c. level is high for a dark picture and low for a bright picture.
(2) Low video frequencies (up to 1 MHz) are responsible for coarse picture detail;
most of the sync. pulse waveform is also contained in this range of frequencies.
(3) Middle video frequencies (from 1 MHz to 3.5 MHz) contribute to most of the
average picture content.
(4) High video frequencies (from 3.5 MHz to 5.5 MHz) are responsible for fme
picture detail and clear definition. These factors are of a special importance in large
screen receivers; in a small portable TV set the size of the display is such that the
eye cannot resolve the details from a normal viewing distance. Satisfactory display
of Teletext also depends on high video frequencies.

6.S Amplification of the D.C. Component

The d.c. component of the video waveform can be amplified only if the coupling
between all the circuits from the detector to the tube is direct-that is, if coupling
capacitors are not used.
80 Television Principles and Practice

Direct coupling in a multistage valve amplifier is not easy to design, since each
succeeding valve requires an ever-increasing cathode bias to offset the d.c. anode
voltage of the preceding stage. Direct coupling between the vision detector and the
video amplifier can present problems if the resultant d.c. level is of negative polarity.
A negative-going signal could drive the valve too far towards cut-off unless the
no-signal bias was very small. In that case, however, the no-signal anode current
could be excessive, and in a Single-stage video amplifier it would also cause a low
anode voltage of the output valve, resulting in a brightly lit screen.
Direct coupling between transistor amplifier stages is easier to design than is the
case with valve circuits. The required forward bias can be derived from the preced-
ing collector (or emitter) rather than from the h.t. supply. A disadvantage of this
arrangement is that if one transistor shifts its operating point, the d.c. conditions in
the remainder of the circuit will be affected. Various circuit arrangements are used
to overcorr.e that difficulty: for example, 'tying' the emitters to a potential divider
will hold the voltages steady; applying negative feedback can stabilise the d.c. con-
ditions; or using a black level clamp will maintain a predetermined d.c. level at the
video output stage.
In view of these considerations the problem of the d.c. component can be
approached in one of the following ways.

(1) A.C. coupling throughout (via capacitors)-the d.c. component is now lost, but
the viewer can introduce it by means of the brightness control. This results in a
simple circuit, but it can produce a picture of poor contrast whenever the illumina-
tion levels change in the transmitted programme.
(2) Part of the circuit is a.c. coupled, after which the video signal is d. c. restored
and the remainder of the circuit is then directly coupled all the way to the tube.
(3) Direct coupling throughout.

The principle of d.c. restoration, or black level clamping, depends upon charging
a capacitor in one of the video amplifier stages (for example, a video coupling
capacitor) to a predetermined voltage. The video waveform is superimposed upon
that fixed d.c. level whose magnitude should correspond to the black level of the
picture. The effect of a black level clamp is illustrated in figure 6.5(a)-if the clamp
voltage forms the d.c. bias of a video amplifier, then the entire video waveform,
including its own d.c. level, can be amplified.
When a black level clamp is not used in the receiver [see figure 6.5(b)], the
coupling capacitor blocks the transmitted d.c. level of the waveform. Consequently,
the correct black level corresponding to the cut-off voltage of the tube cannot be
maintained. If the brightness control is adjusted correctly for the dark picture, then
everything above the resultant black level (A) would indeed be black. Therefore,
should the scene become bright, the tube would still remain cut off between the
signal voltages corresponding to black levels (A) and (B). Conversely, if the receiver
were adjusted for correct level (B), then a change to a dark scene would result in a
picture which would be too bright.
Vision Detectors and Video Amplifiers 8]

Maintenance of the correct black level and the sync. pulse level is also important
to the reliable functioning of many a.g.c. and sync. separator circuits, and it is vital
in colour receivers.
The circuit in figure 6.6 shows a simple diode d.c. restoring circuit particularly
popular in valve amplifiers. Without d.c. restoration, the video waveform has posi-
tive and negative excursions about the zero d.c. level (X' -X'). When applied to the
clamp circuit, negative excursions of the video waveform cause the restoring diode
D to conduct, charging the coupling capacitor C. The capacitor charges to a voltage
which is equal to the peak of the sync. pulse because the discharge time constant
R x C is relatively long and no significant loss of charge occurs between the pulses.
The video waveform 'sits' upon this reference potential so that the d.c. level is now
included, or restored, in the signal.

from video~l to grid of

detector - . - - - - - - - - video output
11' valve
zero r-\
level -'-\f--+X"""-~x~·V'""~lT"VsYnc. D1

waveform without d.c.

restoration

restored d.c.

zero
level
'i1l7\t(
output waveform with
diode restoration

Figure 6.6 Simple diode restorer circuit. Dl clamps the tips of sync. pulses to
zero level (chassis), which automatically gives constant black level, and the video
d.c. level is restored

Since the d.c. restorer clamps the peaks of the sync. pulses, it follows that the
black level of the video waveform would also remain constant. Unfortunately, the
simple circuit described above has to adjust itself to any changes in the amplitude
of the negative-going part of the waveform. The required adjustment, however, is
rather slow because of the long time constant of the diode circuit. Consequently,
the actual black level can be affected by variations in signal strength (e.g. when
changing channels) or even by the picture content.
In order to ensure a stable reference point for the black level, driven clamp
circuits are used. In these the video (or chroma) coupling capacitor is charged to
a d.c. potential via a special circuit which is switched on by the line flyback pulses.
The magnitude of this potential is often governed by the setting of the brightness
82 Television Principles and Practice

control. Such a relatively complex arrangement is universally used in colour

receivers. It is absolutely essential that the three video drives to the colour tube
have stable black level potentials. Any drift in the d.c. bias of one video output
amplifier with respect to that of the other two would alter the tube current
associated with the particular gun and lead to unwanted colour tinting on the
screen. A driven black level clamp effectively 'resets' the d.c. bias of each gun to
a fixed level at the beginning of every picture line.

~----~~------.--+

to picture tube
t.-,V'""R=-:2~ cathode (red)
C,
video in----l.-_~>--t_t
(red)
+

01 brightness
control

02

~..-.~- line pulses

clamp pulses {
to G and B
amplifiers

Figure 6.7 Example of a driven black level clamp circuit in the 'red' output stage
of a colour receiver (a similar arrangement is repeated in the 'green' and in the 'blue'
amplifiers). In this circuit brightness control effectively alters the black level of
video waveform. (Decca)

An example of a driven black level clamp is shown in figure 6.7. Transistors TRI
and TR2 form the video output stage driving the red cathode of the tube. The base
bias circuit for TRI is similar to a conventional potential divider formed by Rl and
R z . The difference is, however, that the lower end of R z is not connected directly
to the chassis, but to the capacitor Cz . Positive-going line flyback pulses are fed to
the circuit via C3 and R4 ; when the magnitude of each pulse is equal to the collector
voltage of TR2, the clamp diode DI conducts and the resulting current flow charges
Vision Detectors and Video Amplifiers 83
capacitor C3 • Therefore the voltage across C3 is dependent upon the difference
between the constant amplitude line pulse and the actual collector voltage, just as
the video waveform is at its black level. During the line scan period the charge on
C3 maintains a relatively fIxed potential at the junction of R z , R3 and Cz . This
provides a reference voltage upon which the picture information is superimposed.
The circuit is repeated for each of the remaining two colours, all three arrangements
receiving the same clamping pulses.
The effectiveness of the clamp circuit in stabilising the black level at the video
output stage can be seen by considering, for example, a fall in the collector voltage
of TR2. Now capacitor C3 in fIgure 6.7 charges to a higher voltage, owing to the
increased potential difference, as outlined above. Consequently, the higher negative
voltage on C3 reduces the collector current of both TRI and TR2. The resultant
rise in the collector voltage of TR2 compensates for the original reduction in that
potential.
In some receivers black level clamping is carried out within an I.C. Suitable
capacitors, equivalent to C2 or C3 in fIgure 6.7, are then connected externally to
the device.

6.6 AmplifIcation of Middle and High Video Frequencies

The middle video frequencies do not present serious problems, and we devote our
attention to the amplifIcation of high video frequencies.
Readers will recall that the high frequency gain of an amplifier reduces with in-
creasing frequency, the principal cause being stray capacitance, which shunts the
amplifier output. The stray capacitance is unavoidably present in the circuit (see
also section 5.1). This problem is worse in valve stages than in transistor amplifiers,
since the high resistance of the anode load means that the stray capacitance offers,
at high frequencies, a comparatively low reactance path to the chassis.
A technique known as frequency compensation is used to extend the bandwidth
of the amplifier at the high frequency end of the spectrum. Two popular methods
of applying compensation are: (1) the use of so-called peaking coils, which are
connected in the output of the video amplifier, and (2) the application of frequency
dependent negative feedback to the amplifier. Both techniques are used in the
amplifier in figure 6.8, and are described below.
Inductor L I in figure 6.8 is a peaking coil. At middle and low frequencies the
reactance of L I is small compared with the resistive load, R 4 . At high frequencies
the reactance of LI increases in value, so that the effective impedance of the anode
load is increased; this results in an increase in gain with frequency. At some fre-
quenc,y the inductance of LI could be made to resonate with the stray capacitance
and give maximum amplifier gain; damping resistor R3 is used to prevent unwanted
oscillations. When connected in the position shown, L I is known as a shunt peaking
coil, since it effectively appears as though it were connected across the signal path.
An alternative arrangement is to connect the peaking coil, together with its damping
resistor, in series with the signal path. Such a series peaking coil could, for example,
84 Television Principles and Practice

replace the parallel combination Cz ,R6 in figure 6.8. The action of the series induc-
tor breaks up the stray capacitance of the circuit into two parts, CSTRAYl (associa-
ted with the valve) and C STRAYZ (associated with the picture tube). This effectively
reduces the capacitance connected directly to the video output valve and the gain
can be increased.

__- - _ to sync. separator

H~-""'-" to picture
I I tube cathode
I I

·c
I I
~ideo _ _ _ _~====t

1.
I Rs I
in I
lOOk

TC st"vl1I
R7
220k
..,..

I
stray (2)

c1 I
I
I
I
1000p I I
I I

Figure 6.8 Video output stage with high frequency compensation given by the
action of L l , C l , Cz and the choice of value for R4

Frequency dependent negative feedback is applied by means of the cathode

resistor R z and its bypass capacitor C l . At middle and low video frequencies the
reactance of C l is high, so that R2 is not effectively bypassed. The resultant
negative feedback reduces the voltage gain of the amplifier at those frequencies. At
high video frequencies the capacitive reactance of C l reduces, which gives a reduc-
tion in negative feedback, leading to an increase in amplifier gain.
The parallel combination R6 , C2 also acts as a frequency compensating network.
At low frequencies the reactance of Cz is high, and the video signal is attenuated by
the potential divider action of R6 and R 7. At middle and high frequencies the react-
ance of C2 reduces, so that an increasing proportion of the signal voltage is applied
to the tube cathode.
It is of interest to point out here why the video output stage has a relatively
high power rating. In order to maintain a good frequency response, it is necessary
Vision Detectors and Video Amplifiers 85

to provide a low circuit resistance-this minimises the effect of stray shunt capacit-
ance. A low value of circuit resistance implies high peak values of current and, when
taken with the high voltages involved, means that the power dissipation in the out-
put stage must be rathe, high. On the other hand, the power taken by the picture
tube itself is supplied by the line output stage and not by the video amplifier.
While one attempts to boost the high frequency gain, the low frequency res-
ponse is sometimes suppressed. To redress the balance some form of low frequency
compensation could be used. This is achieved in the circuit in figure 6.8 by means
of C3 and Rs. At middle and high frequencies Rs is fully decoupled by C3 , so that
it does not affect the amplifier gain. At low frequencies the decoupling action of
C3 is ineffective, which places Rs in series with R4 to give a high gain.
Practical methods of compensation used in other circuits depend on a number
of factors, including the type of picture tube, the characteristics and the layout of
the components used, the Lf. response, etc.

6.7 Luminance Stages in Colour Receivers

The luminance signal in colour receivers is the equivalent of the video signal in
monochrome receivers. Therefore similar requirements regarding the bandwidth and
gain have to be fulftlled in the luminance stages. (A block diagram of a typical
luminance stage is shown in figure 6.18.)
As shown in previous chapters, the processing of the signal in a colour receiver
takes two separate paths. Finally, however, both the luminance and chrominance
are combined to produce the three colour signals. The bandwidth of the chroma
signal is initially 2 MHz, reducing to 1 MHz after the detection; but the bandwidth
of the luminance path is 5.5 MHz. This difference in the frequency response causes
the signal to pass faster through the luminance amplifiers than through the decoder.
Although the theoretical proof is beyond the scope of this book, the problem could
be considered to be similar to the results of square wave testing of electronic ampli-
fiers. When a square wave is applied to the amplifier input, the resultant output
should also be a square wave. Owing to the deficiencies in the amplifier perform-
ance, however, this- is not always the case. If the response to high frequencies is
limited, the leading edge of the output waveform slopes, and the maximum ampli-
tude is reached after a time delay, as shown in figure 6.9. (A similar effect also
occurs at the trailing edge of the waveform.) The net result of this type of response
in the TV receiver would be that the luminance signal would be applied to the tube
a fraction of a second before the corresponding chrominance signal. The visual
effect depends upon the size of the screen and on the actual time delay between
the signals; it ranges from an impression of poor focus to one of double image as if
suffering from 'ghost'. The time difference between the two signals depends on the
amplifier design, and typically lies in the range 0.5-0.9 JJ.S (500-900 ns). If the
width of the picture is, say, 52 cm (approximately 20 in), and since the duration of
one picture line is 52 JJ.S, then the beam moves across the screen by 1 cm in 1 JJ.S.
Should the signal delay be 0.8 JJ.S, the display would consist of two 'pictures' which
86 Television Principles and Practice

would be 8 mm (approximately 1- in.) apart.

To ensure that the two signals are correctly superimposed, the luminance path
contains a delay line. The device is in the form of a long coil with copper strips
placed between the winding and the coil former. The strips are connected to the
chassis to form a distributed capacitance over the whole length of the coil. Such an
arrangement is called a transmission line; one of its properties is the ability to re-
duce the speed of signal propagation. Figure 6.9 shows how a delay line, L 1, is
connected in circuit; components L 2 , L 3 , R 1 and R2 provide the correct termina-
tions to the line. Faults in the terminating components could lead to multiple
'ghosting'.

time delay
14 ~I
.,

{ '<
"0
.~
C.
E , slow rise
co
(poor response)
--+---'
fast rise
(good response)
(a)

--.----------~~-h.t.

R1 1.8k

luma. stage -r:

from previous ....._ _~LA:.2~-"_'_L~1~-'_'__Lu3~.......- - t o next
luma. stage
~elay line

(b)

Figure 6.9 Principle ofluminance delay line: (a) amplifier response to square wave-
form (see text); (b) practical circuit

An open circuit in the luminance path results in an extremely poor definition

picture-due to the chrominance signal only (assuming that time base synchronisa-
tion has not been lost).
The luminance amplifier in a colour receiver also contains a 4.43 MHz notch
filter. This is a tuned rejector circuit whose purpose is to remove the modulated
chrominance information centred around 4.43 MHz. The presence of such a signal
in the drive waveform to the picture tube could cause a very fine dot pattern on the
screen. The 4.43 MHz signal could also produce desaturation of certain colours in
the display. This would occur as a result of the action of the tube which rectifies
Vision Detectors and Video Amplifiers 87

the unwanted signal and adds it to the existing luminance information. The reader
will recall that the only acceptable addition of signals is that of the correct value
luminance matrixed with the demodulated colour difference information.
It will be noted that there are a few monochrome receivers equipped with a sub-
carrier notch fllter-again to prevent the dot interference pattern on the screen. On
the other hand, the rectifying and signal adding properties of the tube could be
beneficial to a monochrome display by improving the apparent contrast of the pic-
ture content.
The presence of the notch fllter affects the high video frequency response of the
luminance amplifier. It can be seen from the graph in figure 6.10 that the response
curve has a notch around 4.43 MHz and a certain loss of fme picture detail is thus
unavoidable. Some designs make provisions for disabling the notch fllter during a
monochrome transmission (when the colour killer operates) so as to maintain good
picture quality.

......_..... luminance
out
luminance
in ---+-I

5.5 MHz
l;; 01--_.... I
~CD I
~:E
IV C

~ri!,
.~.~
E.!! -20 - - - - - - -
.a!

4.43 6 MHz
MHz
video frequency

Figure 6.10 Example of 4.43 MHz notch fllter (Ll' C 1 ) in the luminance channel;
the graph shows the effect of the fllter on frequency response

6.8 R, G, B Drive Circuits

It was mentioned in section 6.1 that the picture tube can be either cathode- or
grid-fed. A colour tube has three electron guns, one for each of the primary colours,
red (R), green (G) and blue (B). The light output associated with a particular colour
88 Television Principles and Practice

is the result of the addition of two signals: the luminance signal and one of the
colour difference signals. This addition can take place in an electronic circuit
known as the R, G, B matrix, or it may take place within the tube itself.
Receivers which use an R, G, B matrix external to the tube are called R, G, B
drive receivers. In these the three primary colour signals are individually applied to
the cathodes of the tube. This method is most suited to transistorised output stages,
and it is widely used for two principal reasons. Firstly, the cathode drive voltage
needed to control a given beam current is lower than that necessary on the grid
of the tube; consequently, the required voltage rating of the output transistors is
reduced. Secondly, the low output impedance of the transistor amplifiers, com-
pared with their valve counterparts, allows the full 5.5 MHz bandwidth to be
easily obtained.
The alternative method of driving the picture tube is known as the colour
difference drive. In this arrangement, as discussed in section 6.9, the matrixing of
the luminance with the colour difference signals takes place within the tube.
An example of an R, G, B matrix circuit is illustrated in figure 6.11. The three
colour difference signals are amplified by TRl, TR2 and TR3, respectively;Rz,R l l
and R 13 are the amplifier load resistors. The luminance input signal to the matrix is

.--+-----------<!r--.-------- h.t .
luminance
input -------.>-----+--+--------+--+-----, to RED
to GREEN output
(figure 6.7)

to BLUE
output - - - -.......

(B'- V') in t-+---(R'- V') in

Figure 6.11 (G' - V') and R', G', B' matrix (Decca). (G' - V') is formed at the
emitter ofTR2 from (B' - V') and (R' - V') fed from TRI and TR3, respectively;
R', G', B' signals appear at the collectors
Vision Detectors and Video AmpUfers 89

fed via frequency compensating networks (R1 C1 , R 9 C4 and R 14 CS ) so that the

signal addition takes place at the collectors of the transistors. The resultant three
primary colour signals, after further amplification (see figure 6.7), are applied to
the cathodes of the tube.
The circuit shown in figure 6.11 includes the (G - Y) matrix. The reader will
recall that the transmitted chroma signal contains only the (B - Y) and (R - Y)
components. The third colour difference Signal, (G - y), is recovered in the receiver
by the addition of differing proportions of the other two signals. This addition, or
matrixing, is done via Rs and R 12 -note the unequal values of resistance which give
the required proportions. The (G - Y) signal is then amplified by the common base
amplifier containing TR2.
The gain of the colour difference amplifier is also adjusted to provide deweight-
ing of the demodulated signals. In order to bring the signal outputs to a common
level, the (B - Y) amplifier needs a higher gain compared with the (R - Y) ampli-
fier. This compensates for the unequal weighting factors applied at the transmitter.
In the circuit in figure 6.11 the deweighting is brought about by including a higher
value of unbypassed resistance in the emitter of TR3 than is the case with TRI. The
d.c. biasing remains the same for each amplifier; therefore the values of (R3 + R 4 )
and R 15 are nearly equal to each other.
The R, G, B matrix, together with the (G - Y) matrix, is often provided in a
suitable I.C., an example of which is discussed in chapter 7.

6.9 Colour Difference Drive

In a colour difference drive receiver the picture tube is used to combine, or to
matrix, the luminance and colour difference signals. The colour difference signals
are fed to the three grids, and the luminance signal is simultaneously applied to all
three cathodes. The resultant beam currents are then proportional to the primary
colour signal content. Although this technique is widely used with valve output
stages, there are also a few designs which employ transistorised colour difference
drive circuitry.
Signal voltages required on the grid of the picture tube are about 50 per cent
above those found in circuits where cathode drive is used. Consequently, the volt-
age and the power ratings of transistors needed in a receiver with a colour difference
drive would have to be relatively high. The reliability of valves is not seriously
affected by such elevated voltages; on the other hand, the design of transistor cir-
cuits tends to favour the lower operating potentials. A comparison of the two
methods of driving the tube shows that the R, G, B drive requires three full band-
width (5.5 MHz) output stages at lower voltages, while the colour difference
receiver needs one 5.5 MHz output stage and three relatively high voltage amplifiers
of only 1 MHz bandwidth.
The colour difference output amplifiers logically belong to the colour decoder,
therefore their arrangement is considered in chapter 7. The associated luminance
output stage is represented by the diagram in figure 6.12. Basically, it is a video out-
90 Television Principles and Practice

put amplifier, the signal being fed to the tube cathodes via drive controls VR2 and
VR3. These adjustments are part of grey scale tracking, which is described in
chapter 12. Other features of the circuit that are worthy of note are

--.----1r---1---~--h.t.

to C.R.T.
cathodes
,
red green blue

luminance
irl
----~I~-.--~--~·

line}
flyback pulses
field

Figure 6.12 Luminance output stage in a colour difference drive receiver. Circuit
includes flyback blanking, brightness control, beam limiting, d.c. restoration, high
frequency compensation and tube drive controls (part of grey scale tracking.
arrangement)

(1) the d.c. restoring circuit (C 1 , D 1 , R 1 );

(2) series peaking coil circuit (Ll' R 3 );
(3) middle and low frequency compensation (C4 );
(4) very low frequency compensation (Cs );
(5) high frequency compensation (C3 , Cs , C9 ).
The action of TRI and the beam limiting circuit will be discussed later in this
chapter.
Vision Detectors and Video Amplifiers 91

6.10 Brightness Control

Brightness control effectively changes the d.c. level or d.c. bias applied to the pic-
ture tube. The setting of the control is normally such that the screen is dark in the
absence of any picture detirll (this corresponds to the black level!). Occasionally a
preset brightness potentiometer is provided in addition to the viewer's control; the
preset is then adjusted according to the makers' instructions, since it often aims at
specified voltage levels.
In most monochrome receivers the brightness control alters the d.c. potential
on the control grid, the cathode of the tube being left free for video signals. The
brightness control voltage can also be applied to the cathode of the tube-either
directly, or indirectly via the video amplifying stages.
Sometimes the control circuit features a facility for switch-off spot suppression.
At the instant when the receiver is switched off, deflection of the beam ceases, but
the cathode of the tube is still hot and the e.h.t. is also present owing to the charge
stored on the final anode capacitor. These conditions allow the beam current to
continue flowing for a while, resulting in a stationary bright spot in the centre of
the screen. In time, the phosphor coating of the tube would become damaged and
create a dark patch in the middle of the picture.
One form of spot suppression circuit, shown in figure 6.13, has a voltage depen-
dent resistor (VDR) in the 'earthy' end of the brightness control. When the receiver
is switched on, normal h.t. is applied to the circuit, and the resistance of the VDR is
low, effectively connecting the lower end of R2 to the chassis. Upon switching off
the h.t. decays, the resistance of the VDR rises, and the result is that the brightness
control appears to be disconnected from the chassis. That way the remaining posi-
tive h.t. is momentarily fed to the grid to increase the beam current and discharge
the e.h.t. capacitance before the scanning has finally collapsed.

h.t.

mains L--<>~} to ~_----"~ to C.R.T.

Ni~
II
grid
supply receiver
I - -h-
I
I

r -----------
I
L------
alternative connection (see text)

Figure 6.13 Brightness control incorporating switch-off spot suppression

92 Television Principles and Practice

The alternative method of connection, shown by a broken line in figure 6.13,

also provides switch-off spot suppression. Instead of using a VDR, the 'earthy' end
of R2 is connected to the neutral on the mains side of the ON/OFF switch. When
the receiver is switched on, the brightness control is connected to the chassis via
the switch and operates normally. Opening the mains switch breaks the chassis
connection and the h.t. is fed to the control grid. This circuit, however, presents a
potential hazard; should the mains supply connections be reversed, the inside of the
receiver would be live despite a double pole switch!

Q) l!l
E g> ~ 'E
g ..t:~:=
~ C E
_0",
o U Q)
PE D
C:o-O
o ~ c:
U _ '"

clamp.
reference
level amp.

delay
1st chroma
chroma amp. line

Q)
2 a;>
gE -0
c:
~~
OJ
E 1:
:J
'"
'"O...J
::; '" 2' u E-O
'"E .., ~
c:= 0
0'- >

e:;
',t:; '+-Cl
.-~-
.E«~
e U cO
e
*
'"
Cl '"
Q,) .c: uQ.. :.:
.J::
cO :J 0 .., Q)"'Q.
-

u-o <ti ..,_ c:

'" 0
'" u u 0 :§~~

Figure 6.14 Luminance and chrominance I.C., TBA 560 C (Mullard). Functions
associated with the chroma path are described in chapter 7
Vision Detectors and Video Amplifiers 93

Another method of spot suppression is based on the use of a large value capacitor
which normally decouples the slider of the brightness control; on switching off, the
capacitor maintains a positive voltage on the grid of the tube.
In some receivers the supply to the brightness control is between the positive h.t.
rail and a negative supply. The negative supply can be obtained by rectifying (and
smoothing) the negative-going flyback pulses-a VDR can be used for this purpose.
Brightness control in colour receivers can present some problems. The colour
tube has three guns whose outputs must be altered equally to prevent mistracking,
which, in turn, would cause colour changes. The brightness control can be connec-
ted to a black level clamp circuit as shown in figure 6.7. The action of this arrange-
ment was described in section 6.5, where it was explained how the overall d.c. bias
of the output stage was controlled by constant amplitude line pulses. In fact, the
amplitude of those pulses is governed by the setting of the brightness control VRI,
since diode D2 clips their magnitude to the value of the potential appearing at the
slider of VRI. (The clipping diode conducts as soon as the amplitude of the pulse
exceeds the voltage from the brightness control.)
In R, G, B drive receivers the brightness control can also be connected to the
three grids strapped together; the arrangement then functions in a manner similar
to that of a monochrome tube circuit. Where colour difference drive is used, the
grids are fed with separate signals; therefore brightness could be varied via a black
level clamp. Alternatively, the d.c. level of the luminance output amplifier can be
SUitably altered by the brightness control, as shown in figure 6.12. Potentiometer
VRI adds a d.c. potential at its slider to the d.c. restoring circuits which fmally
adjusts the d.c. bias of the output valve.
Brightness control can also be carried out in an I.e.; a block diagram of a suitable
device is shown in figure 6.14. Again, the picture brightness is varied by altering the
d.c. level of the luminance signal.

6.11 Contrast Control

It could be said that the effect of contrast control is the video equivalent of a
volume control, its purpose being to vary the amplitude of the video signal. Picture
contrast can be altered by controlling the amount of video fed from one stage to
the next, or by varying the gain of a video amplifier. Alternatively, contrast control
is achieved by changing the gain of the Lf. strip, either directly or via the a.g.c.
Examples of contrast control techniques are shown in figure 6.15. In circuit (a)
the video output is collected at the slider of VRI and applied to the tube. In circuit
(b) the amount of video signal fed from TRI to TR2 is adjustable by means of VRI.
An important feature of this arrangement is that the contrast control is connected
between two points of equal potential marked x and y in the diagram. The choice
of values of R2 ,R3 and the operating point of TRI ensure that there is no voltage
difference along the track of VRI ; this prevents any changes in the d.c. biasing of
TR2 whenever the contrast control is altered. The reader will recall that variations
of this bias would affect the picture brightness.
 to tube
cathode
video
in

video ___
in

C1

(a) (Thorn) (b) (Bush-Murphy)

to tube
~-----i~ cathode
luminance
video in-oo"':--+-I in -00--1--.
luma. to
-II out~ut stage
J....-......
C1

to slider
of colour control
(c) (Thorn) (d) (Pye)

Rl
a.g.c. line to i.f.
amp.
VRl
in
C1
R2

(e) (Thorn)

Figure 6.15 Examples of contrast control arrangements

Vision Detectors and Video Amplifiers 95

The arrangement in figure 6.15(c) uses variable negative feedback to alter the
gain of the amplifier. The capacitance of C2 has a low value to provide high fre-
quency compensation, while C1 is a large value bypass capacitor. The position of
the slider of VRI determines the amount of negative feedback within the stage and
alters the gain.
The circuit in figure 6.1 5(d) shows a method of contrast control found in a
colour receiver. Here both the contrast and the colour (saturation) controls are
interconnected to provide a degree of tracking between the two adjustments. This
is important if the balance between the luminance (contrast control) and the
chrominance (saturation control) is to be maintained.
The arrangement in figure 6.15(e) achieves contrast control by altering the bias
of the a.g.c. detector/amplifier transistor, which, in turn, varies the voltage on the
a.g.c. line, thus changing the gain of the Lf. strip. Contrast control can also be
incorporated in an I.C., as shown in figure 6.14.
Incorrect adjustment of the contrast control can have serious effects on the per-
formance ofthe receiver. Insufficient contrast can produce a weak and noisy pic-
ture, and in some circuits it can cause no colour as well as poor synchronisation.
Excessive contrast may lead to overloading of the video stages, again with a possi-
bility of poor synchronisation. The latter would arise if the signal were so large that
the sync. pulses could not be accommodated within the amplifi~r output.

6.12 Flyback Blanking

The purpose of flyback blanking is to cut off the electron beam of the picture tube
during each line and field flyback interval. If this was not done, bright flyback lines
would spoil the display on the screen. In fact, the presence of Teletext information
at the beginning of each field can affect the brightness level at the top of the screen.
For that reason the field flyback blanking period is often extended to include the
Teletext time.
Flyback blanking is performed by flyback pulses derived from both the line and
field time bases. In order that they may be applied either directly or indirectly to
the tube, such pulses must be of a correct amplitude, suitable polarity and duration.
In most monochrome receivers the suppression pulses are communicated to the
control grid of the C.R.T. so as to override momentarily the setting of the brightness
control. Apart from the pulses being passed through a resistor-capacitor network
which offers the necessary shaping and timing, a blanking transistor may be used to
act as a switch and/or pulse amplifier.
In colour receivers flyback blanking pulses are frequenctly fed to the brightness
control circuit; again they overcome the setting of the control and the tube is
ultimately driven towards cut-off. Similar problems to those discussed in section
6.10 on brightness control apply to the methods of flyback blanking.
The arrangement shown in figure 6.12 causes the luminance output valve to be
switched off by the blanking pulses. Transistor TRI in the cathode circuit of the
valve is normally biased into saturation via R 6 , effectively connecting the lower end
96 Television Principles and Pract:ce

of R2 to the chassis. Negative-going flyback pulses are applied to the base of TRI
so that the cathode circuit of VI is briefly interrupted. This causes the anode volt-
age of the valve to rise to h.t., and the resultant high positive potential communi-
cated to the cathode of the tube cuts the beam off. Diode D2 protects the transistor
base-emitter junction against negative-going pulses.
Flyback blanking can be also carried out inside an I.C., as shown in figure 6.14.

6.13 Beam Umiting

Beam limiting is used in TV receivers to restrict the magnitude of the picture tube
current to a specified value. Some method of restricting the beam current is used in
most, if not in all, colour receivers and it may be occasionally employed in mono-
chrome receivers. An excessive tube current can overload the line output stage and
cause a reduction in e.h.t. which, especially in colour receivers, could spoil the
quality of the picture (poor convergence, purity and focus). A considerable increase
in beam current might cause overheating of the shadow mask which would lead to
serious and permanent colour purity errors. Since the tube current has to be sup-
plied by the line output stage via the e.h.t., the overload condition may also
damage the e.h.t. rectifiers.
An excessive beam current could be caused by a fault in the circuits supplying
the tube; alternatively it may occur on the peak whites, especially if the brightness,
contrast or black level clamps were incorrectly adjusted. The limiting circuit is
designed to sense the magnitude of the beam current and then develop a control
signal which reduces either the brightness or the contrast, or lowers the power
supply voltage.
Examples of beam limiting circuits are given in figure 6.16. In circuit (a) the
beam current is sensed by transistor TRl, whose base potential is governed by the
voltage drop across a low value resistor, R 4 • This resistor is in the cathode circuit
of the line output valve, VI, as the line output stage is also responsible for the
current taken by the picture tube. Transistor TRI is normally cut off by the posi-
tive bias on its emitter from VR2; therefore the brightness control circuit is not
affected. An excessive beam current causes the cathode voltage of VI to rise above
the emitter potential of TRI so that the transistor turns on. The resultant conduc-
tion of TRI lowers the voltage at the brightness control (point X at VRl), effec-
tively reducing the picture brilliance. Capacitor C1 bypasses the high frequency line
scan current.
An alternative method of beam current limiting is shown in figure 6.16(b) and it
is used in a fully transistorised, R, G, B drive receiver. The beam current is sensed in
the 'earthy' lead from the e.h. t. overwind, since the path of this current is from the
e.h.t. winding, then through the e.h.t. rectifier and the picture tube to the chassis.
The return path from the chassis to the winding is via diodes D1 and D2; it would
appear from the above direction of the current flow that Dl conducts in the reverse
direction. In fact, Dl acts as a switch being turned on by the forward bias viaR l
from the positive h.t. rail. It can now be seen that two opposing currents flow
 +-------
to luma.
amp. (see _
figure 6.12)
C1

(a)

D3 D4
..1--_. final anode
~ ~ [Link]
e.h.t. II --1"
..- - "
winding II
II
on line II
output :1 -~h.t.
transformer ,I

URl

1---<_-...... to picture tube

control grids

c S
J~--.~>-----4~__-
(b) (Thorn)

to tube
cathode

(c) (Phillips)

Figure 6.16 Examples of beam limiting arrangements

 R,

fromfinal~ R7 Rg
Lt. amp. - --1 to tube
cathode

L)H
- ( II
,
---I
L5

to sync.
separator
C. C5
0.01!, 80!,
33.5 MHz
w
intercarrier sound --1
to a.g.c. ---.J

Figure 6.17 A complete monochrome video detector/amplifier circuit (Rank)

Vision Detectors and Video Amplifiers 99

through D1-the 'forward' current due to the h.t. supply and the 'reverse' current
due to the beam current. If the tube demand increases to a point where the 'for-
ward' and 'reverse' currents through D1 are nearly equal in value, the diode resist-
ance rises, preventing any further increase in the beam current. The circuit responds
to a change in the average value of the beam current, its a.c. component is by-
passed by Cz and C 3 •
In colour receivers the beam limiting circuit restricts the tube current to between
1 rnA and 1.5 rnA, depending upon the size of the screen.
In monochrome receivers a simple form of beam limiting circuit is sometimes
used, a typical example of which is shown in figure 6.16(c). Here the beam current
flows from the tube cathode via R4 and the parallel combination of R3 and the
video output stage (TRi). When the voltage at the collector of TR1 is higher than
the potential at the junction of R3 and R 4 , the beam limiting diode D1 is biased
ON and the circuit functions normally. An increase in beam current raises the volt-
age at the cathode of D1 and causes the diode to turn off. The tube is then left with
its own cathode bias developed across R3 and R 4 , which should restrict the current;
the video waveform can still be fed to the tube via Cz . This circuit can also limit
the beam current under conditions of excessive tube drive caused by the video out-
put stage. In particular, a fault in the amplifier which results in a very low voltage
at the collector of TR1 would otherwise lead to a considerable beam current. How-
ever, under these circumstances the anode voltage of the limiting diode falls below
the cathode potential and D1 turns off. The beam current is again limited by the
tube self-bias developed by R3 and R 4 •

6.14 Typical Arrangements of Video and Luminance Amplifier Circuits

In monochrome receivers which use a valve output stage the diode detector is a.c.
coupled to the grid of the valve, usually with d.c. restoration at this point. A pen-
tode is used as the video output amplifier, which, apart from feeding the cathode
of the tube, also supplies the sync. separator. Frequently both the a.g.c. and the
intercarrier sound circuits derive their input from the video output stage. Another
popular alternative is to feed the signal to the sound section from the video detec-
tor. In either case a 6 MHz intercarrier sound trap is provided in the video path to
prevent pattern interference on the picture.
In monochrome transistor receivers the video section consists of two amplifying
stages-namely a driver stage (emitter follower) and the output stage (common
emitter). The video detector is directly coupled to the emitter follower, whose
function is to supply a relatively large base current needed by the video output
transistor. The driver stage offers a high input impedance to prevent excessive
loading of the detector and the preceding i.f. stage. The low output impedance of
the emitter follower allows this stage to be used as a distribution amplifier in order
to feed a number of circuits such as the a.g.c., the sync. separator, the sound stages
and the video output itself.
100 Television Principles and Practice

A practical version of a monochrome video detector and amplifier circuit is

shown in figure 6.17. From earlier description the reader will recognise the func-
tions of many components in the arrangement, but some additional features need
the following explanation. Turning to the video detector first: diode D1 produces
a negative polarity output voltage, although its video signal content is positive-going
(see section 6.1). Transformer L 1, L2 is the final i.f. transformer; L3 ,L4 is a trap
to attenuate the sound carrier as part of the i.f. response shaping system. The i.f.
fIlter after the detector is formed by L s , L 6 , C2 , C3 and C6 • The video signal is
developed across the load resistor R 2 , whose chassis return connection cannot be
made directly, as it would upset the d.c. biasing of TRI. Instead the upper end of
R2 is decoupled via C4 and Cs ; two capacitors are used-a high value electrolytic
capacitor to bypass the low and middle frequencies and a relatively low value C4
to decouple at high frequencies. (Electrolytic capacitors, by virtue of their con-
struction, are not very effective at high frequencies.)
The emitter follower in figure 6.17 is biased via R 3 , VR1, R4 ; the preset control
VR1 must be carefully adjusted, since it affects the operating point of the entire
video section (d.c. coupling throughout!). An incorrect setting would clip either the
peak whites ('white crushing') or the sync. pulses. As the intercarrier sound is taken
off at the emitter follower stage (TR1), the 6 MHz signal is fIltered out by Cs , Ls
in the feed circuit to the tube.

sync.
separator sound

from video R
detector G
B

a.g.c. chroma to 4.43 MHz

decoder notch filter

L ____ disabling voltage

from decoder

Figure 6.18 Block diagram of the luminance stages of a transistorised receiver

(R, G, B drive)

Where integrated circuits are used, a considerable amount of signal processing is

done within the I.e. itself, depending, of course, on the type of device. It is usual,
however, to find the video output stage as a separate transistor amplifier because
of the relatively high voltages required for driving the tube.
Vision Detectors and Video Amplifiers 101

In colour receivers the luminance channel follows the pattern outlined for mono-
chrome receivers unless various specialised I.C.s are used, which are described in
chapter 7. For quick reference a block diagram of a luminance channel is shown in
figure 6.18.

6.15 Automatic Gain Control (A.G.C.)-Basic Principles

The chief purpose of a.g.c. is to maintain a relatively constant picture contrast
irrespective of changes in the signal propagating conditions, programme selection,
etc. (Many receivers have a contrast control in the form of a preset adjustment not
normally accessible to the viewer.) An efficient a.g.c. is also essential in order to
maintain a constant level of synchronising pulses, which ensures reliable triggering
of the receiver time bases.
The a.g.c. system is based on a feedback arrangement which includes within its
loop the r.f., the Lf. and the video amplifier stages of the receiver. The system
develops a d.c. control voltage which is dependent upon the amplitude of the detec-
ted video waveform. The resultant control signal is applied to the Lf. amplifiers
(and often to the tuner r.f. amplifier), causing the gain to be adjusted accordingly.
To ensure satisfactory operation, the a.g.c. must respond to the strength of the
received signal. This is relatively easy to do in radio receivers, where the d.c. level
of the detected signal is simply a measure of its strength. However, in the case of
television, the [Link] (average value) of the detected video depends not only on
the signal strength, but also on the picture content. Despite this limitation, mean
level a.g.c. has been used, especially in receivers designed for 40S-line operation.
The more modern type of a.g.c. derives its control voltage from the peak amplitUde
of the detected video; for this reason there can either be a peak level a.g.c. or its
variant, known as the gated (keyed) a.g.c. system.
The design of the a.g.c. controlled Lf. (or r .f.) stage is such that its gain varies
with the applied base bias. In most applications the gain decreases with an increased
forward bias, and the system is known as the forward a.g.c. A few circuits employ
reverse a.g.c., where the gain also decreases, but the base bias must then be reduced.
Both effects are discussed later in this chapter.

6.16 Mean Level A.G.C.

The principle of mean level a.g.c. is shown in figure 6.19. The video input signal to
the circuit may be derived either from the sync. separator or the emitter follower
or from the vision detector. As the input signal varies about its d.c. level, capacitor
C charges to the mean level of the video waveform. The time constant of the circuit,
R x C, is relatively long, so that the voltage on the capacitor is the average of many
picture fields. (The value of C is in the order of several J-lF, while that of R is a few
kil).
This type of circuit has several disadvantages. Firstly, it is affected by long term
variations in the transmitted [Link], as could occur, for example, when the scene
102 Television Principles and Practice

r------ --,
video ~RI a.g.c. ~ control
in I amp. ~ voltage
...J output

IG
L.. _ _ _ _ _

Figure 6.19 Principle of mean level a.g.c. Capacitor charges to the average (mean)
value of video waveform; resultant voltage may be amplified before being fed to
the i.f. strip

illumination changes from dark to bright. Secondly, the system is slow-acting, since
the control voltage is averaged over a number of fields in order to reduce the effect
of fluctuations in picture content. Thirdly, the mean level a.g.c. is prone to in-
correct action known as 'blocking', or 'lock-out', which could occur if the receiver
tunillg is switched from a weak signal to a strong one. Because of its slow response,
the a.g.c. allows a very powerful signal to be fed to the video amplifier, driving it
into saturation. The outcome of a 'lock-out' is a poor quality picture with a possible
loss of synchronisation. Finally, mean level a.g.c. has proved to be unsatisfactory
with negative picture modulation and intercarrier sound. This is because the a.g.c.
'sees' bright picture content as though it were brought about by insufficient signal
strength owing to the reduced d.c. level (see figure 6.5). Consequently, the gain
increases, with the possibility of the video signal breaking into the sound signal.

6.17 Peak Level A.G.C.

In peak level a.g.c. a capacitor is charged to the peak value of the video Signal. This
voltage corresponds to the tip of the synchronising pulse, which, in turn, coincides
with the maximum amplitude of the vision carrier. As the above reasoning implies,
this type of a.g.c. responds to the signal strength and is not affected by the picture
content. The time constant of the capacitor charging circuit can be made relatively
short, leading to a fast response to changes in signal strength-for example, aircraft
'flutter' can be prevented.
Two examples of peak a.g.c. circuit are shown in figure 6.20. In ciicuit (a) diode
Dl is the a.g.c. detector, which is biased so that it conducts during the most nega-
tive part of the video waveform-namely during the sync. pulse period. At the
same time the base bias resistor,.R2, of the transistor provides sufficient current to
drive TRI into saturation. Owing to the connections of diode D 1, polarity of the
'detected' signal stored on C1 causes the transistor to come out of saturation. Con-
sequently, the collector voltage of TRI is low in the absence of any signal, but it
increases with the rise in the amplitude of the received Signal. The output from
TRI is smoothed by the fIlter C3 , Rs , C4 and is used as the bias voltage for one of
the i.f. amplifiers. The point of conduction of D 1 and TRl, and hence, the gain of
the i.f. strip, can be varied by means of a separate d.c. voltage derived from the
contrast control.
Vision Detectors and Video Amplifiers 103

a.g.c. to
Lt. amp.

video in - - - _ - _ - _ - _.......--t-l

detector a.g.c. filter

amp.
(a)

- ......-~_-------4r---_-h.t.

video ___- - + - I .
in

...-_ _"--"....._ _ a.g.c. to

i.f. amp .

detector
_-- ---
.... ......
a.g.c. amp.
and filter
(b)

Figure 6.20 Peak level a.g.c. circuits: (a) a.g.c. diode detector and amplifier circuit
(GEe); (b) a.g.c. transistor detector and amplifier circuit (Decca)

The circuit in figure 6.20(b) uses transistor TR2 as the a.g.c. detector. Unlike
the preceding arrangement, in the absence of any signal TR2 would be biased off
by means of R3 connected between the emitter and the positive supply rail. When
the video waveform is applied to the base of TR2, the positive-going sync. pulses
overcome the initial reverse bias and the transistor conducts. The output from the
104 Television Principles and Practice

detector is fed to the a.g.c. amplifier circuit, TR3. This is a p-n-p' transistor which
conducts whenever [Link] made conductive by the sync. pulses. Hence, an increase
in the value of the video input signal causes both transistors to pass more current,
and the a.g.c. voltage increases in value. The control voltage is smoothed by C3 and
is 'clamped' to a minimum level (when TR3 is not conducting) by the potential
divider R s , R 7 • Similarly, the maximum voltage is also restricted by the potential
divider of R6 , R 7. Contrast control is provided in this circllit by means of VR1,
which, in conjunction with R 1 and R 3 , determines the point of conduction of TR2.

6.18 Gated A.G.C.

Gated, or keyed, a.g.c. is a va!iant of the peak level a.g.c. system. The chief dis-
advantage of peak level a.g.c. is that it can be affected by the field synchronising

a.g.c. delay
video and clamp
in circuit

line
flyback
pulses
(a)

video
in

,
a.g.c. gate smo~thing a.g.c. 1st J stage
filter amp.
(b)

Figure 6.21 Gated a.g.c. system: (a) block diagram; (b) circuit diagram (Bush)
Vision Detectors and Video Amplifiers 105

sequence and by interference pulses. Such pulses extend into the sync. pulse level
(negative picture modulation) and the control system might respond to their pre-
sence as if a strong signal were being received. The long time constant of the a.g.c.
detector and ftlter helps in reducing the effect of unwanted signal; alternatively,
low value ftlter capacitors can be used to remove interference from the a.g.c. detec-
tor-amplifier circuit.
Gated a.g.c. is used to overcome the above difficulties while maintaining fast
speed of response to any changes in signal strength. The system is designed to sample
the amplitude of the video waveform at specified intervals. With negative modula-
tion the sampling points occur in the sync. pulse period, which, as noted earlier,
corresponds to the peak of the vision carrier and, consequently, reflects the signal
strength.
A block diagram of gated a.g.c. is shown in figure 6.21(a). The a.g.c. gate is
biased ON by the line flyback pulses, so that it conducts only during the line sync.
pulse interval. The output from the gate is therefore proportional to the magnitude
of the synchronising pulses. Since the gate is open for only 15 per cent of the total
time, the possibility of interference breakthrough is also reduced. A ftlter is still
required, but, as the ripple is now at the line frequency, smaller values of capacit-
ance can be used, thus increasing the speed of system response.
A gated a.g.c. circuit is given in figure 6.2l(b), in which TRI is the gated tran-
sistor. This device will not conduct until a positive-going flyback pulse is applied
from a winding on the line output transformer via C 1 and Dl. The pulse at the
collector of TRI coincides with the negative-going, sync. pulse portion of the video
waveform at the emitter. Therefore the transistor conducts, allowing C2 to charge
to a voltage dependent upon the magnitude of the sync. pulses; this potential is
also affected by the amount of base bias given by VR1-a preset contrast control.
Components R4 and.C3 form a smoothing filter which, in conjunction with R s ,
applies d.c. bias to the a.g.c. amplifier transistor TR2. The a.g.c. voltage
appears at the collector of TR2 and is fed to the tuner r.f. amplifier via the delay
diode D2, and to the first i.f. stage (TR3) via the clamp diode D3. The operation of
this circuit is considered further in section 6.20.
Because of the relative complexity of the gated a.g.c. s¥stem, it is more popular
in the receivers employing integrated circuits (see figure 6.4).

6.19 Forward and Reverse A.G.C.

The a.g.c. potential, well smoothed and fully decoupled, is applied to the controlled
i.f. (and r.f.) amplifier to alter its gain. The change in amplification is caused by
two factors. Firstly, the gain of transistors depends upon the collector current;
devices are selected for a.g.c. control stages in which this effect is very pronounced.
Secondly, the amplifier gain varies with the collector-to-emiUer voltage-the lower
the voltage the lower the gain.
In practical circuits both effects are employed together, as shown in figure
6.21(b). In the i.f. stage, resistor R12 is fully decoupled to chassis via C6 ; when the
106 Television Principles and Practice

a.g.c. signal is applied to the base of TR3, the resultant change in collector current
alters the p.d. across R12 and changes the collector-emitter voltage of TR3. Readers
will, of course, note that the bias applied to TR3 consists of a d.c. potential sup-
plied by the resistor chain R s , R 9 , VR2, modified by the a.g.c. signal.
The effect of the a.g.c. voltage on the gain of the controlled stage can be seen
from the general curve relating the amplifier gain to the base bias current. The
curve is shown in figure 6.22, and it can be noted that there is a region of maximum
gain-if the bias is either reduced or increased, the result is a reduction in amplifier
gain. When the a.g.c. is applied so that an increase in bias causes the gain to fall, it is
known as forward a.g.c. Reverse a.g.c. exists when a reduction in bias reduces the
amplifier gain. The reader will observe that the two modes of a.g.c. operation corres-
pond to either side of the maximum gain shown on the curve.

reverse forward
.. a.g.c. a.g.c.
..

maximum gain
average
signal
strength

base bias current

Figure 6.22 Effect of amplifier bias on the stage gain

The most common form is forward a.g.c., as may be anticipated from the shape
of the curve in figure 6.22. The gradual and linear reduction in gain to the right of
the maximum suggests that the response is proportional to the applied control sig-
nal over a wide range.
With reverse a.g.c. the change in gain is very rapid and is also non-linear. This
form of control is sometimes used in tuner circuits, as it offers a slight advantage
over forward a.g.c. in u.h.f. amplifiers. The curve in figure 6.22 implies that reverse
a.g.c. does not produce a wide change in transistor currents and, consequently, the
voltages remain relatively constant. In particular, the base-collector potential differ-
ence is important, since this junction is reverse biased with the result that there is a
base-to-collector capacitance effect (generally similar to the action of a varicap
diode). It is, therefore, of some advantage at very high frequencies to maintain this
capacitance at a constant value.
Vision Detectors and Video Amplifiers 107

To decide which type of a.g.c. is used in a particular circuit, one must consider
the effect of a change in the signal strength upon the conduction of the controlled
stage. Returning, for example, to the circuit in figure 6.21 (b), the reader will observe
that an increased signal strength makes the sync. pulses more negative-going at the
input to TRI. This causes C2 to charge to a more negative potential, which, in turn,
reduces the conduction of the n-p-n transistor, TR2. As a result, the collector volt·
age of TR2 rises and so does the forward bias on the i.f. amplifier transistor TR3
(disregarding for the moment the effect of D3 and the associated circuit). Since the
object was to reduce the gain, but as the control signal increases the bias of the Lf.
stage, the arrangement represents forward a.g.c. Using a similar approach in the
case of circuits in figure 6.20, it can be seen that both examples also provide for·
ward a.g.c.

6.20 Direct and Delayed A.G.C.

It was mentioned previously that the a.g.c. signal is always applied to the Lf. strip, and,
in some receivers, it is also fed to the r.f. amplifier in the tuner. In most circuits the
Lf. amplifier is connected directly to the a.g.c. line, so that changes in the control
voltage are communicated to the given stage. For this reason the a.g.c. to the Lf.
strip is referred to as being direct.
In practice the range of change of the control voltage must be restricted. The i.f.
transistor must not be driven either into saturation on very strong signals or towards
the unstable region near the maximum gain (see the curve in figure 6.22). To avoid
this possibility, there are provisions for clamping the a.g.c. voltage to a fixed level.
Looking at the circuit in figure 6.21(b), it can be seen that a clamp diode, D3, is
used. Should the signal strength increase to such an extent that the voltage at the
collector of TR2 exceeds that at the junction of Rs and R9 (see section 6.19), the
diode becomes reverse biased. This leaves the bias of TR3 under the control of the
potential divider R s , R 9 , VR2, and any further a.g.c. action cannot take place.
A.G.C. applied to the tuner, on the other hand, should not become effective
until the full range of Lf. gain control has been exhausted. For this reason tuner
a.g.c. is known as delayed a.g.c.; the delay in this case is in terms of signal strength.
Such an arrangement is desirable, since the tuner mixer/oscillator ought to be pre-
sented with the largest signal possible in order to swamp the unwanted noise gene-
rated in this stage. Any reduction in the tuner amplification may take place only
when the Lf. gain is at its minimum and the signal would still cause overloading
(the symptoms being excessive contrast, poor synchronisation, interference patterns
on the screen and possibly a buzz on the sound).
In the circuit in figure 6.21 (b) the delay is performed by diode D2; normally it
is reverse biased because the collector of TR2 is at alower voltage than that of the
base of the r.f. amplifier. (The r.f. amplifier, not shown here, is fed from its own
potential divider within the tuner unit.) A very strong signal causes TR2 collector
voltage to increase to such an extent that D2 is biased ON. This voltage change is
then applied to the base of the relevant transistor in the tuner, reducing its gain
108 Television Principles and Practice

accordingly. The delayed a.g.c. ought to operate at the point when the clamp diode,
D3, switches off-a. suitable adjustment to the a.g.c. crossover is provided by VR2.
In other circuits the delay can be performed by a transistor which is normally
biased OFF. When a suitable a.g.c. voltage has been reached, it overcomes the
reverse bias and introduces an appropriate degree of tuner control.
Tuner a.g.c. is not always used; instead some form of manual control can be
adopted which performs a similar function by altering the base bias of the con-
trolled transistor.

6.21 Fault-Finding in Video Amplifiers and A.G.C. Circuits

The appearance of the display on the screen serves as a good indication of the
possible fault area, or the nature of the defect. Therefore, the picture may show
such symptoms as smearing (due to instability or to poor low frequency response),
lack of fine detail (poor high frequency response), incorrect contrast or brightness,
and so on.
A video amplifier fault can also result in a blank, but controllable, raster, suggest-
ing a defect in the path of the video signal (assuming other sections of the receiver
are in order). This can sometimes be localised by checking for the presence of
sound and sync. pulses with respect to their take-off points in the video stage.
In d.c. coupled video amplifiers (or directly coupled sections of this part of the
receiver) a fault can often affect the display by creating either a very bright, un-
controllable raster, or no raster at all. In either case the d.c. voltages of the tube
are upset; if the tube is cathode-driven, then a fault which cuts off the video output
stage raises the cathode voltage to h.t. and the beam current is also cut off. Con-
versely, when in the same arrangement the video output voltage falls to a very low
value, the beam current is a maximum, and frequently the operation of the bright-
ness control has no effect. The situation is reversed if the video output stage feeds
the grid of the tube-a high grid voltage results in a bright screen, while a low volt-
age cuts the tube off.
Similar remarks apply to colour receivers, except that in many cases it is only
one gun which is affected; colour casts are then produced of either a primary or a
complementary hue (see chapter 7).
A.G.C. fault-fmding can present difficulties because the system is based on a
feedback loop. Consequently, a defect anywhere within the loop can produce very
similar symptoms, and it is necessary to eliminate those sections which are unlikely
to be responsible.
The a.g.c. line voltage can be measured while either tuning to different channels
or unplugging the aerial. If a range of readings is obtained, and provided they are
within the makers' limits, the basic action of the a.g.c. circuit would appear to be
correct.
The a.g.c. voltage can also be clamped to a flXed d.c. level; the correct proce-
dure would be given in the i.f. alignment notes in the service manual. If this clears
the fault symptoms, then the a.g.c. detector-amplifier circuit could be responsible.
Vision Detectors and Video Amplifiers 109

An oscilloscope is useful, especially when servicing a gated a.g.c. system. Since

the waveforms involved are not symmetrical, they have a d.c. level, and a multi-
meter switched to a d.c. range is useful in measuring the resulting potentials.
Faulty smoothing and decoupling capacitors in the a.g.c. circuit may not neces-
sarily affect the d.c. readings, but they can produce various defects in the display.
For example, a rapidly fluctuating picture, or a display broken into wide
horizontal bars of varying intensity, could be traced to faulty capacitors.
7 Chrominance Signal Stages

7.1 Processing the Chroma Signal

The chrominance signal forms part of the composite video, from which it is sepa-
rated in the receiver and amplified before demodulation. It will be recalled (see
chapter 2) that the modulated chroma is without its carrier. The relative complexity
of the colour decoder is due to the necessity to reinsert the 4.43 MHz subcarrier.
The regenerated subcarrier must be of the correct frequency, and also in the
exact phase relationship with the transmitted signal. Owing to the quadrature
modulation of the U and V signals, there are two sub carrier feeds which are 900
out of pha~e with each other, and in the PAL system there is a 1800 phase change
from one line of the picture to the next. A suitably amplified chroma signal is then
detected using the sub carrier to produce the outputs proportional to the original
colour difference signals (R' - V') and (B' - V'). The third colour difference
(G' - yl) is formed from the proportions of the other two. Finally, the luminance
information, yl, must be added to the three colour difference signals. The resulting
signals produce light outputs from the screen of the picture tuberelated to the red,
green and blue content of the original televised scene.
The reader can refresh his memory by referring to chapter 4 (figure 4.1; also
section 4.5), where the block diagram of the decoder was included as part of a
complete receiver.
In modern receivers integrated circuits are used to perform most of the decoding,
so that it is even more important to appreciate the functions of the individual
blocks and the associated waveforms.
The block diagram in figure 7.1 expands the original arrangement shown in
figure 4.1; the waveforms are based on a colour bar signal (chapter 3), the ampli-
tudes being expressed in volts, peak-to-peak, which may be applicable to an average
receiver.

Figure 7.1 Block diagram of a PAL decoder of a receiver which uses R, G, B drive.
Amplitudes of typical colour bar waveforms are measured peak-to-peak; repetition
times, or frequencies, also shown as appropriate

110
delayed

if12v"lr B'
U64~s U output
amp.

B'

to picture tube cathodes (see figure 7.7)

112 Television Principles and Practice

7.2 Chrominance Take-Off Arrangements

The chrominance signal can be separated from the composite video by means of a
simple diode, whose non-linear characteristics make it act as a mixer. The output
from such a circuit consists of signals whose frequencies result from the subtrac-
tion of the signal frequencies at the input. The feed to the colour decoder is norm-
ally from the vision Lf. strip-for example, from the last i.f. amplifier or the vision
detector circuit. The vision Lf. at 39.5 MHz, when mixed with the chroma sidebands
corresponding to 35.07 MHz, generates the required signal centred around 4.43
MHz. The reader will recall (see chapter 5) that there is also one other Lf.-33.5
MHz-from the sound carrier. Therefore this will mix with both the vision Lf. and
the chroma Lf. to give rise to frequencies of 6 MHz and 1.57 MHz, respectively.
The former is the intercarrier sound Signal, and the latter is the chroma/sound beat;
both are unwanted in the decoder. In order to select only the 4.43 MHz signal,
tuned circuits are used to act as bandpass filters (often in conjunction with rejector
circuits resonant at 6 MHz or 33.5 MHz, depending on their position in the signal
path).
The chroma take-off point can be in the form of a separate diode circuit which
resembles that of the vision detector, its function being to produce the 4.43 MHz
signal. It is sometimes referred to as the chroma detector, but one must not con-
fuse this with the actual demodulators which will be found at a later point in the
decoder.
Since an ordinary vision detector can also produce the chroma beat frequency,
it is sometimes possible to find the take-off point early in the luminance channel-
e.g. immediately after the detector or at the emitter follower stage.

7.3 Chroma Amplifiers

Before any further processing of the chrominance signal can take place, it is essential
to raise its level from, say, 200 mY to about 2 Y. This is done in the chrominance
amplifiers, which may either be within an integrated circuit or in perhaps two or
three cascaded transistor amplifier stages. Each stage normally has a tuned circuit
resonant at 4.43 MHz and giving the necessary overall bandwidth of 2 MHz. To
secure the optimum response, correct alignment of this section is important. The
exact procedure can only be obtained from the makers' manual, since the response
curve here may also be dependent upon the alignment of the vision Lf. strip. The
principle, however, follows the outline given in the wobbulator method in chapter
5; the details, however, must be adapted to suit the decoder.
The first chrominance amplifier usually offers variable gain to compensate for
the strength of the signal at this point. The d.c. biasing can be altered by means of
the automatic chroma control (a.c.c.) potential. Its action is similar to the a.g.c. in
vision Lf. amplifiers, and the reader is referred to chapter 6 for full discussion of
this topic. The derivation of the a.c.c. voltage will be described later, in section 7.7.
In order to prevent coloured 'noise' appearing on the screen, it is desirable to
Chrominance Signal Stages 113

to contrast
negative - going control
line flyback
pulses

from 1st Dl D2 to 2nd

chroma - ••- - t t---.--f'I__- -. .-..----III----.~chroma
amp. amp.

CHROMA AND CHROMA WITHOUT

BURST INPUT BURST OUTPUT

Figure 7.2 Colour control and burst blanking circuit

shut down the chroma path during a monochrome transmission. By removing the
d.c. bias from one of the chroma amplifiers, the signal is blocked at this stage. A
separate circuit, called the colour killer, is responsible for developing a suitable bias
only during colour reception. The action of the colour killer also removes the
effects of a number of decoder faults from the display, so that the viewer is left
with a good monochrome picture.
As mentioned above, the gain of one of the amplifiers is controlled automatic-
ally, but the final balancing between the chrominance and the luminance is carried
out by means of a viewer's control called COLOUR or SATURATION. In effect, its
setting governs the amplitude of the composite chrominance signal, which, in turn,
affects the demodulated colour difference signals and, hence, the saturation of the
displayed colours. The adjustment is normally subjective: in an otherwise correctly
set-up receiver the aim is to achieve a pleasing reproduction of 'skin tones'. The
possible circuit arrangements are similar to those of contrast control in video
(luminance) amplifiers (see chapter 6). Examples of saturation control methods are
given below.
(1) Varying the gain of one of the chroma amplifiers-for example, by altering the
a.c.c. bias.
(2) Varying the amount of chroma fed from one stage to the next; a simple variable
resistor can be used or, alternatively, a diode attenuator of the form shown in
figure 7.2. Diodes D1 and D2 are connected 'back-to-back' with a variable supply
114 Television Principles and Practice

fed to their junction via R I • Now a diode offers a resistance to the signal, but its
value depends upon the actual amount of applied d.c. bias-low bias current means
high resistance, and vice versa. This effect is due to the non-linear characteristics of
the diode. Unfortunately, the non-linearity could also distort the signal; that is why
there are two diodes connected in opposition to cancel out the distortion. This
circuit avoids problems that may arise with the routing of the high frequency signal
from the decoder panel to the remote control.
It is desirable to maintain the relative amplitudes of the chroma and the lumin-
ance signals constant to preserve their transmitted ratio; therefore the colour and
the contrast control may be coupled electrically or mechanically. Changing one of
them will have a similar effect on the other.
The diode attenuator circuit in figure 7.2 can also be used for the removal of the
colour burst from the composite chroma. At this stage the burst has no further use,
and its continued presence may have disadvantages. The burst, if demodulated,
could cause a spurious colour patterning on the left-hand side of the screen. The
demodulated burst could, in some cases, affect the performance of the black level
clamp. In figure 7.2 negative line flyback pulses derived from the line output trans-
former are fed to the junction of D1 and D2. Their amplitude is sufficient to cancel
out the forward bias from VR1, causing the diodes to be reverse biased. The chroma
path is shut off for the duration of the pulse, which, in turn, is arranged to coincide
with the timing of the burst.

7.4 PAL Delay Line

It has already been mentioned that the PAL system, although based on the American
NTSC, improved the quality of the received colour picture. This is due to the alter-
nation of the phase of the V signal from line to line.
We already found in chapter 3 how different colours contain their specific
values of (R' - Y') and (B' - Y'); as shown in the phasor diagrams in figure 7.3,
these two components give a resultant chroma of a specific phase angle with respect
to the horizontal reference. If this angle changes, the (R' - Y') and (B' - Y') com-
ponents also alter, giving rise to a different hue. Should such a change be spontan-
eous in the transmitter-receiver chain, then the resultant colour is incorrect because
it differs from that seen by the camera.
Let us assume that the phase angle is increased slightly owing to the various
imperfections in the whole system. As shown in figure 7.3, during one line when
the V signal is unswitched (diagram a), the result would be an increase in the
(R' - Y') component and a decrease in the (B' - Y'). During the next transmitted
line, when the V signal is switched through 180 0 (diagram b), the additional phase
angle reduces the inclination of the chroma phasor, so that the (R' - Y') is now
reduced but (B' - Y') is increased. It must be pointed out that after the demodula-
tor the (R' - Y') signal of the switched line is reversed so as to coincide with the
(R' - Y') of the previous line. Instead of the desired chroma, we could get one line
Chrominance Signal Stages 115

unswitched line 'error' chroma

U axis

x
co unswitched
'x'"co
>- line >
I
I correct
a: chroma
> U+V

switched line correct

chroma U - V
U(8'- V') axis
(a) (b)

error chroma on

----\\---71
unswitched line

~~~~~: field one

Rt }\
• i 11
'"x
-----J-- .
i - - - A"1
I /1
field two
(interlaced)

/ .V\\.: ~~:ws itched

co
> { - - - - - - - - -

• ~I I sWitche( - -- -- -------}/
/ ./ lerror chroma l lines
/ /' ~ laft:r PAL :

\{===========
•• / Iswltch I
I I
,
U axis
(c) (d)

Figure 7.3 Chrominance phase errors. (a) Effect of phase error on the phasor
diagram of unswitched line (U + V); note that the U and V components are now
different, suggesting a change in the reproduced hue. (b) As above, but for the
switched line (U - V). (c) Mter PAL switching, the V component from diagram
(b) is reversed, and the two consecutive lines correspond to different colours.
(d) Hanover blinds produced on the screen are in pairs oflines, owing to scan inter-
lacing. The continuous lines correspond to the chroma detected during unswitched
lines; the broken lines are produced by the chroma on alternate lines

of one colour followed by a line of another colour. Because of the interlaced scan-
ning the lines of the second field, similarly affected, will fall in between those of
the previous field. The effect on the screen will be a display of coarse horizontal
lines of two different colours. These lines are known as 'Hanover blinds' or 'Hanover
116 Television Principles and Practice

bars' (the PAL system, which is prone to this defect, was invented at Hanover in
Germany).
Seemingly, we have not yet improved the quality of the display at all. In fact,
it is possible to cancel out the effects of the errors in the phase relationship between
the signals comprising the composite chroma. The process of cancellation can take
one of two forms: optical and electronic.
Optical cancellation. The Hanover blinds, as described above, are allowed to be
displayed on the screen. However, the interesting point is that, from a normal view-
ing distance, the different colour lights produced by these bars tend to merge to
create the impression of the original hue, but somewhat desaturated (pale). This is
a rare method of operation, except for test purposes or under certain fault condi-
tions. We can visualise that if such phase errors were relatively large, the two sets of
bars would be of so vastly different colours that the optical averaging would not be
effective, giving rise to most objectionable line structure. When the receiver does
rely on this system, it is called simple PAL, or PAL-So
Electronic cancellation. The chrominance signals corresponding to two consecu-
tive lines are matrixed to produce two outputs. These outputs are equal to the sum
of and the difference between the chroma signals during the switched and the un-
switched lines, respectively. To do this we have to store, or delay for 64 I1S, the
signal received during the previous line, so that it can then be processed together
with the information arriving during the current line period. The more precise value
of the delay is 63.943 I1S, which is slightly less than one line interval because it
must correspond 10 a whole number of the colour subcarrier half-cycles (567 in
this case). On the other hand, the slight difference from 6411S does not affect the
quality of the picture. A block diagram of a circuit used in this method is shown
in figure 7.4(a). This system is known as PAL-D (de-luxe, or delay line).
The delayed signal is the preceding line. Let us assume that it is unswitched; then
the composite chroma =(U + V). The current line must be with its V component
reversed, so that chroma = (U - V). The addition of the two yields
(U + V) + (U - V) = 2U
the corresponding subtraction gives
(U + V) - (U - V) = 2V
At this stage we can see that the method separates the two quadrature signals
(the increased amplitude does not change the principle). Indeed, this is a beneficial
by-product of the process, although correctly adjusted demodulators are also cap-
able of separating the respective colour difference waveforms.
The cancellation of phase errors can be proved by the phasor diagram in figure
7.4(b). Taking the two chroma signals which occur if phase errors were present, as
shown in figure 7.3(c), we find that after the delay line and the ADD/SUBTRACT
matrix, the U and the V components will add separately in the manner shown in
the abC',:!! equations to give the output signals. The resultant chroma formed from
the two pairs of components lies along the correct chroma phasor; therefore its
Chrominance Signal Stages 117

--- ADD r- demodulator

2U to (B'-Y')

-
U+V
~
from chroma
amplifiers delay
line
PAL
delay i
....
driver line

~2V to (R'-Y')
SUBTRACT "- demodulator
amplitude
U-V
direct chroma
and phase
adjustment
-
path

(a)

difference in phasor length =

reduced saturation correct chroma after
"'-/.. _ADD/SUBSTRACT
2U+2V 1/
unswitched line
.... ;'
.... ;,11'
/ I
'error' chroma after
ADD/SUBTRACT
.,

>
..
)(
_;'
.... ......... /
/
/
I
I

/ I
/ I
/ I
I
I

" - switched line 'error' chroma

after PAL SWITCH

U axis

(b)

Figure 7.4 Effect of PAL delay line on phase errors: (a) block diagram of PAL
delay line circuit; (b) phasor diagram showing the result of electronic averaging; the
U and the V components of chroma add in each case (figure 7.3 shows how the
actual chroma phasors were obtained)

phase is unchanged, but the amplitude is slightly less than it would otherwise be.
This means that the ratio of U to V is constant, giving the same hue of slightly
reduced saturation (see chapter 1 for the definitions).
The chrominance delay line is a separate circuit component made of special glass
to very critical dimensions, with two piezoelectric transducers attached to it. The
118 Television Principles and Practice
chroma input signal path inside
signal delay line

---
--- _ polished
surface

delayed chroma
output signal

(a)

h.t.
r- ------ --- - ----,
I I 2V to
I r---~-T--~~----~ R'-Y'
demod.
I
I R6
I 641.1s
I to B'-Y'
delay line
I 2U demod.
I
IL ______________ ...JI
chroma from
precEiding
amp.

c,

(b)

Figure 7.5 PAL delay line: (a) principle of construction; (b) example of the
associated circuit

simplified diagram in figure 7,5(a) illustrates the principle involved. One of the
transducers converts the incoming electrical signals into mechanical vibrations,
which spread through the glass body to be picked up by the second transducer. The
latter reconverts them back into an electrical signal (the idea is somewhat similar to
that of a record player cartridge!).
The speed of propagation of these vibrations in the glass is relatively slow, and
therefore the required time delay in the signal can be achieved. The type of delay
Chrominance Signal Stages 119

line used in the luminance section of the receiver (see chapter 6) would be imprac-
tical, as the delay now is almost a hundred times longer.
There is a loss of signal power in the delay line, which must be made good by a
separate amplifier stage-a delay line driver (occasionally an amplifying stage may
follow the delay line).
A typical circuit diagram of the arrangement shown in block form in figure
7.4(a) is given in figure 7.S(b). The delay line is supplied with an amplified signal
from the collector of the driver transistor TRl, while the direct chroma path is
taken from the emitter. The transducers are connected to the remainder of the
circuit via autotransformers which are usually an integral part of the delay line unit.
They provide impedance matching, and at the same time the output autotransfor-
mer is used as an adding and subtracting circuit. Being a centre tap coil, the winding
produces two outputs of equal amplitude but in phase opposition. The direct
chroma is connected to the centre tap, so that it adds to the delayed signal in one
half of the winding and subtracts in the other half. There are now two outputs from
the outermost terminals of the transformer, which correspond to the separated U
and V signals in the manner shown above. The reader will appreciate that the correct
addition of the delayed and the direct chroma only takes place if the amplitude
and phase of each signal at the point of matrix are equal. Since the signals take
two different paths, adjustments are provided;Ll is primarily used to equalise the
phase and VRI is chiefly used for amplitude compensation. Sometimes the coupling
coils to the delay line itself have adjustable cores to achieve the necessary equalisa-
tion. If the settings are incorrect, instead of cancelling the errors a misadjusted
circuit would introduce them, and the display will produce Hanover blinds.
If the delay line develops a fault, the receiver continues to function, since the
direct chroma path is still available, but Hanover bars are again likely, because the
decoder reverts to the PAL-S mode. Sometimes it may even be desirable to check
the adjustments of the demodulator feeds or the subcarrier oscillator with the
receiver PAL delay line being inoperative (the process of electronic averaging can
mask the defects in the demodulating action).

7.5 Synchronous Demodulators

The reader would have already come across the idea of synchronous demodulation
described in chapter 6 in connection with video detectors. There the process was
desirable, but not strictly necessary, since a simple diode detector will operate
reasonably well. The chrominance signal presents an entirely different problem,
because of the suppressed subcarrier. As outlined in chapter 3, a simple diode detec-
tor would produce an output which bears no resemblance to the original informa-
tion; hence, synchronous demodulation is absolutely necessary. This process requires
a sampling circuit which 'looks' at the signal at the time when the transmitter sub-
carrier passes through its maximum amplitUde. The demodulator behaves as though
it were a switch which closes at that instant, and transfers the sampled amplitude
to the output. The Switching action must be synchronised with the transmitter
120 Television Principles and Practice

colour subcarrier in order to ensure that the output waveform is of correct shape.
The sampling process is steered by the subcarrier regenerated in the receiver, and is
synchronised with the transmitted signal by means of the colour burst.
The composite chrominance signal consists of the U and V components. These
were originally modulated by means of what effectively appeared to be two carriers
of the same frequency, but 90° out of phase with each other (quadrature modula-
tion). The delay line circuit can separate the two components, which doubtless
helps in improving the action of the demodulators and simplifying somewhat their
design. But it has to be emphasised that a correctly adjusted synchronous demodu-
lator can also separate the required component from the composite signal, as would
be the case in the PAL-S receiver. Quadrature modulation means that one subcarrier
feed is reaching its maximum amplitude while the other one is passing through zero.
It follows that the demodUlator will 'see' only the maximum value when it is being
switched by its own, appropriately synchronised, subcarrier, resulting in only one
colour difference signal output. When the sub carrier amplitudes change a quarter
of a cycle later, it will be the other demodulator which produces an output corres-
ponding to its required signal while the previous demodulator is switched off.
Practical circuits are relatively simple because the switching subcarrier has already
been obtained in another part of the decoder. (In a video synchronous detector,
the switching carrier had to be produced from the modulated signal itself, so that
it was economically feasible only by means of an I.e.)
Two arrangements are shown in figure 7.6. In diagram (a) diodes D1 and D2 are
simultaneously switched by the sub carrier , whose phasing has been correctly adjus-
ted to ensure synchronisation with the relevant feed in the coding circuit at the
transmitter. When the diodes conduct, the chroma coupling capacitor C2 will charge
to the amplitUde of the chrominance signal at the time. During the next half-cycle
of the switching sub carrier the diodes are reverse biased and are non-conducting.
The voltage acquired by C2 appears at the output across R 3 • The parallel LI JC I is
a filter circuit, which removes the subcarrier, so that the output waveform is in the
shape of the modulating envelope. Resistors RI and R2 are in series with the diodes
to limit the current during the Switching operation.
In diagram 7.6(b) a four-diode circuit is used to act as a switch. All four diodes
conduct Simultaneously when the junction of Dl and D2 is positive with respect to
the junction of D3 and 04. On inspection, the reader will notice that the bridge
circuit is placed in series with the signal path, which means that when the diodes
conduct, the bridge circuit looks as though a switch were closed to allow chroma
to pass through it to charge capacitor C2 • During the next half of the subcarrier
cycle the diodes are reverse biased, and the voltage from across C2 is transferred via
the filter to the output.
The diagrams shown refer to (B' - V') demodulators, the (R' - V') counter-
part being identical; the only difference is the arrangement of the sub carrier feed
between the oscillator and the respective demodulators to accommodate the 90°
phase shift as well as PAL switching. This aspect of the circuit will be discussed in
later sections.
Chrominance Signal Stages 121

R, D1
T1
colour

~
subcarrier f -_ _---(8'-Y') OUTPUT
from to amp.
oscillator

U signal from
delay line circuit
.. (a)

U signal
from delay -----+--< ~-r----i~ (8'--Y') OUTPUT
to amp.
line circuit

n
T1

colour subcarrier
from oscillator

(b)

Figure 7.6 Examples of synchronous demodulator circuits: (a) two-diode circuit;

(b) four-diode circuit

7.6 (G' - Y') and R, G, B Matrix

As mentioned in earlier chapters, the third colour difference signal, (G' - Y'), is
not transmitted, since it can be 'reconstructed' in the receiver from the specified
portions of the demodulated (R' - Y') and (B' - Y'). This is done in the circuit
called the (G' - Y') matrix. Basically it consists of an amplifier which receives
signals from the other two colour difference amplifiers. The (R' - y') and (B' - y')
amplifiers are fed with signals from their respective demodulators, their gains being
unequal, to compensate for the weighting factors.
In practice, the majority of receivers use R, G, B drive; therefore the luminance
and chrominance paths are combined before being fed to the tube. This is done in
the stage called the R, G, B matrix. One part of the circuit acts as the (G' - y')
matrix, while the other part is the luminance/colour difference matrix; here the
three primary colour signals-red, green and blue-are produced. The reader should
now refer to chapter 6 and look at the circuits in figures 6.11 and 6.7. In figure
122 Television Principles and Practice

6.11 the unequal value resistors Rs and R12 determine the specified proportions
of the (R' - V') and (B' - V') needed to form (G' - V'). In this way the
(G' - V') input is applied to the emitter of TR2, and the amplified output
becomes available at its collector.
If we look at the emitter circuits of TRI and TR3 (figure 6.11), the (B' - yI)
and (R' - yl) amplifier transistors, we note that the total resistance in each case is
390 on or nearly so (R3 + R4 and R 1S ); these values are required to maintain
correct d.c. conditions. However, the gain of the (B' - yl) amplifier is made greater,
since the 220 on (R4) of the emitter resistance is bypassed by the capacitor C2 • In
the emitter of the (R' - yl) amplifier R 15 is unbypassed and the resultant negative
feedback reduces the voltage gain. These differences compensate for the unequal
weighting factors in the U and V signals.
The luminance signal is introduced in figure 6.11 at the three collectors via
R1 jC1, R 9 jC4 , R 14 jCS ; this signal is added to the amplified colour difference
signals at these points to produce the blue, green and red waveforms. The signals
at these points are of an insufficient magnitude to drive the tube, and must be
amplified. A simple output circuit was shown in figure 6.7, together with the
associated black level clamps; a full description was given in the accompanying text.
It is interesting at this point to look at the R, G, B drive waveforms based on a
100 per cent saturated colour bar signal, whose detailed description was given in
chapter 3. The reader will recall that the colour sequence is white, yellow, cyan,
green, magenta, red, blue and black. To produce the corresponding light output,
the cathodes of the picture tube must be driven by the waveforms in figure 7.7. As
discussed in chapter 6, the video signal is negative-going; in other words, the sync.
pulses must take cathodes towards h.t., leaving the control grids at a fixed lower
potential. A falling cathode voltage during normal picture information reduces the
net bias between the tube electrodes and the beam current increases in proportion.
In view of our earlier discussion in chapter 6, the reader will accept the concept
of peak white when all three guns are driven to their maximum output, as shown
in the white bar.
To produce a yellow colour, only the red and the green (see chapter 1) electron
guns. are driven; the blue is switched off at its black level.
Cyan requires blue and green drive, with the red cathode at its black level.
The red bar is the result of the red gun alone being driven, the other two being
at their black level. The reader is invited to verify the appearance of the three wave-
forms for the remaining colours: green, magenta, blue and black.
There are small 'notches' shown between the adjacent 'ON' parts of the wave-
forms; these may be seen on the screen of an oscilloscope when one views the drive
waveforms. They are the resultant of the addition of signals in the R, G, Bmatrix,
but the circuits do not always respond sufficiently fast, and the transition is delayed.
On the screen it often shows up as a darker band between the adjacent bars.
Waveforms of similar shape may be found at different points between the R, G,
B matrix and the tube cathodes; their direction may be inverted, depending on
Chrominance Signal Stages 123

l!l
.~
;:
.2 c: .,c: .,
c:
., .>t.
a; '" '"
'"E :c" :c'""
-0
.<:
;: > l!! l!!
>
" '"
sync.
pulse

(a) Red
cathode

(b) Green
cathode

(c) Blue
cathode

Figure 7.7 Cathode ray tube drive waveforms; R, G, B cathode drive, 100 per cent
saturated colour bars

whether the amplifying stages reverse the polarity ( common emitter) or not (emitter
follower).
The waveforms in figure 7.7 are based on theoretical considerations, but in
practice various overshoots in the vicinity of the sync. pulse may be displayed on
the oscilloscope (depending on the behaviour of the actual receiver circuitry). For
practical purposes these may be ignored, because the flyback blanking (chapter 6)
suppresses any effects of spurious pulses on the actual picture. It has been men-
tioned earlier that, for simplicity, all the waveforms illustrated in this book refer to
100 per cent saturated colours. Some colour bar test signals offer waveforms corres-
ponding to reduced saturation; if this is the case, then, whenever the gun is shown
124 Television Principles and Practice

as switched off, there is, in fact, a drive signal present (except for black, of course),
its amplitude being well below the peak level. The reason is that a desaturated
colour contains some white light in addition to its own hue (chapter 1), but to pro-
duce white we must have all three guns at least partly switched on. Many service
manuals may show such desaturated waveforms instead of the examples given here.
The illustrations in this book suggest that the amplitudes of the signals are
equal in each case. As will be shown in chapter 11, the actual drive required is
different for each gun and is adjustable as .part of so-called grey scale tracking (see
VR2 in figure 6.7). However, the peak-to-peak amplitude of the signal, although
dependent upon the size of the tube screen and other constructional details, is
typically between 100 and 150 V.

_ . . -_ _ _ _
28_5_V_ _ _-.-_ h.t.

(R' - Y') to control

t-----. grid of picture tube
pulses to
grids of
remaining
-(R'-Y') from triodes
preampl ifier

~---+_ _ Ilne flyback

pulses

R,
2.2M to cathodes of
remaining triodes

Figure 7.8 Colour difference output stage (Mullard)

A few receivers of earlier design feature a valve colour difference drive circuit, of
the type shown in figure 7.8.
Preceding the valve is a transistorised preamplifier stage, which also includes the
(G' - V') matrix. In that respect the circuit follows the basic principle outlined
above. Valve VIa, the pent ode section of a double valve, acts as an amplifier to
raise the signal voltage level from about 7 V to, say, 200 V. The output signal at
the anode of VIa is coupled via C2 to the anode of VI b, and from there to the
control grid of the picture tube (the red gun in this case). The triode section, Vlb,
is a black level clamp whose control grid is driven by line flyback pulses supplied
from a suitable winding of the line output transformer.
Owing to the complexity of the preceding decoder, by this point in the circuit
Chrominance Signal Stages 125

the colour difference waveforms would have lost their true zero reference level,
and this must now be restored. It is essential not only to preserve the black level
and, hence, the average d.c. level of the luminance (or video) signal (described in
chapter 6), but also "to maintain a correct reference point for each colour difference
signal. Otherwise a situation arises in which the drive magnitude to the respective
grids could be either excessive or insufficient; this results in either primary or com-
plementary colour casts appearing over the entire picture (even to be present on a
monochrome programme). To ensure a correct 'black level', a large positive pulse
(typically 50 V) is applied to the grid of the triode which coincides with the begin-
ning of each picture line (during back porch when the colour difference signal is
zero). The valve is driven into saturation, and its anode voltage drops to that of the
cathode, which is fixed by the potential divider R 9 , RIO, Rs. This allows the capaci-
tor Cz to charge to a predetermined voltage from the h.t. supply. During the picture
information period VI b is cut off, and the relevant grid of the c.R. T. is held at a
direct potential given by the charge stored in Cz ; the colour difference signal being
superimposed on this level. Capacitor Cz discharges only very slowly via the high
value resistor R 4 , so that the reference level is substantially constant until the
arrival of the next clamping pulse. In some circuits the cathode voltage of VI b is
made variable to form part of the brightness control arrangement.
The circuit shown in figure 7.8 is for the (R' - V') output stage; the remaining
two colour difference output amplifiers are identical, the cathodes of the clamp
triodes being connected together and the line pulse feed is also shared.
The luminance part of the receiver using colour difference drive was discussed
in chapter 6, and the corresponding luminance output stage was illustrated in
figure 6.12. The matrixing of the luminance and colour difference signals takes
place within the picture tube, whose light output is proportional to the voltage
difference between the control grid and the cathode. If the grid is driven by
(R' - V') and the cathode is driven by (-V') (minus sign denotes negative-going
video), the net signal applied to the tube is (R' - V') - (-V') = R'; similarly for
the other two guns.
The waveforms applied to the grids of the tube are of the now familiar shape
(see chapter 3) of (R' - V') (typically 180 V peak-to-peak), (B' - V') (220 V) and
(G' - V') (100 V). The associated d.c. voltages at the grids are then in the order of
120 V, and originate from the triode clamps described earlier.
R, G, B matrixing and other functions described here can also be performed
within an integrated circuit, which often includes other forms of signal processing.
We [Link] at this type of decoder in a later section.

7.7 Regeneration of the Colour Subcarrier

The subcarrier associated with the chrominance signals was suppressed at the trans-
mitter but, as we saw in section 7.5, the subcarrier is necessary to operate the
synchronous demodulators and it must be regenerated in the receiver. This is done
using the colour burst, whiCh is a transmitted sample of the orjginal sub carrier
126 Television Principles and Practice

subcarrier 7.8 kHz

burst to ripple to
a.c.c. circuit ident amp.

composite d.c.
chroma amp.

reactance
gating stage
pulses

4.43 MHz
subcarrier
to demod.

Figure 7.9 Block diagram of colour subcarrier regenerating circuit (a.p.c. loop)

during the back porch period-that is, between the line synchronising pulse and the
beginning of picture information.
The purpose of the colour subcarrier regenerating circuit is to control the fre-
quency of a crystal oscillator; a block diagram of the associated arrangement is
shown in figure 7.9. The burst gating circuit conducts only during the application
of a delayed pulse which coincides with the period of the burst. Usually the pulses
are fed from the line output transformer, and are due to the line flyback that occurs
in time with the synchronising pulse. Therefore they must be suitably delayed to
arrive during the back porch period. The precise method used to shape and delay
them varies from one receiver model to another. Some manufacturers use a com-
bination of resistors and capacitors; others use a tuned circuit which is shocked
into self-oscillation (ringing) by the initial pulse, the frequency of the oscillation
being such that the second half-cycle occurs at the required time to open the burst
gate. It is also possible to use the line synchronising pulse derived from the sync.
separator and suitably delayed.
The diagram in figure 7.10 shows a simplified arrangement of the sub carrier
oscillator and its control circuit. TRI performs the dual function of a burst gate
and a burst amplifier. The base bias on this transistor is not fixed, and it is normally
non-conducting except for the brief period of the positive-going, delayed line fly-
back pulse. The delay takes place in another part of this receiver by means of a
resistance-capacitance network. The diodes, Dl and D2, ensure that the magnitude
of the pulse does not exceed the 25 V supply by clipping any excess; resistors R 1
and R2 act as a base potential divider. The leading edge of the gating pulse is
delayed, and in this receiver the length of the pulse is also controlled to ensure that
the conduction of TRI does not extend beyond the duration of the burst. The
 C 10
subcarrier
to demod.

delayed line
flyback pulses
..

Cg

pulse shaping burst gate phase d.c. amp. reactance subcarrier oscillator
and timing and amp. comparator stage and amplifier

composite 7.8 kHz ripple

chroma input to ident amp.

Figure 7.10 Burst gate and automatic phase control loop (Thorn)
128 Television Principles and Practice

timing circuit consists of D3, C1 , D4 and VRI. Capacitor C1 charges from the
gating pulse via D3 and VR1; during that period of time, the voltage drop across
VRI holds D4 biased OFF, and the conduction of TRI takes place as described
Once C1 is fully charged, the current through VRI and the resultant p.d. are re-
moved; D4 then conducts to divert the remainder of the gating pulse to earth, when
the transistor switches off.
All gating circuits must ensure that the gate is reliably 'shut off' when not
required. This is done here by means of a charge stored in C2 , which applies a
positive potential to the emitter of TRI of sufficient magnitude to prevent the
remainder of the chroma signal (or spurious noise) from turning the transistor on.
The tuned circuit, L 1 , C3 , resonates at 4.43 MHz, which selects the burst signal
and rejects the relatively low frequency gating waveform. With an oscilloscope
connected to the collector of TRI the tuning of L 1 can be adjusted to obtain a
burst of maximum amplitude. The gated burst is then applied to the discriminator
circuit, which consists of DS and D6 (burst discriminator diodes) together with
C4 , Cs ,R s ,R 6 and C6 •
The burst switches the discriminator diodes on, and the coupling capacitors C4
and Cs become charged to equal but opposite polarity potentials. The voltage at
the output from the discriminator Gunction of Rs and R 6 ) is then zero by virtue
of circuit symmetry. However, a sample of the subcarrier oscillator output is fed
back from TR3 via C10 to the junction of the diodes. Now the circuit compares
the phase and, indirectly, the frequency of the oscillator with the incoming burst.
If the regenerated subcarrier passes through zero at the time of maximum burst
amplitude, which is when DS and D6 conduct, the original symmetry of the arrange-
ment is preserved and there is still no output from the discriminator. Should the
phase of the oscillator output change, then the signal fed back biases one of the
diodes more than the other, and the charges on C4 and Cs become imbalanced.
Either a positive or a negative polarity output now appears across C6 • If the oscilla-
tor frequency is incorrect, then similar action takes place, because the subcarrier
zero would be out of step and would be drifting with respect to the received burst.
In both cases the resultant d.c. or slowly changing d.c. (of either polarity) is
amplified and used to adjust the oscillator via a reactance stage until the error
baSically disappears.
For simplicity, we have assumed that the transmitted burst remains constant,
but the reader will recall that the PAL system uses a swinging burst, which means
that its phase is alternating for every picture line. Naturally, this means that the
discriminator always produces an output which also alternates in step with the
phase of the burst. Obviously, the oscillator output, even if it were meant to follow
the burst faithfully, would be out of step at the beginning of each line. A useful
compromise solution exists: the sub carrier oscillator runs at a constant phase
relationship with the original transmitter subcarrier oscillator, and it must ignore
the regular, line-by-line alternations of the burst. However, the 'swinging' phase is
also detected and employed in a separate circuit to identify the switched and the
unswitched lines of chrominance information.
The alternating discriminator output must be smoothed before it is used to con-
Chrominance Signal Stages 129

trol the frequency of the oscillator; the smoothing must be carefully arranged to
allow some 'swinging' discriminator signal to be directed to the PAL identifying
circuitry (so-called ident).
In the circuit in figure 7.1 0 fIltering takes place in two stages. The field effect
transistor, TR2, is a so-called d.c. amplifier, but, in fact, it has to handle the alter-
nating component as well. Its amplitude is first reduced by a.c. negative feedback
introduced by R7 and C7 . The resultant 7.8 kHz ripple is then sent to the ident
amplifier, and is considered in the next section of the book. After the ident take-off
point, final fIltering can take place by means of R12 and Cs .
The reactance stage is formed by the varicap diode D7, whose capacitance is
altered by the applied control voltage. Resistor R13 prevents the 4.43 MHz signal
from being shunted by the remainder of the circuit. The capacitance of the diode
is in series with the self-capacitance of the crystal (Xtal), allowing the resonant
frequency of the circuit to be changed. The quartz crystal is connected to a trans-
ducer (see section 7.4); the frequency of vibration of the crystal depends on its
physical dimensions and offers a great stability of operation. These vibrations are
converted into electrical signals, and the whole arrangement behaves as if it were a
tuned circuit.
The oscillator is based on the Colpitts circuit; the familiar capacitor 'tap' is now
at the junction of C9 and Cll • The output is amplified by means of the tuned cir-
cuit L2 in parallel with stray capacitance; the inductance is adjusted (with the aid
of an oscilloscope) to produce maximum subcarrier amplitude.
The frequency error that may be corrected is limited to a few hundred Hz, which
means that if the free-running oscillator frequency (without the burst input) is too
far from the required 4.43 MHz, the circuit will not be able to synchronise at all
when the burst is applied. This could happen, for example, when the programme
changes from monochrome to colour, or another channel has been selected, or
simply when the receiver has just been switched on. To ensure correct operation,
the amount of d.c. applied to the reactance stage can be adjusted, as if to simulate
the action of the error signal. In the circuit in figure 7.10 this is done by controlling
the amount of d.c. bias of the FET by means of VR2, while the original burst
signal is temporarily reduced in amplitude or even removed. When this adjustment
is being carried out, the makers' instructions must be followed, some requiring the
adjustment to aim at a specified voltage; in other circuits the colour killer may
have to be overridden (see next section) and the display on the screen observed.
If the oscillator frequency only differs slightly from the colour burst, then the
picture will be displayed by the luminance signal, but the colours will slowly drift
across the screen, creating an effect of continuously changing hue. Should the
oscillator be considerably out of adjustment, then the display breaks up into a
drifting and meaningless coloured pattern, often obliterating the picture completely.
The output from the oscillator is usually fed to a buffer stage (an emitter
follower) to limit the loading on the circuit; sometimes, as in this case, a transformer
coupling (tap on L 2 ) is used to achieve the required impedance matching to the
driven circuits.
The complete circuit formed by the burst phase discriminator, the d.c. amplifier
130 Television Principles and Practice

and the oscillator is closed by the feedback connection from the output back to the
input (via C IO in figure 7.10). This is often referred to as the a.p.c. loop (automatic
phase control loop).
In most receivers automatic chroma control (a.c.c.) voltage (see also section 7.3)
is derived from the burst gate amplifier. Part of its output is rectified, sometimes
by means of a voltage doubler, smoothed and fed to the first chroma amplifier to
alter the d.c. bias and, consequently, the gain. The colour burst is used for this
purpose because its amplitude is representative of the actual strength of the
chrominance signal. As with the a.g.c., we need a control voltage which is indepen-
dent of picture content.

7.8 The Ident and the Colour Killer Circuits

In the PAL system the modulated V component carrying the (R' - Y') information
is switched through 1800 at the beginning of every picture line. In order to achieve
synchronous demodulation of the Signal, it is necessary to carry out the phase
reversal of either the subcarrier or the V component in step with the transmitter.
This can only be done if the transmitted switched and unswitched lines are identi-
fied in the receiver. Since the coder switching creates the swinging or switched
phase burst, the phase detector in the a.p.c. loop generates the necessary identifying
signal, as described in the previous section. At that point its amplitude is low and
the shape resembles a charge-discharge curve of a capacitor deliberately produced
by the first stage of inadequate smoothing within the a.p.c. loop. The signal at this
point is called the ident ripple, but in this form it cannot be used in the receiver
PAL switch. The frequency of this ripple is equal to half-line frequency, referred to
for brevity as 7.8 kHz (the 625-line scan frequency is 15625 Hz). The phase dis-
criminator produces a positive-going error voltage during one line and a negative
output during the following line, thus giving one cycle of the ident for every two
lines.
The ident ripple is fed to an amplifier which changes the waveform into a large
amplitude sinewave and then into rectangular pulses. This is usually accomplished
by means of a tuned circuit amplifier, often with a degree of positive feedback
which can drive it so hard as to automatical,y clip the sinewave into an almost
rectangular (square) wave. In most cases further shaping of the ident sinewave is
carried out by means of diodes which follow the amplifier.
An example of an ident amplifier circuit is shown in figure 7.11. The resonant
circuit formed by L 1 , C2 , C3 is tuned to 7.8 kHz (usually to obtain maximum dis-
play of the resultant sinewave on the screen of an oscilloscope connected to the
collector of TR1). The waveform is then fed via C4 to the pulse shaping circuit of
Dl and TR2 with the associated components. Since the amplitude of the sinewave
output is in excess of what is needed to saturate transistor TR2, a square waveform
is developed across the emitter resistor R 7 • Diode Dl clips the negative half-cycles
of the input waveform. The ident pulses are fed to the PAL switch circuit, which
we shall discuss in section 7.9.
 ident pulse colour 2nd chroma
ident amp. squarer killer amplifier
+20 V • • •
I
I
I
R4 I
I
I
I
el ,
R I C"
sa:LTJI1 chroma from
1st amplifier
---tC
I 7
I
I
I
I L, C,
I
~,
+
L2~rOi ~
I
",
" ,,
,, • . . . chroma to
delay line driver
"
",
ident ripple
from
a.p.c.
circuit

Cs ident pulses
I----to PAL switch
bistable

Figure 7.11 Ident amplifier, colour killer and associated second chroma amplifier (GEe)
132 Television Principles and Practice

We now look at the colour killer arrangement. Its chief purpose is to shut down
the main chroma path during a monochrome transmission. This action prevents
spurious 'coloured' noise appearing on the screen which spDils the quality of what
is meant to be a black and white programme. (Normally any noise present would
be swamped by the chrominance signal.) Since colour transmission is always
accompanied by the burst, the colour killer can be operated, either directly or
indirectly, from the burst processing circuit. The most common method is to use
the ident signal to provide the correct d.c. bias for one of the chroma amplifiers. If
the ident is present, the controlled stage operates normally; if it is absent, the cDlour
killer drives the chroma amplifier into cut-off.
The circuit in figure 7.11 shows a simple colour killer arrangement. In the
absence of the ident signal, the second chroma amplifier transistor, TR3, is biased
OFF by a voltage derived from the negative supply rail via R 9 • When ident pulses
appear at the emitter of TR2, they charge the electrolytic capacitor C6 via D2 to a
positive voltage. Transistor TR3 is correctly biased by the potential divider Rs and
R 9 , and normal amplification takes place. The reader will observe that the circuit
of the chroma amplifier otherwise resembles one of the arrangements described in
chapter 5 on i.f. amplifiers.
Some manufacturers use a rather more involved colour killer circuit, with an
intermediate transistor which is used as a switch, either to remove or to apply the
correct d.c. bias to one of the chroma stages. Occasionally the 'unkilling' bias is
derived from the PAL switch, which is, however, linked with the ident circuit and
thus the basic idea remains unchanged.
We have mentioned that the colour killer operates whenever the ident signal is
missing, which brings us to the second purpose of the arrangement. Certain fault
conditions within the decoder will delete the ident; this could cause the picture to
become unviewable, but as the colour killer operates, the receiver is left with an
acceptable monochrome display. By developing this idea further and by connect-
ing the colour killer to the PAL switch, it is possible to spare the viewer the un-
pleasant effects of faults over a considerable part of the decoder by always leaving
him with a good black and white picture.
It is frequently helpful in tracing decoder faults to override the colour killer by
restoring a more or less correct d.c. bias to the chroma amplifier in question. If we
then view the display on the screen, the probable area where the fault lies should
reveal itself. Most makers' manuals in the section devoted to decoder adjustments
indicate the easiest method of overriding the action of the colour killer. In principle,
all that is needed is to complete the open link in the d.c. resistor bias chain to
the controlled chroma amplifier. In the circuit shown in figure 7.11 this is indicated
by means of the resistor Rsa (shown in broken line) connected between the positive
20 V line and the junction of R 9 , Rs, RIO; this forms a bias potential divider con-
sisting of Rsa and R 9 • The value of the resistor is usually given in the service manual,
but could also be obtained from potential divider calculations aiming at the norm-
ally expected voltages on the controlled transistor. We shall return to the problem
of decoder fault fmding later in this chapter.
Chrominance Signal Stages 133

7.9 The PAL Switch

The reader will recall from chapter 2 that the line-by-line switching of the V com-
ponent of the composite chroma at the transmitter must be repeated in the receiver.
If this were not done, as may happen under certain fault conditions in the decoder,
the synchronous detector would produce an output of +(R' - Y') during one line
and -(R' - Y') during the next, while the (B' - Y') component would not be
affected. Since each colour has specific values (and polarities) of the two colour
difference signals, we would see on the screen a line of the correct colour followed
by a line of a completely different colour. This would be the case of very severe,
and most objectionable, Hanover blinds ( see section 7.4). The colours that would
be especially affected are those with a high (R' - Y') content-for example, red or
green; on the other hand, the effect on the blue or yellow will be far less.
In the synchronous demodulator it is the mutual relationship between the V
signal and the reintroduced subcarrier which must be maintained; it is possible to
switch either the chrominance signal or the subcarrier. The latter is only a simple
sinewave; it is relatively easy to change its polarity, and that is the method most
commonly adopted. The switching of the rather complex chroma signal can intro-
duce distortion, resulting in a degraded picture quality.
The PAL switch is under the command of the ident signal to ensure that the
correct line is being switched. Two approaches are found as follows: the first uses
suitably amplified and shaped ident pulses to operate the PAL switch directly; the
second uses line flyback pulses to initiate the switching action, while the ident
signal effects a form of interlocking only to ensure that the correct line is being
switched.
In figure 7.12 the PAL switch operates on the V signal by reversing its phase
every other line to counteract the transmitted reversals. The positive half-cycle
of the large amplitude ident sinewave biases Dl ON, and the chrominance signal is

R,
d.c. level plus sinewave
from ident amplifier ------i

c,
± 2V from
PAL delay line I
+ 2V to (R'- Y')
demodu'lator

Figure 7.12 PAL switch of colour difference signal (Decca)

134 Television Principles and Practice

fed into the primary winding, Ll , of a transformer. In turn, a reversed polarity

signal is induced in the secondary, L 2 , which is then communicated to TRI. During
the unswitched transmitted line, the ident circuit produces the negative half-cycle
of the sinewave which biases D2 ON. This allows the V component to be applied
directly to the transistor. D.C. base biasing is obtained from the collector of the
ident amplifier via the potential divider Rl ,R 2 • Capacitor C2 blocks the d.c. bias
while completing the a.c. signal path through the transformer.
A problem with an 'ident-only' type of PAL switch is that it could switch over
during the actual picture line, giving incorrect colours on one side of the screen and
correct ones on the other, with a visible vertical band of obvious disturbance. This
may happen if the ident sinewave, or the resultant pulse, is unduly delayed with
respect to the beginning of the video line. A double beam oscilloscope could be use-
ful to observe this, and to carry out such adjustment as may be recommended by
the makers.
Another type of PAL switch is shown in figure 7.13. In this case it is the sub-
carrier feed to the (R' - Y') demodulator which is being reversed every other line.
The two transistors, TRI and TR2, with their associated components form a
bistable switching circuit. In operation, when one of the transistors is saturated,
then the other one is cut off owing to the cross coupling from the collectors to the
bases via Rs and R 9 • This state of affairs exists until a suitable negative pulse is
applied to the conducting transistor. The pulse momentarily cancels the forward
bias on the base of, say, TRI so that this transistor is cut off and its collector volt-
age rises. This rise will, in turn, be fed to the base of TR2, which will then be driven
into saturation, its collector voltage dropping almost to zero. Therefore, even if the
original negative pulse is removed, TRI remains in the cut-off condition because its
base supply voltage at the other end of Rs is held at zero. The next line flyback
pulse reverses the situation, and so on. The diodes, D3 and D4, connected to the
transistor bases are called the steering diodes and ensure that the negative switching
pulse is always directed to the saturated transistor. If a transistor is already cut off,
then the pulse has no effect on it.
In effect, the bistable divides the line pulse frequency by 2; its output frequency
is then equal to that of the ident.
The bistable line-by-line switching produces square wave outputs at the two
collectors; however, the waveforms are exactly 1800 out of phase with each other.
The actual PAL switch is shown in the upper part of figure 7.13. The bistable out-
puts are used to bias diodes D1 and D2 alternately ON. The anodes of the diodes
are connected to a fixed d.c. potential as given by the divider R 2 , R3 between the
+h.t. and earth. The cathodes are joined via current limiting resistors R 4 , Rs and
the windings of the transformer, L2 and L 4 , to the collectors of the bistable. When
TRI is bottomed, its collector is almost at earth potential, and D1 is biased ON. At
the same time, D2 is switched off because the collector voltage of TR2 is almost at
h.t. During the next pulse the state of conduction of the two diodes is reversed.
 C,
subcarrier from ----------J
reference oscillator ~ --,

subcarrier to
R, (B'-V'I demod.

L2
D1

3'
D2

subcarrier I
to (R'-V'I L6:: L5
demod. I ,

.---------1~--------_-- + h.t.

TR1

positive ident negative

pulses (see figure 7.111 line flyback pulses

Figure 7.13 PAL switch in the subcarrier path to the (R' - Y') demodulator
136 Television Principles and Practice

The sub carrier from the oscillator is fed to another winding of the transformer,
Ll , via C 1 and C4 .'This induces signal voltages in L2 and L 4 . When D1 conducts, it
directs the sub carrier from L2 to the primary (L 5) of another transformer, the
secondary of which (L6) provides the supply to the (R' - y') demodulator. When
D2 conducts, it connects L4 to L 5, but now the sense of the winding, L4 , is such
that the polarity of the signal is in opposition to that in L2 ; consequently the out-
put to the demodulator is also of a reversed polarity. Capacitors C9 , Cs , C6 provide
the necessary d.c. blocking while completing the a.c. signal path within the circuit.
When the receiver is first switched on, or even when changing channels, there is
a good chance that the PAL switch diodes (D1 and D2 in figure 7.13) could be in
exact opposition to the transmitted reversal of the V signal because the bistable
switch may have started in the 'wrong' phase. It is only a matter of chance which
of the two transistors will be driven into saturation first. If this were allowed to
continue, the line-by-line switching would produce a reverse polarity of the V
signal which means wrong colours would be displayed on the screen. To ensure
that the PAL switch is always in phase with the transmitter, the ident pulse which
occurs every other line (figure 7.11) is fed to one side of the circuit at the junction
of R13 and C7 . The positive ident will thus cancel out the negative line flyback
pulse, rendering it ineffective as far as TR1 is concerned, and the bistable misses
one operation. The next line pulse causes it to switch because the ident is at zero,
and from there on, normal in-phase action continues.
The diagram in figure 7.13 also includes a method of adjusting the 90° (quadra-
ture) phase shift oftlle subcarrier to the (B' - V') demodulator. From the descrip-
tion of the coder at the transmitter (chapter 2) the reader will remember that the
two feeds are in quadrature with each other, and, for correct synchronous demodu-
lation, the same phase relationship must be maintained in the receiver. The adjust-
ment available in this particular circuit is the trimmer capacitor C2 • The signal from
the parallel circuit, C2 , C3 combines with the one from L3 via R 1 to provide the
required phase shift. In practice, C2 is adjusted until the maximum signal ampli-
tude is obtained at the output from the (B' - V') demodulator (as viewed on the
oscilloscope ).

7.10 Integrated Circuits in Colour Decoders

The complexity of the decoder lends itself to circuit integration. Designers try to
achieve a compromise between several factors, including reducing the number of
individual circuits, maintaining reasonable reliability and low repair costs. A wide
variety of circuits are available which combine different decoding functions.
k; yet, it is not possible to provide in I.C. form either large values of capacitance
(in excess of, say, 20 pF) or of inductance. These and certain other components
are connected externally, and some difficulty may be experienced in identifying
their functions. If the reader is well acquainted with the requirements and layout
of the non-I.C. decoder, then a large number of similar features can be recognised.
In order to minimise the number of external components, designers occasionally
Chrominance Signal Stages 137

use circuit techniques which differ from their transistor counterparts. For example,
the ident signal may no longer need a tuned circuit to produce a large amplitude
signal, because a level detector, although quite complex, can be built into an I.C.
The diagram in figure 7.14 shows a complete colour decoder which uses four
integrated circuits. Some external controls and other important components have
been included in a Simplified manner to allow the reader to follow the arrangement
more easily and also to recognise the more familiar functions.
I.C.1 (TBA 560) is a combined luminance and chrominance amplifier. The
luminance signal from the detector is a.c. coupled at pin 3, and d.c. restoration
takes place by means of a black level clamp which is driven by line pulses; the
associated reference charging capacitor is connected to pin 4 (see chapter 6 regard-
ing the principle of such a clamp circuit). Contrast and saturation tracking is main-
tained by the common connection to both the luminance amplifier and the chrom-
inance amplifier. The external chroma bandpass filter between pins 1 and 15 is
designed to shape the frequency response of the chrominance amplifiers which
follow. In common with its transistor counterparts, the response complements th"t
of the vision i.f. in the vicinity of 35.07 MHz. The reader will recall that the chroma
sub carrier is usually placed on the slope of the iJ. response curve (figure 5.11);
thus the lower sideband content is not equal to that of the upper sideband. Many
manufacturers compensate for this unbalance in the chroma amplifiers.
The first chroma amplifier in 1.e.1 is controlled by the a.c.c. voltage at pin 14,
and derived from the burst (I.C.2).
Burst blanking removes the colour burst from the main chroma path, while
allowing the burst to be directed to the burst gate. The gain of the second chroma
amplifier stage is controlled manually by the saturation (or contrast) control at
pin 13 and electrically linked (tracking) with the contrast control (pin 2).
The output from the delay line driver, pin 9, goes to the PAL delay line via a
phase adjustment; the direct signal is applied via an amplitude adjustment to the
'add' and 'subtract' network. The resultant U and V signals are fed to I.C.3 (TBA
990). A removable link (pin 9, I.C.l) is included in the delay line circuit, allowing
the decoder to revert to the PALoS mode for various adjustments (the direct path
only is now available).
I.C.2 (TBA 540) provides most of the services required for demodulation. The
gated burst, whose phase has been adjusted while the decoder delay line was dis-
connected, is fed to the phase detector (discriminator). After filtering between
pins 13 and 14, the resultant d.c. is applied to the reactance stage and then to the
subcarrier oscillator. An external crystal and a trimmer capacitor are provided
between pins 15, 1 and 2 for manual adjustments.
The burst phase detector yields the ident signal from which the colour killer
circuit operates. In the absence of the ident, pin 7 of I.C.2 is at chassis potential;
this earths the slider of the saturation control (pin 13, I.C.1), reducing the chroma
amplifier output to zero. To override the colour killer, the link between the two
I.C.s is removed, allowing the chroma amplifiers in I.C.1 to be biased via the satura-
tion control only.
138 Television Principles and Practice

negal ive feedback

colour killer disabling link

burSI
phase
adjust

I.C.2 TBA540

set XI_I
oscillator
frequency

Figure 7.14 Colour decoder incorporating integrated circuits (Mullard)

 Chrominance Signal Stages 139

h.t.-_-----------,
140 Television Principles and Practice
I.C.3 is basically a synchronous demodulator together with the (G' - Y') matrix.
The sub carrier outputs from I.C.2 (pins 4 and 6) are first phase shifted through 90°
with respect to each other because of quadrature modulation requirement. One
feed is then sent directly to the (B' - Y') demodulator (pin 8) and the other goes
to the (R' - Y') detector via the PAL switch (pin 2). The switch is driven by a
bistable which, in turn, is operated by the line flyback pulses and is interlocked
with the ident signal. The colour difference outputs can be regulated at pins 9, 11
and 12.
Finally, [Link] provides the R, G, B matrix and drive to the video output ampli-
fiers. The three colour difference signals from the preceding I.C. pass through fIlters,
allowing only chroma signals up to 1 MHz to be included, and therefore remove the
subcarrier and demodulator switching frequencies. The luminance signal is obtained
from pin 5, I.C.l. The video driver stages have large amounts of feedback, not only
to stabilise the output amplifiers gain, but also to cope with variations in the video
output h.t. supply (205 V). The amount of blue drive is fixed, but both the green
and the red are variable.

7.11 Fault-Finding in Colour Decoders

Faults in the decoder could be grouped into three major categories: no colour,
incorrect colours, unlocked colours.
Before one attempts to locate a decoder fault, it is essential to establish a good
monochrome picture. This should be possible in most cases, except when I.C.s are
used extensively, where, as we have just seen, the luminance and the chrominance
paths use the same devices. We shall deal with I.C. fault-finding later.
The points to look for are good Signal strength (insufficient signal can give very
weak colours or none at all); good synchronisation (burst gating could be otherwise
affected); correct grey scale (colour casts can create an impression either of colour
when none should be present or of wrong colours).
The easiest method of fault-finding in the decoder is by means of a suitable
oscilloscope which indicates the presence and the magnitude of waveforms. The
oscilloscope must have adequate bandwidth to display chroma signals which are
centred on 4043 MHz. Ideally, its [Link] response should extend to 10 MHz,
while one with a 5 MHz bandwidth is barely adequate. The instrument, especially
of the lower bandwidth type, may load the circuit under test; this would disturb
the circuits and may result in additional faulty symptoms or incorrect readings. A
capacitive probe is often recommended to reduce the disturbance, but it must be
remembered that the V/cm calibration of the oscilloscope is then not applicable,
because some of the available signal voltage is dropped across the probe, leaving
perhaps only 10 per cent for the display. As the amplitudes of the chroma signals in
the initial stages are low, the oscilloscope must have high sensitivity, at least
50 mV/cm or better. Where I.C.s are used, an oscilloscope is invaluable.
Chrominance Signal Stages 141

No colour
Override the colour killer and observe the effect on the picture, as this helps to
identify the possible fault area in the decoder:
(1) Colour still absent. Suspect: chroma amplifiers from the chrominance take-off
point to the delay line driver; alternatively the sub carrier oscillator and its buffer
stage (absence of the sub carrier shuts down both demodulators and removes the
ident).
(2) Unsynchronised colour. Suspect: incorrect oscillator frequency caused by a
fault within the a.p.c. loop, including burst amplifiers and gate, or misadjustments
of available controls.
(3) Colour returns to the display. Suspect: the colour killer itself or the ident
amplifier(s), although in the latter case incorrect colours may appear because of the
PAL switch being out of step (see below).

Inco"ect colours
If at all possible, observe either a colour bar display or a test card, and note how
much the colours deviate from their standard sequence.
(1) Primary colour fault causing an overall colour cast on the screen: excessive drive
to one gun (primary colour cast) or insufficient drive (complementary colour cast)-
either can be easily observed in the 'white' bar; for example, a magenta cast is due
to the absence of green. Suspect: primary colour amplifiers (after the R, G, B
matrix) and black level clamps, incorrect d.c. levels in directly coupled amplifiers
(such as I.e.s), faults in brightness control circuits affecting one gun only. This type
of fault should have been discovered while checking grey scale tracking (see the
introduction to this section).
(2) Missing colour difference signal. Wrong colours result since the addition of the
y' Signal to the colour difference signals is now incorrect. This causes the tube to
be driven by incorrect signal voltages. The colours most affected are those which
would have a high content of the 'missing' signal. There are also slight differences
between the displayed colours if the fault occurs before or after the (G' - y/)
matrix, since this signal is then formed either incorrectly or correctly. If pOSSible,
the reader should experimentally disable the various signals, using an otherwise
co"ectly adjusted receiver and record his impressions of the colour bar sequence
for future reference.
The following effects refer to absent colour difference signals; where there is a
Significant colour change the correct colour is given in brackets. The white and
black bars are not affected by this type of fault.

(a) Missing (R' - Y/). The picture appears predominantly blue and green; the colour
bar sequence is: white, yellow, pale mauve (cyan), green, blue (magenta), dark
142 Television Principles and Practice

green (red), blue, black. Suspect: no V signal from the delay line, (R' - Y') demod-
ulator, subcarrier feed to the demodulator, certain faults in the PAL switch,
(R' - Y') amplifiers.
(b) Missing(B' - y'). The picture appears predominantly purple and pale green;
the colour bar display is: white, very pale yellow (yellow), pale green (cyan),
greenish cyan (green), reddish magenta (magenta), red, black (blue), black. Suspect:
no U signal from the delay line, (B' - Y') demodulator, sub carrier feed to the
demodulator, (B' -- Y') amplifiers.
(c) Missing (G' - Y'). Symptoms on the screen are most difficult to observe, since
this is the lowest amplitude signal and it has least effect. Basically all colours
become somewhat dull (including the green). Suspect: (G' - Y') matrix and
(G' - Y') amplifiers.
(3) Faulty PAL switch (normally the effects are visible only if the colour killer is
overridden-see 'No colour' faults).
(a) PAL switch out of step (with the transmitter). Applicable to circuits which
employ a bistable. Wrong colours may appear since the fault causes the polarity
of (R' - Y') to be inverted. In a picture this produces so-called 'red' grass and
'green' faces, as confirmed by the resultant colour bar sequence: white, greenish
yellow (instead of yellow), pale magenta (cyan), reddish orange (green), bluish
cyan (magenta), dark green (red), blue, black. The fault symptoms may either
appear or disappear upon changing channels, as there is an equal chance of the
bistable starting up either correctly or inc6lrrectly without ident synchronisation.
Suspect: ident circuits, ident feed to the bistable.
(b) PAL switch stopped. Very severe Hanover blinds, which from a considerable
viewing distance may appear as wrong colours. Suspect: bistable not switching
owing to a faulty circuit or lack of line pulses; fault in the ident circuits where a
bistable is not used. NOTE: The bistable output is at 7.8 kHz; consequently either
this or the ident can be coupled via a low value capacitor-say 0.01 J,LF-to the
audio output stage. A high-pitched sound will be heard from the speaker if the
signal is present.

L C fault-finding
(1) Identify the fault area and the associated I.C.s from the fault symptoms.
(2) Check all d.c. voltages associated with the I.C. These devices are usually directly
coupled throughout; if the d.c. readings are incorrect, the signal path will be affec-
ted.
(3) Check all a.c. signals (waveforms) and compare them with the makers' manual.
(4) Check that the circuits preceding (or following) the I.C. are functioning nor-
mally. If necessary, disconnect the feed to or from the I.C. and check the effect on
the relevant instrument readings.
(5) Observe the following precautions:
Chrominance Signal Stages 143

(a) Whenever possible, connect instrument probes away from the pins of the I.e.,
thus avoiding accidental short circuits.
(b) Resistance checks on an I.C. are meaningless and should not be carried out
unless specifically mentioned by the manufacturer.
(c) Avoid overheating the device with a soldering iron.
(d) Before replacing the I.C., ensure that the supply voltage is correct and that there
are no external component short circuits.
8TV Sound

8.1 Principle of Intercarrier Sound

The TV sound signal is transmitted at a carrier frequency 6 MHz above the vision
carrier frequency. The mixer stage in the tuner produces two Lf. frequencies corres-
ponding to the two carriers: 33.5 MHz, sound signal; 39.5 MHz, vision. Both Lf.s
are amplified in the common Lf. strip until the sound information is finally separated
in the form of a beat frequency Signal-that is, 39.5 MHz - 33.5 MHz = 6 MHz.
The frequency difference between the two carriers is then called the intercarrier
frequency. The intercarrier signal is generated by a non-linear device, such as a
simple diode detector which acts as a mixer. As shown in the diagrams in figure 8.1,
the video detector itself is often used to produce the 6 MHz beat frequency (the
chrominance signal can also be obtained on the same principle).
An example of an intercarrier signal take-off arrangement is given in a simple
circuit shown in figure 8.1(b). The vision detector diode is followed by a 6 MHz
tuned circuit L 2 , C2 in series with the feed to the video amplifier to provide a rejec-
tion of sound from the video stages. At the same time L2 ,L3 form a transformer so
that the intercarrier signal is induced in the secondary winding L 3 • Since L 3 , C3 are
tuned to 6 MHz, a large amplitude Signal is communicated to the 6 MHz amplifiers.
In some designs the intercarrier signal is taken off at a later point in the video
amplifier stages. The disadvantage of such a method is that sound/vision or sound/
chrominance beat frequency patterns can interfere with the picture. Conversely,
there could also be a greater possibility of vision-on-sound interference (so called
viSion-buzz). A separate diode is sometimes used to free the vision detector from
handling the 33.5 MHz information of any Significant amplitude and to lessen the
chance of interference patterns.
An alternative method to the intercarrier principle would be to amplify the
33.5 MHz signal separately. Such an arrangement would introduce unnecessary
circuit complexity and, in the event of the tuner oscillator drift, the sound carrier
could fall outside its own Lf. response curve. Should that happen, the resultant
audio output could become either noisy or distorted. The intercarrier signal is
always at the same frequency; consequently good quality sound is possible provided
both the signal strength and the receiver Lf. alignment are satisfactory. The same
conditions apply in order to prevent the sound signal interfering with the picture.
144
TV Sound 145

from __
tuner
Lt.
strip r- vision
detector ~
6 MHz
trap
_ to video
amplifiers

i
i ntercarrier
acceptor circuit I-
6 MHz limiting
amplifier I-

(a)
f.m.
detector r-- audio
amp.
-a1
4Jt:h
r--------------<.~ to 6 MHz amp.

D1
----
L3
I4
+---_ _ _ _ _~ to video
amp.

from final
Lt.

(b)

Figure 8.1 Intercarrier sound: (a) block diagram of TV sound as part of a receiver;
(b) intercarrier take-off arrangement at vision detector

8.2 Intercarrier Amplifiers and Sound Demodulators

The TV sound carrier is frequency modulated, and therefore it is relatively unaffec-
ted by amplitude modulated interference signals. Interference may come primarily
from the vision signal, which, on peak white, could cause accidental amplitude
modulation of the sound signal. The reader will recall that the luminance signal is
not allowed to modulate the vision carrier below 18 per cent of the carrier ampli-
tude; this ensures that a sufficiently large intercarrier signal is always present. How-
ever, vision buzz would become evident if the correct relation between the ampli·
tudes of the two carriers were lost in the receiver.
The sound system will enjoy freedom from interference provided the intercarrier
amplifier offers so much gain that the peaks of the amplified signal can be 'sliced
off'. Most of the interfering amplitude modulation would then be eliminated and
the sound detector would be presented with a 'pure' f.m. signal. Such amplifiers
are called limiting amplifiers because they restrict the amplitude of their output
signal. Where transistors are used in the intercarrier section, a diode may be connec-
 146 Television Principles and Practice

ted across the 6 MHz circuit in order to clip any excessive amplitude. Alternatively,
the biasing of the tuned circuit amplifiers is such that self-limiting takes place with-
in the stage.
Ceramic filters, which replace tuned circuits, can be found at the input to the
intercarrier amplifiers. Their behaviour is similar to that of a crystal vibrating at a
specified frequency. To improve the selectivity and amplifier gain, two ceramic
fIlters following each other may be used. For TV applications the filters are tuned
to 6 MHz; they are also used in f.m. radio receivers, where they are tuned to 10.7 MHz.
Ceramic fIlters offer a narrow bandwidth, which results in excellent selectivity; they
have good stability and present no alignment problems.
Modern designs use integrated circuits to provide amplification of the 6 MHz
signal. The same I.C. may also contain the sound detector as well as an audio pre-
amplifier.
Where the detection of the audio signal is done by discrete circuits, two main
methods are used-namely a ratio detector or a slope detector.
A circuit diagram of a ratio detector is given in figure 8.2. The detector trans-
former, L J , L 2 , is tuned to 6 MHz and the tertiary winding, L 3 , injects an addi-
tional voltage to the centre tap of L 2 • When the incoming signal remains at its

VRl

10k C6 0.051'

intercarrier

..
input

I
6 MHz
detector response curve

Figure 8.2 F.M. ratio detector circuit (Rank). The S-curve represents the response
of the circuit-an audio output signal is produced as the intercarrier frequency
deviates from 6 MHz
TV Sound 147

carrier frequency, the two diodes conduct by equal amounts and the reservoir
capacitor, C7 , charges to a d.c. voltage dependent upon the signal strength. Because
of the balance under this condition, the potential across VR2 is zero. When the
modulating signal causes the intercarrier frequency to deviate from 6 MHz, the
balance of the circuit is upset. This results in a current flow during one half-cycle
through D2, chassis, C4 ,L3 and back to L2 ; during the next half-cycle of the audio
signal the current is in the opposite direction via Dl, L 3 , C4 , chassis and back to
Dl. The audio output is developed across C4 and its parallel network of R 6 , Cs ,
C6 and the volume control VR2.
In common with f.m. radio broadcasting, TV sound is subjected to pre-emphasis
at the transmitter. It means that high frequency audio signals are boosted by com-
parison with middle and low frequencies. The emphasised high frequency informa-
tion swamps most of the subsequent noise, which also tends to have a predomin-
antly high frequency content. In the receiver the signal is given de-emphasis or
treble cut. The original high frequency audio signals are restored to their proper

intercarrier C1 R1
input from
limiting
----11--.. . . .-----..--1
amp.
VR1

to audio
amp.

(a)

5.9 MHz

frequency

(b)

Figure 8.3 F.M. diode slope detector. The tuning adjustment (L3) places 6 MHz
on the slope of response curve (b)
148 Television Principles and Practice

level, which automatically reduces the effects of noise. A simple filter, consisting
of a resistor and a capacitor, achieves the required effect. Pre- and de-emphasis are
expressed in terms of the time constant of such a filter; in the UK it is 50 J1S. In the
circuit in figure 8.2 the filter consists of R6 and Cs , whose time constant is
10 kn x 4.7 nF = 47 J1S, which is near enough to the specified value. Of course,
other combinations of component values are possible, but their method of connec-
tion with respect to the signal path remains as shown in the diagram.
Rejection of a.m. signals relies on the detector circuit being perfectly balanced.
Therefore the total resistance of the circuit associated with Dl must be the same
as that of D2. In figure 8.2 variable resistor VRI compensates for imbalance caused
by differing forward resistances of the diodes and also for the tolerance of R 5 • The
adjustment of VRI aims at a minimum buzz on sound. The relatively large capacit-
ance of C7 also helps to suppress any a.m. interference, while Cs is a 6 MHz filter
capacitor.
There are many versions of the ratio detector circuit, but the fundamental
principle remains the same for all of them. Often the centre tap in the secondary
of the detector transformer is not in the winding, but is formed by the junction of
two capacitors (in place of C2 in figure 8.2).
A slope detector circuit is shown in figure 8.3(a). The intercarrier signal is applied
to the tuned circuit formed by L3 and C3. The response of this circuit is such that
6 MHz is on the slope of the resultant curve-see figure 8.3(b). As the modulated
intercarrier frequency swings within narrow limits (maximum ±50 kHz), the output
from the arrangement varies accordingly and provides the audio signal; de-emphasis
components are R 1 and C4 . The alignment of this type of circuit can be made fairly
simple by tuning L3 for a maximum undistorted output.
The final method of detection to be considered is quadrature detection. This is
usually incorporated in an I.e., which often includes a multistage limiting amplifier
and an audio preamplifier. A block diagram, illustrating the principle of quadrature
detection, is given in figure 8.4(a). Transistors TRI and TR2 produce separate
pulsed outputs from the amplified and limited intercarrier signal inputs. When the
frequency is exactly 6 MHz, the two pulses are 90° out of phase (or in quadrature)
with each other; capacitor C at the input to TR2 provides the necessary phase shift.
The quadrature condition is maintained at the resonant frequency of 6 MHz by
the tuned circuit L 1 , C1 • Transistors TRI and TR2 are driven between saturation
and cut-off, and the resultant outputs are fed to the third transistor circuit, TR3.
The conduction of TR3 depends upon the degree of coincidence between its two
inputs. As the modulation deviates the intercarrier frequency from the centre value
of 6 MHz, the presence of the tuned circuit alters the original 90° phase shift. The
net result is that the two sets of pulses fed to TR3 coincide either for longer or
shorter periods of time. In effect, the input to TR3 is in the form of variable width
pulses which switch the transistor on, either for shorter or for longer time intervals.
The resultant output from TR3, after suitable filtering, produces an output that
varies in sympathy with the original frequency modulation.
An example of a complete circuit used in a TV receiver is given in figure 8.4(b).
TVSound 149

A signal from the video detector is fed to the 6 MHz ceramic fIlter (C.F.), from
which the intercarrier frequency of a very narrow bandwidth is applied to pin 2 of
ICI. After several stages of limiting, the frequency modulated signal enters the
quadrature detector. Components L 1 and C 1 form the external tuned circuit-

f.m. intercarrier audio

input I--.......--I~ output

IUL
variable phase
shift between
output pulses due
to action of L,
and C,
I
variable
pulse width
input signal
Ice ,

(a)

h.t.

I.C.l
@ to audio
1----<>--1J---I~ amp.

from
vision
detector

feedback
from audio
output
stage

(b)

Figure 8.4 Quadrature Lm. detector: (a) block diagram showing the principle of
quadrature coincidence detector; (b) complete practical circuit using I.C. type
MC 1358Q (Thorn)
150 Television Principles and Practice

decoupled to chassis via C2 • The quadrature coil, L 1 , can be simply adjusted for
maximum amplitude, un distorted audio output.
The volwne control, VRl, acts indirectly by varying the d.c. voltage fed to a
separate attenuator stage in the I.C. The signal is then coupled via C4 to a built-in
preamplifier circuit. Finally, the audio output is fed to a power amplifier which
drives a loudspeaker.
De-emphasis components are often less obvious when I.C.s are used, because the
required arrangement may be part of the integrated circuit itself. In figure 8.4(b)
capacitor C6 , in conjunction with the frequency dependent feedback network R 3 ,
C7 at pin 7, offers the necessary compensation.

8.3 Audio Frequency Amplifiers

In this part of the receiver the power level of the detected audio signal is raised
from a fraction of a milliwatt to perhaps several watts in order to drive the loud-
speaker.
Where valves are used elsewhere in the receiver, it is not uncommon to find a
valve audio output stage. Since a power output valve may require an input voltage
swing of several volts, a transistor or a valve preamplifier is necessary.
Audio output amplifiers may be divided into two groups according to their d.c.
biasing: class A and class B. In class A amplifiers, with no signal applied, the valve
or the transistor is biased so as to carry a relatively high output current, known as
the quiescent current. When the audio signal is added to the bias current, the out-
put current increases further during one half-cycle, and reduces below the quiescent
value in the next half-cycle. To prevent signal distortion, the biasing must be care-
fully arranged to ensure that, with the maximum signal amplitude, the device is
neither driven into saturation nor into cut off. Class A output amplifiers favour
valve circuits because the quiescent current is only in the region of, say, 20 or
30 rnA. In transistorised power output stages the 'standing' current may be of
several hundred rnA. This tends to create excessive heat diSSipation; it also requires
a high power rated transistor and leads to a large current drain from the power
supplies. Incidentally, class A amplifiers are used in most other parts of a TV
receiver, where signal amplitudes are usually low.
Class B amplifiers require two transistors (valve circuits of this kind are not used
in TV receivers). They are arranged so that, theoretically at least, one transistor
conducts during positive half-cycles of the audio signal, and the other transistor
amplifies the negative half-cycles. Often one of them is a p-n-p device and the
second one an n-p-n; the arrangement is called a complementary symmetry output
stage (see figure 8.5). In this the p-n-p transistor deals with the negative-going part
of the waveform; the n-p-n device amplifies the positive half-cycles of the input
signal. In practice, it is not possible to rely on the transistors to be switched on by
the signal alone, because a certain minimum value of base voltage must be reached
before conduction can begin. Unless a d.c. bias of approximately 0.2 V for ger-
manium and 0.6 V for silicon devices is provided, the output will be seriously dis-
torted. This type of distortion is called crossover distortion, for it occurs when one
TVSound 151

stabilised + h.t. 11.6 V unregulated + h.t.

effect of crossover
distortion in the output
for a sinusoidal input

coil12n

R7 470
audio audio bias complementary
preamp. driver stabiliser output transistors

Figure 8.S Complementary symmetry audio output stage. Voltage levels show
how the d.c. output voltage is maintained at about half the h.t. supply potential
(Thorn)

transistor is approaching zero current and the other one is due to take over. The
effect of crossover distortion on a sinusoidal waveform is shown in figure 8.5. The
reader will appreciate that the presence of such a defect would be particularly
noticeable at a low volume of sound.
An example of a class B complementary output stage is given in figure 8.5. An
important feature to note is that the whole stage is directly coupled, except for
the loudspeaker, which is fed via capacitor Cs . Therefore the transistors not only
act as amplifying devices, but also provide correct d.c. biasing for one another.
Because of the direct connection between the individual stages, even slight drift in
d.c. voltage in one of them could spread to the remaining stages and the biasing
throughout the whole amplifier would be seriously upset. To prevent this happen-
ing, d.c. feedback is used to stabilise the operating point of the circuit.
In figure 8.5 transistor TRI is responsible for maintaining correct overall bias
for the stage. The base of TRI is connected to a relatively conventional potential
divider R 1 , R 2 , R 3, R 5, but its emitter is effectively returned, via R 6 , to the junc-
tion of the output pair TR4 and TRS. Because of the symmetry in the output
circuit, the voltage at the junction of Rs and R9 is approximately one-half that of
the supply voltage. Transistor TRl, however 'senses' the value of the voltage at
152 Television Principles and Practice

the junction of TR4 and TRS; any change in this potential is communicated to the
emitter of TRI to adjust its conduction. If, for example, the 'centre point' output
voltage increased above 6 V, TR1 would be forced to conduct less. As a result the
collector voltage would rise and, because of direct connection to the base of TR2,
this transistor would also conduct less. In turn, the base voltage on TR4 would fall;
consequently, the emitter voltage of TR4 would be brought back to its nominal
6V.
Transistor TR3 in figure 8.5 acts as a variable resistor in the base bias circuit of
the output transistors. As mentioned earlier, the output stage requires a certain
amount of base bias in order to prevent crossover distortion. However, if full
advantage is to be taken of class B operation, the d.c. biasing must be kept to a
minimum. In this circuit the adjustment is carried out by means of VR1, which
varies the conductivity of TR3 and, in effect, alters the biasing of TR4 and TRS.
The setting of VR1 is critical-if the quiescent bias is too high, the transistors can
become overheated and destroyed. The manufacturer's adjustment procedure for
this particular receiver is to connect a milliammeter in the h.t. feed supplying the
output transistors, and then adjust VR1 until the current is 4 rnA (volume control
set to minimum).
Class B audio amplifiers in other receivers may be without a variable bias control.
There are also circuits which use output transistors of the same polarity (either
both p-n-p or both n-p-n) instead of complementary types.
The audio signal path in the circuit in figure 8.S can be followed in a similar
manner to the change in the d.c. levels discussed above. The reader is invited to
work out for himself what happens when the positive half-cycle increases the con-
duction of TR1 and the negative half-cycle reduces it. illtimately, the positive-going
portion of the input waveform causes the current in TR4 to rise, while cutting off
TRS; the load current then flows through TR4, Cs , the loudspeaker and the chassis
-charging C s in the process. During the negative half-cycle the signal current in the
speaker changes its direction as TRS conducts by discharging capacitor Cs .
Apart from the d.c. feedback, there is also a.c. negative feedback, which carries
out a number of functions, including preventing instability, improving the frequency
response, reducing distortion, etc. This is achieved in figure 8.S by: C3 ,R 5, R3 , C6 ,
C4 and, finally,R 7 .
Full understanding of the operation of this and similar types of directly coupled
circuits is important in fault-finding. The effects of a fault in one part of the circuit
can affect the biasing of the remaining sections, which makes the interpretation of
voltmeter readings difficult. Incidentally, a circuit somewhat similar to that in
figure 8.5 may also be used in the field time base of a TV receiver.
We shall not consider details of class A audio amplifiers, since they follow the
usual circuit techniques where either a single valve or a transistor is used.
The diagram in figure 8.6 shows a circuit of an audio output stage which em-
ploys an I.C. This may be preceded by a demodulator I.C. of the type illustrated in
figure 8.4. Integrated circuit amplifiers are capable ofvery high gains; consequently,
their frequency response must be carefully tailored to obtain the desired bandwidth
TV Sound 153

and to prevent instability. The shaping of the amplifier frequency response in

figure 8.6 is by means of the negative feedback networks: R 6 , C6 from the output
pin 6 to pin 9 (via C7 ) and R 3 , C3 from pin 16 to the input pin 1. Components
R 4 , C4 (pin 16) and Cs (pin 15) are also used to determine the low frequency and
the high frequency response, respectively. The d.c. bias of the preamplifier is
obtained via Rl ,R 2 (decoupled by C2 ) between pins 1 and 2. The amplifier output
(pin 6) is shunted by a compensating network C7 , R 7 ; it ensures that the effective
loudspeaker load is resistive rather than inductive. The latter arrangement provides
for a better response to rapidly changing signals (transients); it also offers protection
to the output transistors in the I.C. against overvoltages which could occur if the
effect of the speaker inductance was not compensated.

r
C2 h.t.
I.C.l other pins not in use

CD @

.. audio

CD
a.f. from preamp.
detector

@ ®0 @ ®

Figure 8.6 Audio output amplifier using I.C. type SN 76013/ND/7 (GEC)
9 Synchronising Pulse
Separator and Field Time
Base Circuits

9.1 Separation of Synchronising Pulses from Composite Video

The synchronising pulse separator circuit produces a train of relatively large ampli-
tude rectangular pulses from the composite video waveform. The input to the cir-
cuit is from one of the video (luminance) amplifier stages. The separated pulses,
after further processing, are then available for synchronising the receiver time bases
with the scanning action of the transmitter camera tube.
The process of separation relies on the fact that the synchronising pulses in the
composite signal always begin at a fixed reference point of the video waveform-
the black level. Therefore the required circuit is a voltage level detector. The separa-
tor circuit is made non-conducting during picture information, but it is driven into
saturation when the video waveform enters the 'blacker than black' region. Inter-
ference signals can also be present in the synchronising pulse part of the waveform;
therefore some circuits may have additional features to render the separator rela-
tively immune to unwanted Signals.
A simple transistorised circuit is shown in figure 9.1. Negative-going video signal
(hence, positive-going pulses) from the video driver stage is fed to the separator
transistor, TRI, via C 1 and R 1 • When the transistor is driven into saturation by the
synchronising pulse portion of the input waveform, the collector voltage falls, and
negative-going output pulses are generated. At the same time the input coupling
capacitor, C 1 , is charged to the peak value of the incoming waveform. After the
synchronising pulse interval, the amplitude of the input video waveform falls, which
effectively places the charged capacitor across the base-emitter junction of TRI.
The negative potential on the right-hand plate of C 1 causes the transistor to be cut
off, and the collector output voltage rises to h.t. This state is maintained during the
active picture period as the capacitor discharges via both the preceding amplifier
stage and the high value resistors R2 and R 3 . When the next pulse interval arrives,
C1 is recharged and the process is repeated. Resistors R2 and R3 provide a potential
divider which applies a small forward bias to the base of TRI. This, by itself, is
insufficient to turn the transistor on, but it ensures a much quicker changeover
from cut-off into saturation when the pulses are applied.
154
Synchronising Pulse Separator and Field Time Base Circuits 155

+ h.t.
---<1..-----_-
25 V

sync. pulses
t - -___ to time base

video input ---t

0.2211

Figure 9.1 Transistorised sync. separator circuit

composite _ sync. pulses

video input output

composite
video input

D1
r C3
tosync.
separator

(b)

Figure 9.2 Noise gated sync. separator: (a) block diagram-interference pulses on
video waveform are used to operate the gate circuit; (b) practical circuit; Dl and
VRI set the threshold of interference level and TRI is the noise cancelling gate
(Rank)
156 Television Principles and Practice

Negative-going output pulses are produced by the circuit in figure 9.1. Should
positive synchronising pulses be required, either an inverting amplifier might be
included or the take-off point would be chosen from a suitable luminance (video)
amplifying stage. Alternatively, a p-n-p transistor sync. separator circuit may be
found in an arrangement of n-p-n transistors.
A number of receivers use a noise gated synchronising separator. An ordinary
separator circuit is likely to amplify interference signals and produce spurious pulses
at its output. It will be shown later, however, that both the line and the field time
base oscillators are not easily affected by such unwanted signals. For that reason
most receivers do not take any special precautions against noise in the separator
stage.
Noise gating, or noise fIltering, is based on the fact that the amplitude of inter-
ference pulses is much greater than the amplitude of synchronising pulses. A block
diagram of a noise cancelled synchronising separator is shown in figure 9.2(a)-the
gate renders the actual separator circuit inoperative during interference periods. The
practical circuit may be either contained in an I.C. or can use discrete components
as given in figure 9.2(b). In this example both the synchronising pulses and the
accompanying interference are positive-going. The gating diode, Dl, is normally
reverse biased from the potential divider formed by R 1, VRI and R 2 • The amount
of bias is adjusted by VRl, so that the synchronising pulses cannot, by themselves,
turn the diode on, and the pulses continue to the separator via C2 • Large amplitude
negative-going interference pulses bias Dl ON, which, in turn, switches the gating
transistor, TR1, on. In effect, a momentary short circuit is placed across the signal
path to the separator and the noise is removed. Capacitor C3 is of a relatively low
value (330 pF); it removes the high frequency chrominance information, which
could also extend into the synchronising pulse level. The presence of C3 alone

R4

to field
time base

to line
time base
composite
video
input
..
R5 C2

Figure 9.3 Valve sync. separator circuit

Synchronising Pulse Separator and Field Time Base Grcuits 157

would have helped in reducing the amount of interference reaching the synchronis-
ing separator. In many receivers such a simple capacitor fllter provides all the
necessary noise cancellation.
A valve synchronising separator circuit is shown in figure 9.3. The principle of
operation is otherwise similar to that described for its transistor counterpart. The
video input has positive-going synchronising pulses which drive the valve into satura-
tion and charge C 1 at the same time. For the remainder of the waveform the charge
stored in the capacitor maintains the pentode in a cut-off state until the arrival of
the next pulse. Therefore the outputs from the anode circuit are in the form of
negative-going pulses. The screen voltage of the valve, set by R 4 , R 5, is relatively
low, which ensures that the cut-off state is also maintained under nO-Signal condi-
tions. The valve is usually part of a triode-pentode arrangement; the other section
may be used in one of the time bases.
Finally, the synchronising pulse separator circuit can be incorporated into an
I.C. The device would, naturally, provide other functions, as will be shown later in
this chapter (see figure 9.14).
The output pulses from the sync. separator must be of the correct magnitude,
shape and timing. These conditions are fulfllied by the values of the components in
the circuit. Should a fault develop in this area, the separator might, for example,
switch during picture information, or it could produce pulses of insufficient ampli-
tude. Such defects would cause lack of proper synchronisation, or lead to the
synchronisation being upset by the picture content. Both the line and the field time
bases could be affected by a faulty sync. separator. However, the flywheel circuit
in the line time base can sometimes mask the effect of separator faults on horizon-
tal synchronisation (see section 10.1).

9.2 Processing of Field and Line Synchronising Pulses

The output from the sync. separator is in the form of either positive- or negative-
going pulses: the line, field and equalising pulses are all present in succession. They
must now be sorted out in order to make the two time bases respond only to their
respective synchronising information. That is done by means of simple resistor-
capacitor networks. Depending upon the values of the components and their method
of connection, these circuits can be called either integrating or differentiating
circuits. The names are derived from mathematical operations, because such net-
works can, apart from their TV applications, perform the mathematical equivalents
of either integration or differentiation in, for example, certain computer circuits.

Integrating circuits
An example of such a circuit is given in figure 9 A(a). Capacitor C is allowed to
charge via resistor R from the driving circuit. The graph in figure 9 A(b) shows
how the voltage across the capacitor rises if the input, VIN , is maintained constant
for a relatively long time. Ultimately, the capacitor would become fully charged
158 Television Principles and Practice

r --- - -o------j
I R
I

(a)

applied input
V 1N ------------- B Vmax initial voltage
.~
Cl.
co
"III
1:
"co
~
+01
O.37Vmax
initial "0
linear rise >

charging time f-----I discharging time

T= RC

(b) (c)

Figure 9.4 Pulse integrating circuit: (a) basic circuit-input waveform charges C
via R; (b) graph of voltage rise across capacitor during charging (assuming constant
[Link] at the input to the circuit); (c) graph ofvoltage fall across capacitor during
discharging

and the output voltage would be equal to the input. The speed with which the
potential across C rises depends upon the value of the series resistance, R, and the
capacitance, C. The product R X C is called the time constant of the circuit (T).
From the mathematical properties of the charging graph it can be proved that after
a time interval equal to the time constant the voltage C'n the capacitor reaches 63
per cent of the maximum driving voltage. The initial shape of the graph up to, say,
10 per cent of the maximum is reasonably linear, after which the graph becomes
increasingly curved.
If the circuit is fed with synchronising pulses, the voltage, VIN , is sustained only
for the duration of each pulse. Therefore the time constant of the integrator is
deliberately made long by comparison with the duration of the input pulse. Now
the capacitor voltage will rise only along the initial portion of the charging curve in
figure 9 .4(b).
To sum up: an integrating circuit has a long time constant when compared with
Synchronising Pulse Separator and Field Time Base Qrcuits 159

- - ,.-- r-- ,.--

~~~J--~'==~~~~==~
line pulses equalising pulses
_____
(a)
~~~~~~~~~~~~~~
field pulses equalising pulses

field time base

trigger level

(b)

Figure 9.5 Formation of a field time base trigger pulse (see also figure 3.8):
(a) output from sync. pulse separator; (b) output from field pulse integrating circuit

the duration of the applied pulse, and the output is taken from across the capacitor.
This type of circuit changes the shape of a rectangular, or square, pulse input wave·
form into a sawtooth type of output waveform. An integrating circuit is used to
produce a field time base trigger pulse from the incoming field synchronising
sequence. Another application is in the generation of sawtooth scan waveforms and
certain convergence correction waveforms in the line and field time bases.
The diagram in figure 9.5 explains how the field synchronising sequence of pulses
is changed into a field trigger pulse by the action of an integrating circuit. The
pulses produced by the sync. separator are shown in figure 9.S(a); these are then
fed to an integrator. The resultant output waveform across the integrating capacitor
is illustrated in figure 9 .S(b).
In a practical version of the circuit in figure 9.3 the value of resistance can be
18 kn and that of the capacitance 2000 pF; the resultant time constant T=R x C=
18000 n x 2000 pF = 36 JiS. This is relatively long compared with the duration of
the various input pulses. Therefore, turning to figure 9.5, the equalising pulses
(2.3 Jis) charge the capacitor only to a low voltage because of the long time constant
of the circuit. During the intervals between successive equalising pulses the capacitor
loses most of its voltage. When the broad field pulses (27.3 Jis) are applied, the
charging period is longer than the discharge time; consequently, the capacitor out-
put voltage is in the form of 'serrations' of increasing amplitude. The intervals
160 Television Principles and Practice

between the equalising pulses which follow the field pulses allow the capacitor to
discharge to its initial voltage level. The actual calculations of the amplitudes of the
charging and discharging steps in the waveform are beyond the scope of this book.
When the build-up of the pulse voltage across the capacitor reaches a predeter-
mined value, the field time base oscillator is ~ynchronised to initiate field flyback.
A possible trigger level is suitably marked in figure 9 .S(b). The first part of the
waveform also indicates that line synchronising pulses (4.7 ps) charge the integrat-
ing capacitor to a relatively low potential. This is below the field oscillator trigger
level and has no effect. Similarly, short duration interference pulses cannot produce
sufficient output to upset field synchronisation.
The need for equalising pulses can also be seen from figure 9.S(b). These are
necessary to maintain good interlace of the two fields, which form a picture
(frame). The first field ends on a complete line (this corresponds to the waveform
in figure 9.5) so that the integrating capacitor can discharge almost completely
before the field synchronising sequence begins. The second field ends on halfline;
the reader will observe that the remaining capacitor voltage would then have a
higher value at the start of the next field pulse sequence. The slight differences
between the two fields would cause the fixed trigger level to be reached later on
the first field than on the next, resulting in poor interlace. Equalising pulses 'reset'
the capacitor voltage always to the same value, giving the broad pulses an equal
starting potential at the end of either field.

Differentiating circuits
An example of a simple differentiating circuit is given in figure 9.6. It consists of a
capacitor and a resistor, but now the output is taken from across the resistor. This
means that the output waveform is proportional to the capacitor current. When a
pulse is applied to the input, a large initial current flows, its value being liinited only
by the resistance in the circuit. As the charge on the capacitor builds up, the current
becomes less until, fmally, it reaches zero. The complete process follows a graph
which is identical in shape with that in figure 9.4(c).
In a differentiating circuit the value of the time constant (R X C) must be small
when compared with the width of the pulse applied to it (readers are reminded that
for an integrator the time constant must be longer than the pulse width). A short
time constant ensures that the charging is completed rapidly, and the output falls to
zero for the greater part of the input pulse.
When the pulse is removed, the fully charged capacitor behaves like a storage
battery and it discharges through R and the driving circuit. However, the flow of
the discharge current is in the opposite direction to that of the charging process-
hence a negative-going pulse in the output, as indicated in figure 9.6(c).
To sum up: a differentiating circuit has a short time constant when compared
with the duration of the applied pulse, and the output is taken from across the
resistor. The shape of a rectangular (or square) pulse input waveform is changed
into two sharp 'spikes' at the output, which coincide with the leading and the trail-
Synchronising Pulse Separator and Field Time Base Grcuits 161

C
o-----JJ---+-----o
Jl I l (b)

input output

(a)

(c)

Figure 9.6 Pulse differentiating circuit: (a) basic circuit-input waveform charges
C via R; compared withfrgure 9.4, output is now across resistor, and the time
constant is much shorter; (b) example of an input waveform-one narrow pulse
followed by a wide pulse (part of field synchronising sequence); (c) output wave-
form as a result of differentiating the input pulses

ing edges of the original pulse. Differentiating circuits are chiefly used to produce
sharp pulses in order to switch another circuit-for example, a time base oscillator
or a bistable in the PAL colour decoder.
Since the output from a differentiating circuit is relatively independent of the
input pulse width, this type of circuit produces line time base trigger pulses from
the narrow equalising pulses, from the broad field pulses and, of course, from the
line synchronising pulses themselves. This effect is illustrated in figure 9 .6(b) and
(c), where two different input pulses result in identically shaped o~tput pulses.

9.3 Receiver Time Base Requirements

The purpose of the time base circuit is to supply a current waveform to the deflect-
ing coils whose resultant magnetic field will deflect the electron beam in the tube.
A correctly synchronised 'linear' picture should be scanned on the screen. Good
linearity suggests that the picture is free from distortion, as is evident when simple
geometric forms are displayed. In colour receivers the time base must provide
additional waveforms for convergence and other raster correction purposes.
162 Television Principles and Practice

The current waveform in the scan coils is principally a sawtooth-the relatively

long rise in the waveform generates the scan and the rapid fall initiates the flyback.
A simple integrating circuit, of the type shown in figure 9.4, can be modified to
produce a sawtooth waveform. Charging the capacitor via the resistor at a relatively
slow rate gives the scan portion of the waveform. At the end of the scan period the
capacitor must be rapidly discharged by means of an electronic switch in the form
of a time base oscillator connected across the capacitor. The discharge time is very
brief because the resistance of the switch is low (for example, a saturated transistor
or a valve), giving a very short time constant. The resultant sawtooth waveform is
further amplified before being fed to the scan coils.
The presence of the inductance of the scan coils and, to some extent, that of the
transformer can also modify the applied waveform. If a rectangular voltage pulse is
fed to an inductor, the resultant current will rise relatively slowly because of the
back-e.m.f. in the coil. The graph of the current growth is then identical in shape
with the voltage rise across a capacitor [figure 9 .4(b)] . The rate of increase again
depends upon the time constant, which for an inductive circuit is given by L/R (L
is the inductance of the coil and R its resistance). This type of circuit can also act
as an integrator, provided the time constant is long compared with the duration of
the applied voltage pulse.
The length of the field scan is approximately 1/50 s, or 20 ms (ignoring the dura-
tion of flyback); therefore the time constant of an integrating circuit which pro-
duces the field sawtooth waveform would have to be, say, 100 ms. This is easily
achieved with a capacitor and a resistor of suitable value. The alternative method
of using the scan coils to generate a sawtooth waveform is not possible. The induct-
ance of the coils is, for example, 50 mH, and their resistance is in the region of, say,
50 n, which gives a time constant of only 1 ms.
The situation is considerably different in the line time base, since the scan con-
sists of two periods of 26 fJ.S each. Therefore the time constant of the scan coils is
sufficiently long to generate the required sawtooth, as will be shown in chapter 10.

9.4 S-correction and Scan linearity

So far it has been assumed that the scan current waveform is in the form of a per-
fectly linear sawtooth. The practical waveform is somewhat different because a
wide angle tube has an almost flat screen. Consequently, the deflection centre (the
point at which the beam begins to deflect) does not correspond to the centre of
the screen curvature. A 'flat' screen implies that its radius of curvature is relatively
long; therefore equal changes in the deflection angle-based on the deflection
centre-dO not result in correspondingly equal distances on the screen. This effect
is illustrated in figure 9.7(a)-the distances covered by the beam near the centre of
the screen are shorter than iliose away from the centre. A linear sawtooth would
produce a picture which was elongated at the edges. To correct for that, the rate of
change of scan current must be slowed down towards the extremities of the scan.
The resultant current waveform in the coils resembles a somewhat slanted letter
Synchronising Pulse Separator and Field Time Base Circuits 163

/
/
/

S-shaped //~/

---r
scan waveform // //
/ ,/
distance across
/ ,/
screen produced
//~/// by deflection
V;..-::: angle a
electron beam ________
from gun / ~---
-=.-:.. _..::
, ' , - ___ a f
,, , ,
centre of deflection " ' , -

" " increased

'~'
,
distance across
,, screen produced
by the same
(a) deflection
angle a

R L

scan
current

input voltage
waveform

required scan
cu rrent waveform

voltage across R
due to scan current

u
voltage across L

J due to scan current

required resultant
input voltage
waveform

(b)

Figure 9.7 Some of the factors to be considered in the formation of scan drive
waveforms: (a) S-correction, without which the picture would be stretched near
the edges of the screen; scan waveform is flattened towards its peaks, thus reducing
progressively the amount of deflection and producing a 'linear' picture; (b) 'trape-
zium' drive waveform compensates for the combined inductance and resistance of
scan coils
164 Television Principles and Practice

'S', and the correction which must be applied to the original sawtooth is called
S-correction. In field time base circuit S-correction is a part of picture linearity
correction arrangements within the field amplifier stage.
There are other factors which could cause picture distortion; consequently, the
amplifier drive waveforms must be further modified in order to obtain a linear
display on the screen. The effect of the time constant of the scan coils on the
shape of the current waveform has already been discussed. If the coils have a high
inductance (long time constant), the drive waveform would have to be rectangular;
if the inductance is very small (very short time constant), the required voltage wave-
form ought to be a sawtooth. Practical coils may need a trapezium-shaped voltage
waveform which, as shown in figure 9. 7(b), is the sum of a rectangular waveform
and a sawtooth. When the scan coils are coupled to the output stage via a trans-
former, waveform distortion can again be introduced. This is caused by the satura-
tion of the core owing to the combined effect of the no-signal direct current and
the superimposed peak of the scan waveform. The field output stage is a power
amplifier driven alternately between cut-off and near-saturation. Such a large signal
swing could easily introduce distortion if the bias conditions of the output valve or
transistor were incorrect. In fact, some degree of waveform distortion under these
circumstances is inevitable.
linearity controls are provided to compensate for the effects that the above
factors have on the scan waveform. Suitable shaping circuits pre distort the desired
drive waveform, so that the effect of subsequent distortion can be cancelled out.
linearity correction in the field time base is achieved by means of negative feed-
back. Capacitors and resistors are often used in the feedback path, which further
modify the shape of the signal before reintroducing it to the input. The complete
arrangement acts as a waveform adding network-the basic sawtooth is developed
by partly charging a capacitor and then adding suitable linearity correction wave-
forms. Their actual shape and amplitude are adjusted by linearity controls until a
correct picture is displayed on the screen.

9.5 Valve Field Time Base Circuits

Since the approach to circuit design differs between valve and transistor receivers,
it is necessary to consider the two types separately. The block diagram in figure 9.8
shows the basic functions of a typical circuit where valves are employed.
The trigger pulse shaping circuit develops a synchronising pulse for the field
oscillator from the pulses arriving from the sync. separator. Basically this consists
of some form of an integrating capacitor circuit, as described in section 9.2. Addi-
tional relative complexity at this stage ensures correct interlace; any unwanted
pulses that could otherwise cause premature flyback are often removed by an
interlace filter. An example of this is given in the diagram in figure 9.1 0 as part of
a complete time base circuit which will be discussed later.
Synchronising Pulse Separator and Field Time Base Orcuits 165

trigger pulse
sync. pulses _ _ shaping
frQI![Link]. circuit

11[gJ
separator

field field
oscillator

field charging output

(sawtooth) capacitor transformer

Figure 9.8 Block diagram of a valve field time base. The oscillator acts as a switch
allowing Cto charge viaR (long time constant), and to discharge via the oscillator
itself (short time constant)

Valve time base oscillators

The field oscillator produces pulses which allow the field charging (sawtooth)
capacitor to charge during the scan period, and then to rapidly discharge it to
initiate the flyback. The circuit behaves like a changeover switch connecting the
capacitor alternately to the supply line or to chassis. In valve field time bases the
most common form is a multivibrator or, occaSionally, a blocking oscillator.
The two basic forms of multivibrator are: anode to grid cross-coupled and a
cathode coupled circuit. The circuits in figure 9.9 show the fundamental arrange-
ments of each. The more frequently encountered cross-coupled multivibrator is
shown in figure 9.9(a). The oscillator is free-running as the valves are driven alter-
nately between cut-off and saturation. The oscillator starts up when the circuit is
first switched on, since the inevitable inequality between the two valves combined
with the anode-grid coupling drives, for example, V2 into saturation while VI be-
comes cut off. During this stage, the cross-coupling capacitor C 1 charges from h.t.
via Rl and the grid-cathode of V2. The situation is then reversed; VI becomes
saturated and V2 cuts off-now C2 charges from h.t. viaR2 and the grid-cathode
of VI. At the same time the previously charged capacitor C 1 is discharging through
the saturated valve, VI, and resistor R 4 • In effect, the negative plate of C 1 is
connected between the grid of V2 and the chassis; therefore the valve remains cut
off. The rate of discharge depends on the time constantR 4 C1 , so that after a cer-
166 Television Principles and Practice

mark space

output to
J ~ l
output waveform

C, charges
sawtooth
capacitor grid
cut·off - -
level I
I
V2 I :
waveform I C, discharges
alternative
L--.---Ir-- sync. on grid of V2 I I : '
sync. input input
: I
sync. ___ ...n.. __ n__ ~_.J1
I :

pulses (A) (8) (A) (8)

(a)

t----o output
synch. t-----<l output

nn
C,
input

output
waveform

(b) (c)

Figure 9.9 Basic forms of field time base oscillators: (a) cross-coupled multi-
vibrator; (b) cathode-coupled multivibrator; (c) blocking oscillator. NOTE: All
waveforms are idealised. Effect of sync. pulse timing is shown in (a): premature
pulse at (A) cannot trigger the oscillator

tain time interval the voltage on C 1 can no longer maintain the negative grid bias
and V2 switches on. The negative plate of the previously charged capacitor C2 is
now connected to the grid of VI and this valve is driven into cut-off. The time
constant R3 C2 governs the timing of the second changeover as C2 gradually dis-
charges through R3 and V2.
The basic multivibrator output waveform is rectangular in shape, owing to the
~ynchronising Pulse Separator and Field Time Base Circuits 167

mode volt(ige switching between the [Link] (cut off) and almost chassis potential
:saturation). The mark space ratio and the frequency of the output waveform
jepend upon the time constants of the cross-coupling components. Therefore, if one
Df the resistors is made variable, the oscillator frequency can be altered. Synchronis-
ttion of the multivibrator can be achieved by applying a suitable pulse to one of the
~ids. A positive-going trigger pulse applied to V2 while the valve is cut off overrides
the negative bias of C 1 and switches the valve on. Alternatively, a negative-going
,ynchronising pulse could be applied to the grid of VI to take the valve out of
;aturation into cut-off, which would, in turn, switch V2 on. The timing of the
trigger pulse is important, as indicated on the oscillator waveforms in figure 9.9(a).
Synchronising pulse A is applied too soon, so that it cannot overcome the cut-off
Jias on the grid of V2 and synchronisation does not take place. Pulse B, however,
is timed to occur just before the 'free-running' changeover point; consequently V2
can be triggered into conduction. This system ensures that the oscillator is relatively
immune to interference pulses occurring during the scan, but it also means that the
multivibrator free-running frequency must be close to, although slightly lower than,
the required value.
A cathode-coupled multivibrator is shown in figure 9 .9(b). There is only one
cross-coupling capacitor (Cd, the other feedback connection being between the
cathodes. When C1 is charging from the h. t. via R 1, triode V2 is driven into satura-
tion. Its high current produces a voltage drop across R 5 of such a magnitude as to
keep the other valve, VI, cut off. When the capacitor is fully charged, the current
in V2 falls, the cathode bias reduces and VI starts to conduct again. This, however,
effectively places the negative plate of Cion the grid of V2, which, in turn, switches
the valve off. The reduction in the current of the second triode further reduces the
:::athode bias; VI conducts more and is driven into saturation. The capacitor is now
discharging via VI, Rs and R4 until a point is reached when its negative charge is
no longer effective and V2 is again allowed to conduct; recharging C 1 repeats the
above process. A negative-going synchronising pulse can be applied to the grid of
VI to take it out of saturation.
Another form of pulse generator occasionally found infield time base circuits
is the blocking oscillator shown in figure 9 .9(c). This requires a transformer, which
could be an added disadvantage, since at an operating frequency of 50 Hz the size
and cost may not be justified. When first switched on, the valve conducts heavily
because of lack of cathode bias-the rising anode current induces an e.m.f. in the
,econdary winding of the transformer. The direction of the e.m.f. is such that it
charges capacitor C1 via the grid-cathode circuit of the valve. Once the valve reaches
,aturation and its output voltage is nearly at chassis potential, no further increase
in the anode current is possible and the e.m.f. in the secondary winding falls to
zero. In effect, this places C 1 between the grid and the cathode-the negative
potential on the capacitor cuts the valve off and the anode voltage rises to h.t.
The capacitor slowly discharges via R 1 until the negative bias is sufficien tly low to
cause the valve to start conducting again. A positive-going synchronising pulse can
be applied to the grid to overcome the cut-off bias and trigger the oscillator.
168 Television Principles and Practice

Practical field time base circuit

The circuits discussed in figure 9.9 were the basic fonns of the theoretical arrange-
ments. Their practical versions often differ to such an extent that it can be diffi-
cult to recognise the original principle. In older designs a simple multivibrator was
followed by the power output stage. The tendency in more modern circuits is to
combine the multivibrator with the output amplifier and thus eliminate one valve.
A triode-pentode is used, the triode section acting as one part of the multivibrator,
while the pentode serves the dual purpose of being the second part of the oscillator
as well as the field output amplifier.
An example ofa complete field time base circuit is given in figure 9.10. Negative-
going pulses from the sync. separator are fed to the integrating circuit R 1 , C 1 which
develops the field trigger signal (waveform A-the 'serrations' from figure 9.5 could
be seen if the oscilloscope time base were set to display line frequency pulses). The
interlace filter is formed by R2 , D 1, C2 and C3 • Its function is to establish a voltage
level at which the field synchronising pulse initiates flyback. The interlace diode,
Dl, clips the pulse developed across C 1 and allows it to be coupled to the oscillator
via C3 • When the diode conducts, a negative trigger pulse is fed to the grid of V2
which cuts the valve off.
The two valves (actually a triode-pentode valve) fonn a multivibrator as well as
the power output amplifier. Because of the presence of the amplifying stage, the
circuit departs from that shown in figure 9.9. In the practical circuit in figure 9.10
VI is driven alternately between cut-Qff (scan) and saturation (flyback) by means
of the cross-coupling from the anode of V2. The coupling via C9 and the network
of RIO, R 11 , R 12, R 13, C 10 and C 11 ensures the correct timing of the oscillator
waveform. Manual frequency control is obtained by VR1-thefield hold control.
However, the second valve, V2, is no longer switched in the typical multivibrator
fashion. Instead, its grid is fed with a SUitably shaped drive wavefonn which the
pentode amplifies and delivers the current to the field output transformer, Tl.
The drive waveform starts as a sawtooth (wavefonn C); this is due to the field
charging (sawtooth) capacitor Cs , which is charged from the boosted [Link] via
R9 and VR2. The capacitor charges when VI is maintained in its cut-off state by
the negative portion of wavefonn B. The field charging circuit is an integrating
circuit with a relatively long time constant. Assuming that VR2 is set to its mid-
value of 1 Mil, then the time constant T= (R9 + VR2) X Cs = 1.47 Mil x 47 nF =
69 ms, which is longer than the scan time of approximately 20 ms. The boosted
h.t. originates from the line output stage; this voltage is much higher than the
normal h.t. and it can be anything between 400 V and 1000 V. The use of the
boosted h.t. supply has a number of advantages. Firstly, Cs can charge to a much
higher voltage and still follow the more or less linear portion of the charging graph
(figure 9.4). Secondly, a high potential across Cs will suffice to drive the output
valve fully and without any intennediate amplification. Finally, the boosted h.t.
has a relatively high output impedance, so that simple voltage stabilisation by means
of a voltage-dependent resistor, VDR2, is satisfactory. Regulating the supply at this
 • • h.t.
flyback blanking

Cg

scan
coils

,;tC l3

_ sync. pulses in

V\;ov,-, ~3VP-P
A B c D
Hv,-, Dv"
n U v ,-, v ,-,
\,/1v,-,
E F G

Figure 9.10 Valve field time base circuit diagram. Expected waveforms are shown
by their corresponding letters in the diagram (Rank)
170 Television Principles and Practice

point helps to maintain constant height of the picture. The magnitude of the voltage
across Cs governs the drive to-and the output from-the power amplifier; therefore
VR2 is the height control.
The sawtooth as developed so far is further modified to ensure picture linearity
(see section 9.4). The modifications to the waveform produced across Cs are carried
out by the linearity correction network. Since it is impossible to design a fixed
circuit to provide the required waveform shaping, two variable controls are incor-
porated: VR3 (so called 'top linearity') and VR4 (,overall linearity'). The correction
is applied by means of negative feedback from the anode ofV2 to its grid via C g ,
R 7 , R g , C6 , VR3, R 4 , and Rs. These components alter the shape of the waveform
fed back, which is then added to the basic sawtooth coupled in via C7 . A second
linearity correction circuit, formed by Cg , R6 and VR4, is effectively placed across
the output from the valve to modify the waveform applied to the output trans-
former TI.
The field charging capacitor, Cs in figure 9.10, is connected to the cathode
bypass capacitor C4 in preference to a direct connection to the chassis. This intro-
duces yet another form of linearity correction. At first glance it appears that C4
bypasses the cathode resistor R3 in the usual way. In fact, another waveform is
generated across the capacitor which, together with the slope resistance ofV2,
forms a long time constant integrating circuit. The basically sawtooth cu"ent flow-
ing in the output stage produces a voltage waveform called a parabola (waveform D).
This parabolic waveform is, in effect, added to the input to improve the linearity.
Waveform E applied to the grid of V2 consists of a suitably shaped 'sawtooth'
and a large negative pulse of a relatively short duration. This pulse cuts V2 off to
initiate the flyback, which, in turn, causes a high amplitude positive pulse to appear
at the anode of the valve (waveform F). The effect is produced by the sudden inter-
ruption of the scan current and by the energy stored in the scan coils. If this energy
is not quickly dissipated, the output circuit can be shocked into self-oscillations
(ringing) that could interfere with the flyback as well as continue into the next
scan. Damping is provided by R 15 and to some extent by VDRI across the primary
of the output transformer. The voltage-dependent resistor reduces the amplitude
of the flyback pulse, because the resistance of the device falls rapidly as the voltage
across it increases. If the pulse magnitude was not restricted, the insulation of the
associated components, chiefly that of the capacitors and the transformer, could
be damaged. The flyback pulse across the secondary of the transformer is shaped
via R 16 and C 13 , and then applied to the video amplifier to provide field [lyback
blanking (see chapter 6).
The field scan coils are connected to the secondary of the output transformer
via a thermistor, R 17. Physically the thermistor is attached to the coils to sense
their temperature. As the temperature increases, the resistance of the scan coils
rises, which tends to reduce the amplitude of the scan current and, hence, the
picture height. The resistance of R 17 falls, owing to the negative temperature
coefficient (n.t.c.) of the device, so that the effective resistance of the entire circuit
remains constant. Waveform G across the thermistor represents the current in the
coils, although it is not normally given in the makers' manual.
Synchronising Pulse Separator and Field Time Base Circuits 171

9.6 Transistorised Field Time Base Circuits

The block diagram in figure 9.11 shows the circuit arrangement of a typical transis-
torised field time base. Some of the functions can be easily recognised by compari-
son with the valve counterpart discussed in the previous section. Now the oscillator
tends to be an independent unit followed by a multistage amplifier. In most cases
there is no output transformer; the scan coils are coupled via a suitable capacitor
to the output stage.

from
sync.
separator

field linearity driver

oscillator amp. amp.

' ' ';'"' I

field charging C
linearity
correction

Figure 9.11 Block diagram of a transistorised field time base

Field oscillator
The field oscillator in some receiver designs may be either a cross-coupled multi-
vibrator or a blocking oscillator. The principles of operation are similar to those of
their valve versions described previously; any differences are primarily due to (he
d.c. biasing requirements of transistors. However, a somewhat modified multi-
vibrator circuit is shown in figure 9. 12{a). When first switched on, the cross-coupling
capacitor C1 charges up via R2 and the base-emitter junction of TR2. The charging
current drives TR2 into saturation, which, in turn, causes TRl to cut off owing to
the lack of forward bias. When C1 is fully charged, TR2 is left without any forward
base bias and the transistor switches off. Transistor TRI turns on (its bias restored
via R s ), which allows C 1 to discharge through R 3 , YRl ('hold' control) and TRI.
The capacitor again recharges via R2 and TR2, so that the above sequence is repea-
ted. The oscillator output is in the form of short duration, positive-going pulses
developed across Rs. The charging time constant associated with C1 is short, which
corresponds to the flyback, while the discharge time constant is long to allow for
 172 Television Principles and Practice
--.-----------+- + h.t.

TR2

TRl

Jl.
sync. field time
pulses in ........_ _ _-II-_ _-' base trigger C,
pulses out

R5 10k R2 Jl
output

(a) (b)

sync. trigger
pulses in pulse out

(c)

Figure 9.12 Field time base multivibrator circuits. These arrangements trigger the
field charging capacitor charge/discharge circuits: (a) field time base multivibrator
(Thorn); (b) complementary symmetry multivibrator (basic circuit); (c) complemen-
tary symmetry multivibrator (Thorn)

the duration of the scan. Positive-going synchronising pulses switch TRI off, the
charging of C1 begins and the resultant output pulse initiates field flyback.
The fundamental principle of another form of oscillator is illustrated in figure
9.12(b). Transistors TRI and TR2 are complementary types and their method of
interconnection resembles that of a two-transistor equivalent of a thyristor (see
Synchronising Pulse Separator and Field Time Base Qrcuits 173

section 14.6). When the circuit is first switched on, C 1 charges via R 1 ; therefore
the emitter of TRI is close to the chassis potential and the transistor remains· cut
off. Similarly, TR2 is non-conducting, because of the absence of any forward base
bias. The charging process of C1 fmally raises the emitter voltage of TRI above the
base potential derived from R4 , R 5 and the transistor turns on. This switches TR2
on, and the two transistors are driven rapidly iilto saturation as they provide base
bias for each other. The resultant current produces a positive-going output pulse
across R 3 • Capacitor C 1 is discharged through the conducting transistors; conse-
quently the emitter voltage of TRI falls below the base voltage, both TRI and
TR2 cut off and the capacitor charging process starts again. Synchronisation can
be obtained by the application of a 'turn-on' pulse to the base of one of the transis-
tors. Field hold control could be provided by making R 1 variable.
The circuit shown in figure 9. 12(c) is based on the above principle. Initially,
however, both transistors are switched on, owing to their respective biasing arrange-
ments. During that time capacitor C 1 charges from h.t. via R 3 , the forward biased
base-emitter junction ofTR1, Dl,R, and TR2 (the path ofleast resistance). When
C1 is fully charged, its negative plate is in effect connected to the emitter of TRl,
which causes the transistor to turn off; the base bias to TR2 is now interrupted and
this transistor also switches off. Capacitor C 1 slowly discharges via R 2 , VRI (field
hold control), the h.t. decoupling capacitors (not shown in the diagram),R 3 and
R 6 • Eventually the capacitor voltage falls to a value where Dl becomes forward
biased, so that the conduction of TRI and TR2, as well as the recharging of C 1
start again. The output from the oscillator is in the form of short duration negative-
going pulses that occur during the conduction periods of the transistors. A positive-
going synchronising pulse applied to the base of TR2 switches the transistor on,
which allows the charging of C1 to take place.
A number of receivers use a form of thyristor, known as a silicon controlled
switch (SCS 1), in the field oscillator as shown in figure 9.13. (The principle of
operation of the thyristor is described in detail in chapter 14.) In this circuit the
thyristor has two gates available: the cathode gate (connected to Rs) and the anode
gate (connected to the junction of Rl andR2).1n operation the SCS is not con-
ducting until the anode is positive with respect to the anode gate. In the circuit in
figure 9.13 the anode gate potential is fixed by the divider R 1 , R 2 , but the anode
voltage is variable because of the repeated charging and discharging of capacitor C 1 •
As C1 charges slowly via VRI (field hold) and R 3 , the anode voltage is low and the
SCS is held off by its gate potential. However, a point is reached when the SCS is
triggered into conduction as the anode voltage rises above the gate potential. During
this period the timing capacitor C1 is rapidly discharged through the SCS (and so
is the field charging capacitor C2-to initiate flyback). Positive-going synchronising
pulses are applied to the anode of SCS I to advance its 'firing' point; diode D1 pre-
vents the synchronising pulses being diverted viaR 4 and C 1 to chassis. Alternatives
to the arrangement described here do exist, but their operation is always based on
the provision of a suitably timed change in the voltage level between the anode and
the anode gate of the SCS.
 R'G + h.t. blanking
pulses to
video amplifier

R,
~

TC,_ ;=C 2

01 ~
field s~nc. .. I •
pulses .n '

VR5

field oscillator sawtooth linearity driver output

generator amp. amp. amp.

~V ~ fV\ $J tej
-{)/ 10V
o
A B C 0 E F G
Figure 9.13 Transistorised field time base circuit diagram. Expected waveforms and their peak-to-peak values are shown
encircled. (Philips)
Synchronising Pulse Separator and Field Time Base Circuits 175

Some field oscillator circuits have no vertical hold adjustment, although a suit-
able modification can be sometimes introduced if a field hold control is required.

Practical field amplifier circuit

At first sight the field time base amplifier circuit shown in figure 9.13 tends to
resemble an audio output stage (see chapter 8). The loudspeaker is now replaced
by the scan coils! The reader is asked to refer to section 8.3 (and figure 8.5) and
note the discussion on crossover distortion and bias stabilisation in this type of
circuit. Additional features of the amplifier in figure 9.13 dictated by the require-
ments of the field time base will now be discussed.
The field waveform is initially developed across the field charging capacitor C2
connected to the h. t. through VR2 (height control) and R6 . A linearity correction
waveform is fed back from the emitter of TRI via RIO and VR4 ('top linearity
contro!'); it is shaped by the integrating capacitor C4 and coupled in via C3 • A para-
bolic waveform appears at the collector of TRI because of the action of the inte-
grating circuit R 12 , C6 • A sample of that waveform is also fed back via R 9 , Cs , Rs
and VR3 ('bottom linearity contro!').
Field scan takes place when SCS 1 is switched off and C2 is charging. The time
constants associated with C 1 and C2 , respectively, are so arranged that the charging
potential across C2 is lower than that on C 1 and the oscillator isolating diode D2 is
biased OFF. To initiate flyback, SCS 1 is switched on, D2 becomes forward biased
and C2 is rapidly discharged.
The output from the linearity amplifier, TR1, is coupled to the driver transistor
TR2. Stabilisation of its d.c. operating point is achieved by feeding the base bias
current from the junction of the output transistors, TR3 and TR4, via the scan coils
and the voltage adjustment VRS. Additional linearity correction takes place by
feeding back, via R 22 , a sample of the output waveform developed across the low
value resistor R 26 (1.2 n in this design). There is also a feedback connection from
the output through the so-called bootstrap capacitor Cs .
D.C. biasing of the output transistors, TR3 and TR4, is accomplished by the
combined effect of TR2, VR6 (eliminates crossover distortion), R 21 , bias stabiliSing
thermistor R 20 , R 19, R IS and R 17' During the first half of the scan the drive wave-
form makes TR3 conduct while TR4 is switched off. The second half of the field
scan is completed by the conduction of TR4, but the transition from one output
transistor to the other is gradual in order to avoid crossover distortion. If this were
present, non-linearity would appear in the middle part of the picture. However, any
adjustments to the bias resistors (VR5 and VR6 in figure 9.13) must be carried out
according to the manufacturers' instructions. There could otherwise be a danger of
overdriving and possibly destroying the output transistors.
Field flyback is caused by the rapid discharge of C2 , which switches off TRl,
whose negative-going output pulse turns off the driver (TR2) and, fmally, the out-
put transistor TR4 cuts off as well. The sudden collapse of the magnetic field in the
scan coils induces a large back e.m.f.; its polarity tries to keep the current flowing
176 Television Principles and Practice

towards the junction of R24 and R 2S (that is, in the same direction as at the end
of the scan). The magnitude of such a pulse could be damaging to the output tran-
sistors; however, the flyback pulse turns D3 on, so that the back e.m.f. from the
scan coils is clipped at a level just above the supply voltage (see the output wave-
form). The flyback pulse is also fed via Zener diode D4 and the shaping compon-
ents CIO and R2 7 to the video amplifier to provide retrace blanking, as described
in chapter 6.
A number of receivers use a single transistor output stage. The scan coils are then
fed via a double-wound transformer or an autotransformer, or directly from the
collector decoupled from the h.t. by means of a choke. A suitable diode circuit is
again included to damp the flyback pulse to a safe value, and the action of the whole
output stage is very similar to that of its valve counterpart.

9.7 Integrated Circuits in the Field Time Base

I.C.s can be used to provide synchronising pulse separation, the field time base
oscillator, the driver and the field output amplifiers. External components are
necessary to provide, in addition to d.c. biasing and decoupling, such functions as
integration of the composite field trigger pulse, field charging capacitors, height
and hold controls, linearity correction feedback circuits, etc. There are, of course,
a number of combinations of different functions in any particular type of I.C. to
suit the designer's approach.
An example of an integrated field time base is shown in figure 9.14, where two
devices are used to complete the circuit. The reader is advised to refer to the other
arrangements where discrete components were used in order to recognise many
similarities in the operation of the circuits.
Integrated circuit I.C. 1 in figure 9.14, apart from being shared with the line
time base to be discussed in chapter 10, includes the synchronising pulse separator
and the field oscillator. Composite video is fed in at pin 9 (I.C. 1) via C1 , where the
capacitor together with resistor Rl is part of the sync. separator time constant
(compare with C1 and R3 in figure 9.1). A low value capacitor C2 (470 pF) shunts
any high frequency content of the video, e.g. colour burst and noise. The sync.
separator supplies its output pulses to the line oscillator (top half of I.C. 1 in the
diagram) and to the field oscillator (bottom half of the circuit). The integration of
the field trigger pulse and the timing of the free-running frequency is determined
by C3 and VR1 (field hold control) connected to pin 12.
The field sawtooth is developed by charging C4 and Cs from h.t.2 via the height
control circuit-VR2 and R 2 • The resultant waveform, including linearity feedback
connected to the junction of the two capacitors, is applied to pin 7 of I.C. 2. The
latter is a power amplifier which drives the field scan coils coupled to pin I via
Cs . The waveform developed across the small value resistor Rs is then fed back via
R4 and VR3 (linearity control) to the input in a similar way to that shown in
figure 9.13 (see components R 26 and R22 , also VR4 and RIO)' Diode D 1 between
the field charging capacitors (C4 and Cs ) and the field oscillator is the isolating
 h.t.l
h.t.l pulses from J. h .t . l field flyback
line --, 9 C,o
T ~o blanking pulse
output
line
VR4~to line
hold driver
I.C.2
TBA64l-BXl C9

o@ o@ o@ !@ o@@o

I.C.l
SN76544N-07

h.t.l

driver I o@ @ ® ® o@ @ 00 CD
;I c,
h.t.l
21
c3

7.

Figure 9.14 Complete field time base using I.C.s. Circuits include part of line time base and the sync. separator. (Thorn)
178 Television Principles and Practice

diode (equivalent to D2 in figure 9.13). Diode D2 connected across the output

from I.C. 2 provides positive-going pulses for field flyback blanking (flyback blank-
ing pulses could also be obtained from pin 14 of I.e. 1). Capacitor C9 is the 'boot-
strap capacitor' (equivalent to Cg in figure 9.13); resistors R6 and R 7 -pin 8-
provide d.c. biasing for the I.C.; C6 is a decoupling capacitor. R3 and C7 form a
filter to prevent possible high frequency instability in the amplifier.

9.8 Class D (Switched Mode) Field Time Base

The previously discussed class A and class B field output stages dissipate large
amounts of power in the output devices. This is particularly high in colour receivers
employing 110° tubes, because of their additional convergence circuitry. These
power requirements make it difficult to provide a suitable I.C. for such a receiver.
A switched mode field time base reduces the power consumption of the stage
and, consequently, most of the circuitry can be accommodated in an I.C. (for
example, Mullard TDA 2600). This is achieved by using the output transistors as
switches which are pulsed either ON (high current, low voltage, low power loss) or
OFF (no current, no power loss). The drive waveform to the output devices is a
train of 150 kHz pulses of continuously changing mark space ratio. The resultant
output pulses are converted into the scan sawtooth by an exiernal filter consisting
of a suitable choke and capacitor (see the discussion on integrating circuits). Exter-
nal energy recovery diodes are provided to maintain the flow of current in the scan
coils when the output transistors are cut off by their input pulses.
The width of the high frequency drive pulses is varied at the required field rate-
that is, 50 Hz. The I.C. contains a 150 kHz triangle waveform generator, the output
of which is modulated by a separately derived 50 Hz sawtooth. Their timing com-
ponents are external to the I.C. This type of modulation is known as pulse width
modulation: at the beginning of the scan the 'on' period is short, while the 'off
interval is long; as the scan progresses, the 'on' period becomes longer at the expense
of the 'off interval.

9.9 Fault-Finding in Sync. Separator and Field Time Base Circuits

Faults in the synchroniSing pulse separator circuits affect the quality of time base
synchronisation. Because of the interaction between the various sections of a
receiver, it is possible that a fault outside the separator circuit could produce
similar symptoms. The necessary process of eliminating the effect of 'external'
stages could be helped, for example, by observing the quality of the picture (and
sound!), the effect of contrast control, time base 'hold' controls, etc. Preset adjust-
ments should, however, be carried out in accordance with the makers' instructions
where applicable.
In principle, faults in the sync. separator circuit produce either no pulse output,
or pulses of insufficient amplitude or of a distorted shape or incorrectly timed. The
resultant symptoms can be divided into the following groups: (1) loss of both line
Synchronising Pulse Separator and Field Time Base Qrcuits 179

and field synchronisation, (2) loss of only one time base synchronisation, (3) line
'pulling' and (4) field 'bounce'.
To ensure both the correct biasing and the required time constants, the values
of components used in the separator circuit are quite critical. In addition, the
characteristics of the valves or transistors in this type of circuit are important. The
only satisfactory way of checking the valves is by substitution, but transistors and
I.C.s are less prone to gradual changes, and should not be removed unnecessarily.
The indirect method of line time base synchronisation (flywheel) often causes only
the field time base to be affected by sync. separator faults.
Faults in the field time base can be divided into a number of groups as shown
below, although at times at least two effects may be present. (It is assumed that the
possibility of faulty valves has already been considered.)
(1) Incorrect field frequency. Apart from faults in the sync. separator (or elsewhere
in the received), this could be caused by a defective field oscillator-in particular,
in the trigger pulse circuit or the oscillator timing components (the latter often
leads to a critical field hold.
(2) Loss of vertical deflection ('frame collapse'), resulting in a thin horizontal line
across the screen. This can be caused by a failure of either the field oscillator or the
field output stage (including the scan coils, or even the convergence circuitry in
colour receivers). The defective section can be easily isolated with the aid of an
oscilloscope,However, a voltmeter can often indicate whether the oscillator is
functioning, since the resultant waveform usually develops a reverse grid (or base)
bias. If the oscillator output could be fed to the audio amplifier via, say, a 0.01 f.1F
capacitor, a low frequency harsh note from the loudspeaker would confirm the
presence of a field waveform. The field output stage could be tested by applying a
suitable signal to its input. In valve time bases a.c. mains waveform can be injected
to the grid of the output valve through a 0.1 f.1F capacitor from the lower voltage
end of the heater chain. Should the display open up after a fashion, then the out·
put is functioning. Alternatively, touching the input terminal to the field amplifier
with a meter test prod may result in a momentary 'bounce' of the horizontal line if
the stage is not faulty. Any such short-cuts must be carried out with some foresight,
especially when applied to transistor or I.C. stages, as semiconductor devices can be
damaged by excessive amplitUde 'test' signals. Faulty field output transistors are
often caused by a defect in a different part of the circuit-particularly, in directly
coupled stages, which can easily produce grossly incorrect bias conditions.
Whenever investigating a frame collapse type of fault, it is essential to reduce
the brightness to a minimum; alternatively, the beam could be cut off by removing
the drive to the tube, switching the guns off or similar measures. This is necessary
because the energy of the tube current will be concentrated over a very small area
in which the screen material could be damaged. Some receivers have a protection
arrangement that automatically cuts the beam off in the absence of a field drive.
(3) Insufficient height. This is often associated with poor picture linearity or pic-
ture foldover. The fault area is difficult to predict, because both the oscillator and
the field amplifier stages can be responsible. An oscilloscope is valuable in establish-
180 Television Principles and Practice

ing whether correct amplitude waveforms are present. In valve circuits the common
cause of this type of fault is low h.t. (or boosted h.t.) supplying the field charging
circuit; alternatively the resistors in the height control circuit often increase con-
Siderably in .value.
(4) Poor linearity. This can be associated with some lack of height, as mentioned
above. This defect often affects only part of the picture-namely cramping at the
top, cramping at the bottom or compression at the centre. The latter can only be
found in transistor circuits, owing to the presence of crossover distortion. Non-
linearity can be caused by incorrect d.c. biasing in field amplifiers or faulty linearity
correction components. In valve field output stages the cathode resistor and its
bypass capacitor are a common source of trouble. The resistor fixes the d.c. bias,
while the capacitor provides a correction waveform; thus an open circuited capaci-
tor causes cramping at the bottom of the picture together with a reduction in
height. Similar symptoms can occur if the capacitor (or resistor) is short circuited,
since the valve would be driven into its non-linear region during the second half of
scan. A high value cathode resistor causes non-linearity at the top, because of the
excessive negative grid bias voltage at the beginning of the scan. Generalisations
on fault symptoms in transistorised circuits are difficult, because of their relatively
greater complexity. The underlying principles, however, are similar, except per-
haps for the flyback circuit, where a number of components are involved-both at
the field output and in the field oscillator stage.
(5) Miscellaneous faults.
(a) Poor interlacing. This exaggerates the line structure of the picture; it can be
caused by a fault in the interlace filter (where present) or the pulse integrating
components, allowing interference or stray line flyback pulses to trigger the field
time base oscillator.
(b) Trapezium distortion. Either the top or the bottom edge (or both) of the
picture is slanted, which causes part of the screen not to be scanned-this is due to
a fault in one of the field scan coils. In colour receivers a fault in the raster correc-
tion network can be responsible (see under 'symmetry control' in chapter 11).
(c) Pincushion distortion. 'Bowed' top and bottom edges of the picture-deferred
to chapter 11.
10 Line Time Base Circuits

10.1 Line Time Base Synchronisation

Horizontal scanning of the screen is under the control of an oscillator which is
synchronised with the action of the electron beam of the camera tube at the trans-
mitter. In the 625-line system the line repetition frequency is 15 625 Hz, and each
line period of 64 p.s consists of 52 p.s of actual picture information (active picture
line) and 12 p.s line blanking. During the latter period there is a line synchronising
pulse which lasts 4.7 p.s. These timings are important, because they explain the
different design of line time base circuits when compared with their field counter-
parts described in chapter 9. The common factor between them is the requirement
to provide an approximately sawtooth deflection current waveform in the scan
coils.
The problems involved in the synchronisation of the line oscillator will now be
discussed. It was explained in chapter 2 how interference pulses can affect the
synchronising pulses when negative picture modulation is used, since the amplitude
of any such noise rises towards the synchroniSing pulse level. Synchronisation of
the field time base is not seriously affected by interference, because of the presence
of a trigger pulse integrating circuit. The capacitor which develops the field trigger
pulse tends to 'absorb' short term spurious noise. This is not possible in the line
time base, because of the nature of the synchronising pulses; here they are of
relatively short duration-and so is the interference. We saw in section 9.2 that the
line synchronising information is sorted out from the field pulses by means of a
differentiating circuit. Such a circuit has a short time constant; therefore when a
narrow interference pulse is applied to the input, a differentiated pulse will appear
at the output. If the line oscillator were triggered directly from the differentiating
circuit, the synchronisation of the time base would be disturbed by the unwanted
'spikes'.
Modern receivers use a form of buffer stage between the sync. separator and the
line oscillator which offers considerable immunity against incorrect triggering. This
method is called flywheel synchronisation, because its effect in the circuit is similar
to that of a flywheel in a mechanical system whose speed of rotation is maintained
relatively constant.
Precise line synchronisation is obviously necessary to ensure a viewable picture;
however, it is also needed for correct gating of the colour burst and for the function-
ing of gated a.g.c. where applicable.
181
182 Television Principles and Practice

sync. sync. pulse

I n ~
d.c. control
pulses
from separator
~ polarity
splitter
line sync.
discriminator filter r- oscillator
to line

I l j ~
sync . pulses
to fie Id time base 1J pulse pulses from lin e output
integrati ng __ - - - - - transformer (0 r sample of
circuit
lJ line oscillator output)

(a)
r - -...... - - - - - - - - - - - - - _ - + h.t.

to line
from sync. oscillator
separator - - - I I

----- ----
C~[Link]
GOo.P.T.

phase
splitter
flywheel
discriminator
_--.........---' ---......-
-.... flyb;;'ck
pulse
filter

(b) integrator

;~v~~orm _A~b~cA __ zerolevel ____ __ __ ~ ~

L
Junction

l L
.,....--~--[/ 7--~---~
i

l
of Oland 02 I I I

~,":'i:m"~ ~;:; D1

current through
02via C3 r I (c)
~~;~~~~ 02 r (d)
r
Figure 10.1 Flywheel synchronisation of line time base: (a) block diagram;
(b) practical circuit diagram (Thorn); (c) and (d) diode sampling of the integrated
line output waveform. In (c) synchronisation is correct-diodes sample zero level of
integrated flyback waveform. In (d) synchronisation is incorrect-flyback occurs
too soon, diodes sample negative part of flyback sawtooth. Note: Current in diodes
is due to sync. pulse differentiation
Line Time Base Qrcuits 183

A block diagram and the corresponding circuit arrangement of flywheel line

time base synchronisation are shown in figure 10.1. The purpose ofthe circuit is
to compare the phase and frequency of the synchronising pulses with those of the
line time base oscil4ttor. This is done by checking the coincidence of the actual
flyback with the synchronising information. Any discrepancy generates a d.c. con-
trol voltage to adjust the oscillator frequency until the error disappears.
In figure 10.1(b) pulses from the sync. separator pass through the polarity splitter
amplifier, TRI, so that equal but opposite outputs are obtained at the collector and
the emitter, respectively. TRI provides a switching waveform for the diode line
synchronising pulse discriminator (D1 and D2). When the two diodes are simul-
taneously biased ON, they charge the coupling capacitors C2 and C3 • The capacitors,
in conjunction with the low resistance of the diodes, provide short time constant
differentiating circuits during their charging period. In the absence of synchronising
pulses, TRI is switched ON, as implied by its waveforms; D1 and D2 are biased
OFF by the charges on C1 and C2 • The capacitor discharge path is via the transistor
and the loop formed by R4 and Rs. Because of the symmetry between the top and
bottom half of the circuit, the current due to each capacitor is equal, and the potential
differences across R4 and R 5 are equal in value but of opposite polarity. Therefore
the junction of R4 and R 5 is at zero potential with respect to chassis-in effect,
there is no output from the circuit. However, the junction of the two diodes is fed
with integrated flyback pulses from C4 in addition to a d.c. potential derived from
VRI. The presence of these two inputs modifies the output from the discriminator.
The integrated flyback pulses indicate the actual timing of line flyback and, in-
directly, the frequency and phase of the line oscillator. Initially, the pulses from
the line output transformer, at Cs , are rectangular in shape, which is changed into
sawtooth by the action of the integrating circuit R 9 , C4 • The coupling capacitor,
Cs , ensures that the waveform has no d.c. component, so that the sawtooth also
varies about the zero level [figure 10.1(c) and (d)] . The time base circuit is set up
so that when the oscillator frequency is correct, both Dl and D2 conduct at the
time when the integratedjlyback sawtooth passes through zero, as shown in figure
10.I(c). Therefore the flyback waveform does not contribute anything to the
output from the discriminator and the line oscillator remains unaffected. If, how-
ever, owing to an incorrect timing of the flyback, the diodes conduct when the
integrated sawtooth is, for example, negative [figure 10.1 (d)] , the balance of the
entire circuit will be upset (D1 conducts more than D2). Consequently, the dis-
criminator will also produce a negative-going output to correct the oscillator fre-
quency accordingly. Conversely, a positive control voltage will appear if the diodes
switch during the positive portion of the integrated waveform.
The application of a d.c. voltage to the junction of D1 and D2 results in a d.c.
level at the output from the discriminator. This serves as a d.c. bias for the line
oscillator and any correction voltage swings above or below that fixed value.
Effectively, VRI alters the oscillator frequency and it becomes a line hold preset.
It should be adjusted according to makers' instructions; usually this means that
VRI is set to produce a specified output voltage from the discriminator while the
184 Television Principles and Practice

sync. separator has been disabled.

The output from the diode circuit is smoothed by Rs and C6 , followed by a
damping filter RIO and C7 to prevent 'hunting'. 'Hunting' would cause the line
oscillator frequency to rise and fall before finally settling down-for example,
whenever channels were changed over.
The flywheel effect is created by the value of the smoothing filter components.
Any short-term changes in the output from the discriminator are simply absorbed
by capacitor C6 in figure 10.1 (b). Similarly, variations in the line synchronising
pattern during the field synchronising sequence are also swamped by the filter. Only
long-term changes can alter the charge on the filter capacitors sufficiently to affect
the line oscillator.
A number of variations on the above circuit are possible. In some receivers
polarity splitting can be performed by a centre tapped transformer; in others there
may be two antiphase feeds from the line output transformer and only single
polarity synchronising pulses.
In the above-mentioned circuits three separate signals were needed. A somewhat
simpler alternative is shown in figure 10.2; there synchronising pulses are applied
directly from the separator. The underlying principle is similar to that of the pre-
vious arrangement, but the actual mechanism of its operation is different. The
coupling capacitor, C1 , having been previously charged, is quickly discharged by
the negative-going synchronising pulse when the two diodes are switched on. If at
the same time the flyback sawtooth passes through zero [see figure 10.1 (c)] ,C1
becomes fully discharged; should the sawtooth be, respectively, either positive or
negative, the coupling capacitor will acquire a net charge of the resultant polarity.
This charge is then transferred to the output, where, after smoothing, it becomes
a d.c. control voltage.

output to
,-----1>-----1- filter. etc.
(see figure 10.1)

output transformer

Figure 10.2 Line sync. pulse discriminator {alternative arrangement)-polarity

splitting no longer needed
Line Time Base Circuits 185

There are several versions of this type of arrangement. For example, in some the
sawtooth feed may be applied at the output terminal; in others provision for adjust-
ing the control voltage (,line hold') can be included.
The flywheel discriminator can also be contained inside an I.e. An exampl~ of
such a circuit was given in the previous chapter, where it was combined with the
sync. separator, the line oscillator and other functions. The reader is asked to refer
to figure 9.14. Flyback pulses from the line output transformer are fed in at pin 6
and, after suitable shaping, they are communicated to the phase comparator,
which performs the function of the discriminator. Here the pulses are not integrated
as they were in the previous circuits, since the comparator 'looks' at the displace -
ment between the edges of the synchronising and the flyback pulses. The output is
then smoothed by the filter capacitor, Cll , and the 'anti-hunt' network R 9 , C IO
connected externally to the I.e. (pin 7). The resultant d.c. voltage, controls the line
oscillator.
Other I.e. types are also pOSSible; a different external component arrangement
would then suggest an alternative design approach to circuit operation.
Semiconductor flywheel circuits are now commonly used. Slightly older receivers
employed a synchronising pulse coincidence detector in the form of a valve-for
example, a triode-pentode. The synchronising pulses could be applied to the grid
of the triode while the integrated flyback waveform was fed to the pent ode section.
The two outputs at the anodes would then be added, but the shape and the average
value of the resultant waveform would depend upon the degree of coincidence
between the respective inputs. After a filter the average value would become the
oscillator control voltage.
The frequency stability of the flywheel-controlled line time base can have its
disadvantages when a video recorder is to be connected to a TV receiver. Owing to
slight variations in the speed of the tape/head mechanism, there is a tendency for
the recorded synchronising pulses to occur at somewhat irregular intervals. A stable
line oscillator might not be able to respond quickly to those variations; therefore
modifications are necessary to such receivers as may be recommended by the
makers. Some TV sets have built-in alternative flywheel filter components for this
purpose.

10.2 The Line Oscillator

Both the blocking oscillator and a multivibrator are used to provide the line time
base waveform. In addition, most modern receivers designed in Britain tend to
favour a sinewave oscillator, usually of the Hartley type (the exception being found
in I.e. time bases). The reason for this is the improved frequency stability of a
tuned circuit oscillator over one of the other types. That, of course, further en-
hances the flywheel effect already introduced by the discriminator described in
section 10.1.
At first sight it may be puzzling how it is possible to use a sinewave in order to
obtain a sawtooth scanning waveform. What happens in practice is that the sinusoi-
186 Television Principles and Practice

dal shape can be easily converted into a more or less rectangular form, which is, as
we shall see, the required drive for the line output stage (especially in transistorised
circuits).
The frequency of the line oscillator must be controlled automatically by means
of a d.c. voltage derived from the flywheel discriminator. If the oscillator uses a
tuned circuit, then its frequency depends only on the effective value of its capacit-
ance and inductance. Therefore the control voltage must change one of those
properties, which implies that the effective reactance of the circuit has to be altered.
This problem exists elsewhere in a TV receiver-namely in tuner a.f.c. and the
automatic control of the colour sub carrier in the decoder. In both applications
there is normally a varicap diode whose capacitance changes with the applied d.c.
voltage. However, the line oscillator frequency is relatively low (15 625 Hz), and
the amount of capacitance change available directly from a diode would be in-
sufficient. A separate amplifying stage, known as the reactance stage, is incorpora-
ted in the sinewave line oscillator. The reactance stage is connected to the original
tuned circuit in such a way as to behave like a capacitor (or sometimes as an
inductor) whose effective value depends upon the amount of the applied d.c. bias.
Two examples of a sinusoidal line oscillator with its reactance stage are given
in figure 10.3; circuit (a) is based on valves and circuit (b) is transistorised. Consider
the valve 'arrangement first.
The connections to the pentode section are such that in effect it functions as if
it were a double valve. The oscillator circuit is formed by the control grid and the
screen, so that the electron flow to the anode is controlled by the action of the
oscillator; the anode circuit functions in the usual amplifier fashion. Positive feed-
back is applied via C3 to the grid to sustain the oscillations. The tapped, tunable
inductor Ll can be recognised as being the Hartley circuit. Its parallel capacitance
is formed by C1 together with the effective capacitance of triode VI. The latter
effect is produced by the signal taken from the tap on the coil and fed to the
cathode of VI via C2 . The combined phase shift introduced by R 1 , C2 and R2 is
such that an angle of 90° lead is created between the input current to the triode
and the voltage induced in L 1 • Thus VI behaves as if it were a capacitor; the
amount of capacitance depends on the gain of the valve, which, in turn, varies with
the amount of control d.c. applied to the grid.
The fundamental difference between the circuit used in the line time base and a
normal sinusoidal oscillator is the shape of the output waveform. In most applica-
tions it would be expected to generate as pure a sinewave as possible. Now the
output must be in the form of a pulse whose shape has to be adjusted to suit the
line output stage (see anode waveform of V2). The overall shaping is done in two
stages: (1) by a suitable biasing of the combined oscillator and amplifier valve V2
so that it is driven alternately between saturation and cut-off (the resultant clipping
is not necessarily symmetrical); (2) by a suitable shaping network [Rs, Cs in figure
1O.3(a)] connected across the output.
The transistorised version of the line oscillator shown in figure I 0.3(b) appears
to be similar to its valve counterpart. The chief difference, however, lies in the
Line Time Base Orcuits 187
.------...._-----_------h.t.

Vl
d.c. control
voltage from
discriminator

reactance Hartley tuned oscillator and

stage circuit buffer amp.
(a)

d.c. control
voltage from
discriminator

reactance oscillator
stage circuit

Figure 10.3 Examples of sine wave line oscillators: (a) GEe; (b) Thorn. Note
differences between the output waveforms required in a transistorised circuit and
in a valve circuit

actual method of introducing the 90° leading phase shift between the reactance
stage (TRl) and inductor L 1 • The phase shifting network consists of C2 and R 2 ;
the signal developed across R2 is fed to the emitter of TRI through the coupling
capacitor, C1 • The actual amount of 'leading' current supplied by the reactance
188 Television Principles and Practice

stage depends on the d.c. bias applied to the base of TRI.

The sinusoidal waveform developed within the tuned circuit drives TR2 alter-
nately from saturation to cut-off, so that the output at the emitter is almost a
square wave. The differences between the drive waveforms required by a transis-
torised and a valve line output stage, respectively, will be explained later in this
chapter.
In some receivers the reactance stage does have a varicap diode. However, its
effective capacitance must be increased, which is done by means of a transistor
amplifier known as the capacitance multiplier.
The Hartley oscillator circuit can be tuned by means of the ferrite core of the
inductor; its adjustment becomes the line hold preset. This is often in addition to
the line hold potentiometer provided in the line sync. pulse discriminator [VR1 in
figure 10.1(b)]. Neither adjustment is normally available to the viewer, since it is
important that the recommended procedure be observed.
The action of the synchronising pulse discriminator opposes the effect of any
slight lack of accuracy in the oscillator free-running frequency. This may create the
impression that precision in carrying out the adjustments is not important because
of the apparent 'play' in the system. Unfortunately, the automatic line oscillator
control circuit can only correct a limited range of frequency errors. Furthermore,
this range is somewhat wider when the circuit is already synchronised. If the
oscillator is to be 'pulled in' from its free-running state (for example, when the
receiver is first switched on, or after channel changeover) the discriminator can
deal successfully with a much lower frequency error.
The adjustments, as already mentioned in section 10.1, should be performed

output to
driver h.t.
transistor

d., ~'rnl I~II

I,,::
voltage from - - - - - : - - ' ) 1 I
discriminator II
II

c,

Figure 10.4 Line time base blocking oscillator. Conduction is 'blocked' by C 1 ,

which, when fully charged, applies a positive bias to the emitter
Line Time Base arc/~its 189

with the automatic control feature disabled; in some circuits it could be done by
short circuiting the discriminator diodes. The tuning of the coil can be altered
until a 'floating' picture is obtained; this should lock in as soon as the action of the
sync. discriminator has been restored to normal.
Where the time base uses a blocking oscillator, the circuits are very similar to
their basic forms; a valve version was shown in figure 9.9, and its transistorised
counterpart is given in figure 10.4. The high operating frequency gives a smaller size
transformer compared with the design used in the field time base. The blocking of
transistor conduction in the arrangement in figure lOA is brought about by C1 in
the emitter circuit. Positive feedback is applied from the collector to the base by
the action of the transformer (windings L 1 and L2)' When the transistor is driven
into saturation, C1 charges to a positive voltage. Once the collector current has
reached its maximum, the e.m.f. induced in L1 falls to zero and the transistor
switches off. The charge on the capacitor makes the emitter positive with respect
to the base and the cut-off state is maintained until C1 has discharged sufficiently
via R 1 • At this point the d.c. control voltage on the base takes over, the transistor
is again driven into saturation and line flyback begins. The output from the oscilla-
tor is coupled to the next stage via tertiary winding L 3 •
The inductance of the blocking oscillator transformer produces a large back
e.m.f. each time the transistor switches off. In some circuits a diode may be connec-
ted across the primary winding to discharge the magnetic energy and the large
voltage pulse is avoided. The switch-off pulse tends to be followed by self-oscilla-
tions (ringing) of the tuned circuit formed by the inductance of the windings and
their stray capacitance. In the arrangement shown in figure 10.4 ringing is damped
by resistor R 2 •
Control of the blocking oscillator frequency is performed by the synchronising
pulse discriminator. This time the d.c. voltage is fed to the base of the transistor to
override the reverse base-emitter bias and initiate flyback.
Multivibrator circuits are not discussed here, since they are rarely used in modern
receivers, because of their poor frequency stability. If encountered, the circuit will
resemble one of the basic arrangements illustrated in chapter 9.
It is interesting to note that some makers who do not use a sinewave oscillator
in the line time base include a tuned circuit in conjunction with their multivibrator
or blocking oscillator. This is done to improve frequency stability of the oscillator,
making it relatively unaffected by 'stray' trigger pulses.
A number of receivers use an I.e. line time base oscillator. This usually incor-
porates other functions, and the reader is asked to refer to chapter 9, figure 9.14.
Mter the phase comparator and the nIter (pin 7, figure 9.14), the d.c. control volt-
age is fed to the oscillator. Its frequency is governed by the external capacitor, C12
(pins 4 and 5), and the manual 'line hold' adjustment, VR4. The oscillator in an
I.C. is often designed to generate a triangular waveform in the first instance, which
is then shaped within the device to produce a rectangular output. Frequency
stability of the oscillator itself can be obtained by means of internal circuitry, so
that a sinewave circuit is not needed.
190 Television Principles and Practice

10.3 Transistorised Line Driver Circuits

The purpose of the driver stage is to amplify the oscillator output waveform before
feeding it into the line output transistor. This type of arrangement is needed in all
power amplifiers (audio, field and video output), because of their requirement for
a high amplitude input current at a relatively low input voltage.
In the line time base the driver stage may also help to shape the oscillator wave-
form to produce rectangular output pulses.

h.t.

~~: ~I
oscillator C,

Figure 10.5 Line driver amplifier (Thorn)

A typical circuit is shown in figure 10.5. The oscillator waveform (A) is coupled
via Ct to the driver stage. The somewhat modified waveform (B) switches the driver
transistor alternately on and off. The output at the collector is basically a square
wave. Its large amplitude voltage swing must be stepped down by the driver trans-
former, Tl. As shown by the waveform (C), when the transistor switches off, there
is a voltage spike accompanied by 'ringing'. This is due to the initial back e.m.f.
from the transformer winding and the subsequent self-oscillations in the collector
circuit. Excessive ringing on the drive waveform could have a detrimental effect on
the line output transistor; therefore damping is applied by means of C3, R3 and R4
connected effectively across Lt. When the transistor switches on, R4 also limits the
maximum d.c. current taken from the h.t. supply.
Diode Dl across the base-emitter junction of TRI protects the device by pre-
venting a large negative signal being applied to the transistor input.
Line Time Base Orcuits 191

10.4 Equivalent Circuit of a Line Output Stage

The design and operation of the line output stage differs considerably from that of
the field time bas~. The difference is caused by the effect of the long time constant
of the line scan coils and the associated circuitry on the deflection waveforms. It
was shown in section 9.2 how an integrating circuit produces a sawtooth current
waveform in a suitable inductor. The line output transistor acts simply as a switch
which either connects the scan coils across a supply (usually via a transformer) or,
at the end of a line, disconnects them, to initiate flyback. When the voltage is
applied to the coils, the current through them rises gradually; if the circuit time
constant is relatively long, the sawtooth deflecting waveform is obtained.
The behaviour of the line output stage can be better described with reference
to the simplified equivalent circuit shown in figure 10.6; L is the effective induct-
ance and R is the effective resistance of the stage. Parallel capacitor C represents
not only the inevitable stray capacitance, but also additional capacitors which tune
the entire circuit.
At first, with switch Sl (the line output transistor) open, no current flows and
the beam is at the centre of the screen. When the switch closes (the transistor is
turned on by its input drive waveform), the current through the coil rises at a rate
given by the time constant L/R; this causes the beam to be deflected from the
centre to the right-hand side of the screen. At the end of the picture line, Sl opens
(the output transistor is turned off) and flyback begins. The scan current cannot
disappear at this instant because of the energy stored in the inductance; instead the
parallel circuit formed by Land C begins to oscillate (ring). Initially, the sudden
interruption of the main current path produces a very large positive voltage pulse
from the scan coils towards the upper terminal of switch S1. At the same time, the
first half of the beam flyback takes place as the current in the inductor falls to
zero, having charged up the parallel capacitor. The second half of the flyback is
produced when capacitor C discharges into the scan coils. The lower plate of the
capacitor would have previously become positive; therefore the flow of current is
now as shown by the broken line arrow in figure 10.6.
When the beam has reached the left-hand side of the screen, the relatively slower
scan begins. The oscillations must be stopped, because both their timing and wave-
form (sinewave instead of sawtooth) are no longer useful. The energy which would
have maintained the ringing could be dissipated in a suitable damping resistor, but
this would be extremely wasteful, since the process is repeated every 64 fJ.S. Instead
the switch is closed at the end of the flyback, the h. t. supply with its negligible
resistance prevents any further oscillations and the scan current falls at a linear rate.
The beam is, therefore, deflected from the left-hand side to the centre of the screen
and the first half of the scan is completed. During this process the energy stored in
the coil is gradually transferred to the supply capacitor, where it will not be wasted-
it is, in fact, available for the second half of the scan. If the switch remains closed,
the current in the coils builds up in a sawtooth manner and the deflection continues
from the centre to the right-hand side of the screen. Oscillatory flyback then
192 Television Principles and Practice
+ h.t.----_-------,

C
circuit supply
L stray and decoupling
scan additional capacitor
coils capacitance

switch
(line output
transistor or valve)

Figure 10.6 Simplified equivalent circuit of a line output stage. Continuous line
arrow indicates direction of current in scan coils when the beam is deflected be-
tween screen centre and the right-hand side. Broken line arrow shows direction of
current during deflection between centre and the left-hand side of the screen.
When current is zero, the beam is at the centre. As switch opens (flyback), L, C
form a tuned circuit; with the switch closed, R, L form an integrating circuit

follows and the process is repeated as described above.

The whole set-up resembles a kind of pendulum which is slowly moved from its
vertical position to the right (part of scan); it is then released to swing back to the
extreme left with the aid of the potential energy acquired previously (flyback). The
speed of that return movement depends on the length (tuning) of the pendulum.
When the pendulum has reached the topmost left position, a suitable brake must
be applied to prevent further swings and to control the rate of descent (next scan).
Energy has to be added to the system only to lift the pendulum from the vertical
to the top right.
The reader will recall that the video waveform allows approximately 10 IJ.S for
line flyback. If this were not achieved, one would be faced either with width
problems or with picture distortion in the left-hand side of the screen. Therefore
the frequency of flyback ringing is relatively critical; this means that the value of
the tuning capacitor C in figure 10.6 is critical too. This capacitance also governs
the amplitude of the initial voltage pulse developed at the beginning of flyback;
the lower the value of C the greater the back e.m.f. from the coil.

10.5 Simple Transistorised Une Output Stage

The practical line output stage shown in figure 10.7 is now considered. The circuit
is of a relatively simple design, since it is used in a low power, battery/portable type
of receiver.
Line Time Base Circuits 193

h.t.2 -4-~~---_-

T2 ~ .. e.h.t.

~03
.. to a.g.c.
~
~--------_

L6
R,
/~~
h.t.1 -...-----i__-----. ~L7 04 ~C5 h.t.3
02
;;;;;
r - - - - - - - I__- _ _ . - - - - h.t.4

to sync. pulse
discriminator L . -_ _ _ _ _ _ to tube
grid

T1

~II
Figure 10.7 Line output stage ofa battery/portable receiver (Thorn)

It will be noted that, from the scan current point of view, the parallel-connected
deflecting coils LI and L2 are in series with the output transistor TRI via CI. The
receiver power supply (h.t.1) is thus used to 'top up' the energy losses in the scan
circuit in addition to the provision of various h. t. levels to other sections of the
receiver.
The line output transistor has no fIxed d.c. bias, as it is alternately switched on
and off by the base drive waveform from the driver transformer Tl. When the
transistor is turned on, it discharges the scan coil S-co"ection and d.c. blocking
capacitor CI , which results in a current flow through L I and L2 in the 'upward'
direction in the diagram in figure 10.7. At the end of scan the drive waveform
switches TRI off, but the e.m.f. in the coils tries to maintain the original direction
of the current. That means that the parallel combination of C2 and C3 , together
with any stray capacitance, are now charged; the polarity of the flyback pulse is
such that diode D1 is reverse biased. Flyback oscillations begin, and the beam is
returned to the left-hand side of the screen. It is then that the direction of the fly-
back e.m.f. reverses and diode D1 is biased ON. The low resistance of the diode
damps the oscillations immediately and the current gradually decays-flowing in
the 'downward' direction (figure 10.7) in the deflection coils L I , L 2 • The fIrst half
of the scan is thus produced from the energy already stored in the scan coils. The
efficiency of the output stage is therefore high, and for this reason diode D 1 is
called the efficiency diode.
194 Television Principles and Practice

In many receivers, especially in those where the line output stage is fed from a
relatively high supply voltage (100 V or more), the efficiency diode is not used. The
line output transistor itself provides efficiency action by conducting in reverse. This
type of design will be discussed in section 10.6.
The reader can now appreciate why the line drive waveform (see figure 10.5) is
almost square (the mark/space ratio is approximately 1: 1.3). The mark period
s,witches the line output transistor on, while the space interval turns it off. During
the transistor off-state, flyback takes place, followed by the first half of scan, when
the efficiency diode conducts. The exact timings depend upon the practical design
of the circuit, but lack of continuity between the changeover from one state to
the other could cause picture non-linearity towards the centre of the screen.
The parallel capacitors, C2 and C3 in figure 10.7 tune the line scan coils during
flyback thus affecting the width of the picture. Since C2 and C3 are equivalent to
capacitor C in figure 10.6, they also govern the magnitUde of the flyback pulse,
and therefore, protect the output transistor against excessive voltages.
Capacitor C1 in figure 10.7 performs two functions-firstly, blocking the d.c.
from the scan coils, and, secondly, improving the linearity by introducing S-correc-
tion (see section 9.3). The latter occurs because the charging of the capacitor tends
to be slightly non-linear, which, in turn, counteracts the tendency to overscan
towards the edges of the raster.
As far as the scan coils are concerned, the line output transformer, T2 acts as
a choke in the h.t. feed. A current flows through it during scan to 'top up' the
charge on C1 and to supply power to a number of auxiliary h.t. rails; similarly,
flyback pulses are also fed to the transformer. This means that a number of a.c.
outputs are available from the different windings. These outputs are rectified and
smoothed to provide power supplies to other sections of the receiver. The scan
part of the induced waveform is normally used to supply low voltage, high current
demands; conversely, the flyback generates high voltages at relatively low currents.
The latter produces e.h.t. for the final anode of the picture tube, which in a
battery portable type of receiver would be approximately 10 kV.
The advantage of using the line output stage to supply various circuits in the
receiver lies in the fact that the frequency of the resultant a.c. is high (15 625 Hz),
making the ftltering very simplified and giving a smaller transformer size when
compared with a 50 Hz mains unit. In battery portable receivers the main supply
is limited to 12 V d.c., but the line output transformer can supply the necessary
higher voltages. For example, in figure 10.7 h.t.3 of approximately 330 V is fed
to the screen grid and focus control of the tube, h.t.4 of 90 V to the video output
stage, h.t.2 of 24 V to the line and field oscillators as well as to the video driver
stage.
Supply h.t.2 is also known as the boosted h.t. This is obtained by adding to
h. t.l (11 V main receiver supply from the battery) a rectified and smoothed supply
across C4 (derived from L3 and boost diode D2). If the voltage across C4 is 13 V,
then the total with respect to receiver chassis is 11 V + 13 V = 24 V. Of course,
h.t.2 cannot be established until the line oscillator and the line output stage
Line Time Base Qrcuits 195

become operational. Therefore, when first switched on, the h.t.2 rail is at a reduced
voltage of less than 11 V fed from h.t.l via D2. This is sufficient to start the sys-
tem up until normal action is finally reached.
Some designs do not employ a parallel efficiency diode (Dt in figure 10.6);
instead the series boost diode is used to provide the energy recovery action
during the first half of scan. This method is similar to the one adopted in a valve
line output stage, as will be shown in section 10.7.
The line output transformer in figure 10.7 also provides a number of pulses:
winding L6 communicates puls~s to the gated a.g.c. circuit; the tap on L 8 , L9
gives negative pulses to the grid of the picture tube for line flyback blanking; the
flywheel line sync. pulse discriminator is supplied from the lowest point on the
transformer (L4)'

10.6 High Power Transistorised tine Output Stage

In large screen, mains-driven receivers the arrangement of the line output stage is
somewhat more involved. The design is even more complex in a colour TV set.
This is due to a far greater power demand placed upon the line output stage,
which, in turn, is handled by the output transistor. The main design features are:
high level of h. t. voltage (100 V or more) and high amplitude currents (1 A or
more). The available transistors are also subjected to flyback pulses of nearly
2000 V and are close to the limits of their operating capability; they have to be
protected against excessive conditions of drive current and voltage.
The diagram in figure 10.8 is of the line output stage in a colour receiver which
employs 110° deflection. In this type of circuit the power requirements are parti-
cularly high because of the wide angle scanning. The two line output transistors,
TRI and TR2, are connected in series, basically to share the unusually large fly-
back pulse in view of the relatively high h.t. (275 V in this receiver). The flyback
tuning capacitors, C3 and C4 , are of equal value, so that they ensure a balanced
division of voltages across each device. The waveform from the driver stage is
coupled via transformer T1, current limiting resistors R z , R4 and adjustable
inductors Ll and L2 to the bases of the transistors. When TRI and TR2 are driven
hard into saturation, it is essential to their survival that they be bottomed as
quickly as possible. This means that the collector-emitter voltage drops rapidly
to less than one volt before the output current has risen to any extent and power
dissipation in the transistors can then be kept relatively low.
A similar requirement applies when switching off, but this is not so straight-
forward. A heavily saturated power transistor takes a certain amount of time to
turn off after the removal of its base drive. If the input signal were suddenly inter-
rupted, it would make the collector current decay slowly, while the voltage across
the device would rise sharply, and the power rating might be exceeded. The base
inductors, L 1 and L 2 , are used to reduce the base current gradually-when the volt-
age waveform from the driver reverses the coils oppose the sudden change. However,
when the delaying action of the chokes finally stops, the reverse drive applied to
196 Television Principles and Practice
h.t.

various {
h. t. supplies to
convergence
correction

Ie
circuits

Tl
T2

from
driver
~
stage

overload
trip to t~ L9 : to tube
heaters
driver amp. '--------1_

Figure 10.8 Simplified circuit diagram of line output stage of a colour receiver
(Rank)

the bases rapidly turns the transistors off. In the arrangement shown in figure 10.8
the inductors are variable, to ensure balanced operating conditions of each transis-
tor. The adjustment must be carried out according to the makers' instructions, and
usually involves a measurement of an out-of-balance current in the circuit. This
current is often indicated by a p.d. across a fusible resistor (not shown in figure
10.8) connected between the junction of the two output transistors and another
point in the circuit. The adjustment aims at a minimum reading across the resistor
(less than 1 V); incorrect settings can lead to the destruction of the transistors.
In many other designs separate base inductors are not used; their function is
Line Time Base Circuits 197

then taken over by the inductance of the driver transformer itself.

line flyback takes place in the manner already described; however, the reader
will notice that an efficiency diode is not used in the circuit in figure 10.8. The
output transistors are allowed to conduct in reverse, to provide the current during
the first half of scan. This action occurs when the back e.m.f. in the scan coils
reverses at the end of line flyback. The emitter potential of each transistor is now
positive with respect to its collector; consequently, the base is also positive in
relation to the collector (via either Rs or R3 in parallel with the respective second-
ary windings of T1). Efficiency action takes place because the base-collector junc-
tions are forward biased, allowing current to flow. It will also be noted that the
'new' base biasing from the emitter terminals turns the transistors on; it appears
that the original functions of the emitter and the collector are interchanged. This is
possible because the name n-p-n (or p-n-p) implies the same material structure of
the device in either direction.
Efficiency action continues until all the stored magnetic energy has been used
up; the scan current is then at zero and the electron beam is at the centre of the
screen. Before the efficiency period ends, the drive waveform is again inverted by
the action of the oscillator, ready for the 'normal' conduction of TR1 and TR2 to
be resumed for the second half of scan.
The above method is often adopted when the h.t. supply to the line output stage
is in excess of 100 V. The use of a transistor in place of an efficiency diode has one
major disadvantage. When the transistor conducts in the conventional direction, the
voltage across the device is less than 1 V, but during part of the reverse conduction
period this potential difference is in the region of 5 V. The latter value consists of
the base-collector forward bias and the base-emitter 'cut-off drive fed from Tl.
These differences can cause scan non-linearity at the beginning of each picture line.
The effect of the inequality cannot be swamped if a relatively low supply voltage
is used in the line output stage.
The scan coils, L3 and L 4 , are connected to the output transistors via the
S-correction, d.c. blocking capacitors, Cs and C6 • In series with the coils are two
linearity co"ection inductors, Ls and L 6 , with their permanent magnets. It will be
recalled that in the field time base linearity correction took place before the wave-
form was fed to the output amplifier. This is no longer possible in the line output
stage, which merely functions as an ON-OFF switch, and is unable to deal with
anything but a rectangular (square) waveform. The linearity coils apply correction
directly to the line scan current waveform. Ls and L6 vary their inductance as the
amplitude of the current increases; this effect is caused by the permanent magnets
placed close to the coils. Magnetic saturation is produced in the cores of the induc-
tors as the flux from the magnets adds to the flux set up by the rising scan current.
Consequently, the total inductance of the entire circuit falls towards the end of
the line period, which, in turn, alters the shape of the waveform. This effect is
particularly noticeable near the edges of the piCture.,
The core of a linearity coil can be adjusted too. The overall inductance will then
be varied, which affects both the width and the linearity. In many receivers a vari-
198 Television Principles and Practice

able inductor, without a magnet, is used in series with the scan coils to provide
width control.
Additional inductors in the scan circuit can be shocked into self-oscillations by
each flyback pulse. This ringing is damped by parallel resistors-for example, Rs
and R9 in figure 10.8. If the oscillations were allowed to continue, the beam
would tend to move to and fro at the same time as it traversed the screen. This
would lead to a display of alternate dark and light vertical bands, known as stria-
tions, extending from the left towards the centre of the screen.
A colour receiver needs further scan correction by means of convergence wave-
forms, described in the next chapter; such waveforms are also derived from the
line output transformer (L 7 and T2).
Resistor R t in the h.t. feed to the output stage in figure 10.8 limits the current
when the transistors are switched on-especially in the event of a tube flashover.
The 'earthy' end of the stage is connected to the chassis via VR1 and the
associated resistors. This arrangement senses the current in the circuit; if it is exces-
sive a 'turn-off bias disables the driver and the line output transistors remain in the
'off state (assuming that the fault is not within the transistors themselves!).
The capacitors in the base circuits of the line output transistors (Ct and C2 ) are
also a form of protection against tube flashovers-they divert any resultant pulses,
which could otherwise damage the transistors.

10.7 Valve Line Output Stage

The behaviour of the scan coils again, is, similar to that described. in the equivalent
circuit in section 10.4. However, the difference between the design of transistorised
and valve output stages is due to the following factors; valves do not act as low
resistance switches even when fully saturated; valves cannot conduct in reverse, but
are capable of withstanding relatively high voltages-especially those which occur
during flyback.
A Simplified circuit diagram is shown in figure 10.9. The line output valve, VI, is
a high power pentode. The oscillator waveform suitably shaped is applied to the
grid via the coupling capacitor, Ct. The cathode is connected directly to chassis,
and its biasing is derived from the drive waveform together with a correction bias
fed via R3 and its associated circuit (Rt, R 2 , VR1, VDRl, Cs ). The latter stabilises
the performance of the line output stage-that is, the width of the picture, the
magnitude of the boosted h.t. (which will be described later) and, to some extent,
even the e.h.t. This is done by means of feedback from the line output transformer
via Cs . The amplitude of the line pulses at that point depends upon the amplitude
of scan current which represents the picture width. At the same time, the scan
current is dependent upon the pentode biasing and the resultant anode current.
Since e.h.t. is derived from suitably stepped-up line flyback pulses, the feedback
arrangement indirectly senses the magnitude of the e.h.t.
The voltage-dependent resistor, VDRl, rectifies the pulses from Cs . The rectifi-
cation action is due to the fact that the resistance of the device is high during the
Line Time Base Orcuits 199

L1

~==:::!....t~e.h.t. to
tube anode
V3
line drive -----------I
from oscillator -------.I--+-t-!
C1

Figure 10.9 Valve line output stage, including width stabilisation circuit

low amplitude scan waveform and is low during the large amplitude, but opposite
polarity, flyback waveform-exactly as in a rectifier. Its connection to the line out-
put transformer is so arranged that a negative voltage is developed by the action of
the VDR and capacitor Cs . The larger the pulse-suggesting excessive line output
conditions-the more negative is the feedback control voltage. The drive to VI is
thus reduced by this voltage and the output brought back to normal. The feedback
voltage is added to a positive potential obtained from the divider R 1 , VRI which
allows the conduction of VI to be preset as required.
VRI, or its equivalent, may be labelled: 'set boost h. t. " 'set e.h. t.' or 'width',
depending on the manufacturer. Its setting must be carried out according to the
makers' instructions, since excessive drive could result in damage to the valve itself
or to the line output transformer. Alternatively, e.h.t. which is above normal can
also cause serious problems. In most cases the adjustment aims at a specified value
of the boosted h. t.
200 Television Principles and Practice

The line scan coils, L 13 and L 14, are fed from a section of the line output trans-
former between L 8, L9 and L 10 via the S-correction capacitor, C4 •
When the receiver is first switched on, the line output valve is connected to the
h.t. supply via L 3 , L 6 , L2 and diode V2. As the drive waveform turns the pent ode
on harder, its anode current increases. By transformer action a corresponding
current flows through the scan coils, causing the beam to be deflected from the
centre to the right-hand side of the screen. At the end of the line period VI is
rapidly cut off by the oscillator to initiate the flyback. A large back e.m.f. pulse
appears at both the anode of VI and the cathode of V2. For this reason the two
electrodes are brought out to the top of the valve glass envelope in the form of a
top cap (marked T.e. in figure 10.9). This solves the problems of adequate insula-
tion between the adjacent pins of the valve base.
The inductance of the line output stage (the transformer and the scan coils) plus
its stray and flyback tuning capacitance are again shocked into self-oscillations
when VI switches off. Beam flyback takes place in the manner described in section
10.4; during that time diode V2 is reverse biased by the back e.m.f. in the trans-
former. When the polarity of this e.m.f. changes at the end of flyback, V2 switches
on to damp any further oscillations. The scan current gradually decays as C3 charges
via V2 and the first half of the scan is produced. The energy stored in the scan coils
'drives' the windings L 8 , L 10, which, by transformer action, induces a voltage and
current in L4 and L 5 • Since the diode allows energy recovery to take place, it is
called the efficiency diode.
During the 'efficiency' part of the scan C3 is charged by the sum of two voltages:
h.t. supplied to the circuit by V2 and the e.m.f. induced in the transformer Wind-
ings L 4 , Ls. As a result, the potential on this capacitor, with respect to the chassis,
is well in excess of the main receiver h.t. The supply rail fed from C3 is called the
boosted h. t.; the efficiency diode is also known as the boost diode and C3 is referred
to as the boost capacitor. The value of C3 is usually between 0.22 Jl.F and 0.47 Jl.F.
The reader is invited to compare this arrangement with the derivation of the boosted
h.t. in the transistorised circuit shown in figure 10.7.
The boosted h.t. in valve circuits can have any value from 400 V to 1000 V,
depending on receiver requirements. It is used to supply some of the electrodes of
the picture tube (screen grids-or the so-called 'AI' and focus in monochrome
receivers) and the field charging capacitor in valve field time base circuits. The line
output valve is also fed from the boost circuit once the receiver is operating nor-
mally.
In the previous discussion it was implied that the line output valve was com-
pletely switched off during the first half of scan. In practice VI in figure 10.9 is
biased ON before the beam has reached the screen centre. Power losses in a valve
circuit are higher than in the transistorised equivalent; therefore the stored energy
is used up before the first half of scan is completed. The drive waveform is shaped
and timed in such a way as to ensure a smooth takeover from the efficiency diode
to the line output valve. Failure to achieve this can cause picture non-linearity.
E.H.T. is generated by rectifying the large flyback pulse, which is stepped up by
Line Time Base Circuits 201

an additional winding, L 7 , known as the e.h. t. overwinding.

Diode V3 is the e.h. t. rectifier; its heater is supplied, via a current limiting resis-
tor, from a separate winding of just a few turns on the line output transformer
(LI2). It is not practicable to connect the heater to the main valve heater chain
(see chapter 14), because the fllament would then be at a relatively low voltage
with respect to chassis, while the cathode would be maintained at e.h.t. Such an
arrangement would create problems with the insulation between the heater and
the cathode of the rectifier.
The line output transformer, in both valve and transistor circuits, supplies a
number of pulses to other sections of the receiver; in figure 10.9 only a tap feeding
the line sync. pulse discriminator (L9' L 10) is shown. There are, however, other
pulses-namely to the video amplifier (or directly to the tube) to provide line fly-
back blanking, black level clamp (mostly in colour receivers), a.g.c. gating (where
applicable) and the colour decoder (burst gating, burst blanking, PAL switching).

10.8 Generation of E.H.T.

In most receivers e.h.t. is obtained by rectifying a stepped-up line flyback pulse in
the manner illustrated in figures 10.7-10.9. The value of e.h.t. can be from 5 kV
(small screen, battery portable) to 25 kV (large screen colour receivers). The output
from the e.h.t. overwinding can be of the required amplitude--only rectification is
then necessary; or it may be at a lower level, in which case voltage multiplication is
used.

Jl
8.5 kV pulse
input c,
I
D5 C ~ 25.5 kV output
e.h.t. 5 i e.h.t. to final anode
overwinding
7/77;

Figure 10.10 Voltage tripler. This arrangement is used in a colour receiver which
requires a relatively high focus potential. Final capacitor C s is formed by the tube
body
202 Television Principles and Practice

Semiconductor e.h.t. rectifiers are commonly used, because of their negligible

heat dissipation, greater reliability and lack of X-ray radiation. X-rays can be emit-
ted whenever electrons are accelerated through a large p.d. and then suddenly
decelerated by hitting the anode.
Voltage multiplication is only economical in semiconductor circuits, since a
large number of rectifier diodes are needed. The circuit shown in figure 10.10
represents a typical voltage tripler. The energy supplied by pulses from the over-
winding is transferred to the final capacitor, Cs , via the intervening C 1 , C2 , C3 and
C4 through a 'pumping' action of the diodes; each flyback pulse initiates a 'pump-
ing' stroke. The first pulse after switching the receiver on charges C2 via D1 to
8.5 kV. During scan the input from the overwinding falls to zero; D1 is cut off, but
D2 conducts, charging C1 from C2 ; each capacitor ends up with just over 4 kV. The
next pulse recharges C2 , and during the following scan the potential on C1 is raised
again, this time to over 6 kV, and so on, until there is almost 8.5 kV on both capa-
citors. C3 and C4 charge in a similar manner; when C1 and C2 have been charged,
pulses are steered to C3 via D3, and then D4 charges C4 . Let us assume that the
final capacitor, Cs , has not yet been charged, so that it 'sees' the series combina-
tion of C2 and C3 (as well as C1 and C4 ); since there is 8.5 kV on each, Cs can
initially charge to 17 kV via D5. When this voltage is reached, the last pulse in the
whole charging sequence will raise the potential on Cs to 17 kV + 8.5 kV = 25.5 kV.
The whole process takes some time to arrive at its state of balance, because each
time the input to the tripler falls to zero, the existing charges on the capacitors are
being distributed around the circuit.
The tripler is a self-contained unit with the exception of capacitor Cs , which is
normally formed by the body of the picture tube, as will be shown later.
In colour receivers a relatively high voltage is needed for the tube focus electrode
and it can be conveniently tapped from the tripler. The potential across C2 in
figure 10.10 is approximately 8.5 kV, which is then reduced by means of a divider
formed by R 1 , VR1 and R 2 • Occasionally a special VDR with a sliding contact can
be employed. Some designs use an arrangement known as a thick film potentio-
meter. This could be compared to an integrated circuit, since a number of resistors
(for example, R 1 , VR1, R2 in figure 10.10) are deposited in the form of a mm over
a ceramic substrate. These units are obviously not repairable, but they have good
reliability and, unlike large value carbon resistors, their stability is high.
Some receivers have an e.h.t. of less than 20 kV, and a voltage doubler may be
used. In that case the circuit action is similar to that of the tripIer, but the section
comprising C3 , D3, C4 and D4 (figure 10.10) is omitted.
Some designs use a diode split e.h.t. winding instead of a tripler. The winding is
split into a number of sections which are linked by diodes. Each diode rectifies the
flyback pulse from its own section, while the smoothing is performed by the 'stray'
capacitance that exists between adjacent layers of the winding. The final value of
the e.h. t. is equal to the sum of the d.c. levels obtained from each section. Relia-
bility of diode split line output transformer is good, because the high voltage capa-
citors normally used in triplers are not required.
Line Time Base arcuits 203

Modern receivers use IuJrmonic tuning in the generation of e.h.t. The overwind-
ing, by virtue of its design, has a relatively high value of inductance and it also
possesses the inevitable stray capacitance. Therefore each flyback pulse will shock
the circuit into self-oscillations. The 'ringing' not only affects the input to the
e.h.t. rectifier, but is also transformed to other sections of the line output trans-
former-notably to the collector of the output transistor.
At this point the reader must not confuse the flyback tuning with the harmonic
tuning. The latter is primarily used to help to produce the required e.h.t. by taking
advantage of the ringing of the e.h. t. overwind. Flyback tuning is necessary to time
the return of the electron beam at the end of each scan.
Instead of damping the self-oscillations of the e.h.t. overwind, they are en-
couraged, provided the frequency of such oscillations is strictly controlled. The
winding is tuned by its inductance in conjunction with suitable, low value external
capacitors. The capacitors do not have to be connected to the e.h.t. coil itself-for
example, C7 in figure 10.9.
Depending upon the frequency of self-oscillations, either third harmonic or fifth
IuJrmonic tuning could be used. In this context it means that the frequency is linked
('in harmony') with the value of the flyback frequency (which, as we know, is tuned
separately). The mathematics of the associated theory is beyond the scope of this
book, but it can be shown that the desirable frequency of e.h.t. ringing is approxi-
mately three times the flyback frequency; hence third harmonic tuning. The oscilla-
tion waveform adds to the flyback pulse waveform in the overwind, so that the
resultant is an even more 'peaky' output to the e.h.t. rectifier. This means that the
input pulse is increased in magnitude without the necessity for a greater number of
turns in the winding. At the same time, by transformer action, the induced ringing
subtracts from the flyback pulse appearing in the main winding connected to the
output transistor (or valve). Thus, the advantages of third harmonic tuning are:
increased e.h.t. pulse often making a tripler unnecessary; reduced flyback pulse in
the main winding-an important feature when transistors are used; the ringing is
effectively damped at the end of flyback. The explanation of the last feature in-
volves complex mathematics, but from a practical point of view, if the e.h.t. ring-
ing extended into the scan period, it could again produce vertical striations (see
section 10.6).
With the advent of colour receivers fifth harmonic tuning has been widely used.
In this case the frequency of e.h.t. ringing is apprOximately five times the flyback
frequency. The resultant e.h.t. pulse does not increase in its peak amplitude; there-
fore a tripler may be necessary. The principal advantage of fifth harmonic tuning is
that it produces a relatively constant e.h.t., since the output pulses from the over-
wind become broad-almost rectangular in shape. This is particularly important in
colour receivers, where the beam current is relatively high and the e.h.t. must remain
stable. The disadvantage of fifth harmonic tuning is that the flyback pulse at the
collector of the output transistor tends to be more 'peaky'. Careful deSign, however,
ensures that the ratings of the output device are not exceeded.
There is no obvious way of deciding from the circuit diagram which of the two
204 Television Principles and Practice

methods is used; details may only be obtained from the makers' circuit description.
Harmonic tuning is sometimes made adjustable, in which case manufacturers' in-
structions must be consulted.

10.9 TIlyristOrs in Line Output Stage

The chief disadvantage of a transistorised line output stage is the risk of damage to
the device if either the current or the voltage limits are even momentarily exceeded.
Since the transistor only acts as a switch, it might appear that a thyristor could be
used instead. Thyristors have been used in power electrical engineering for many
years, and devices of high voltage and current capacity are easily available. The
principle of operation of a thyristor is described in chapter 14; briefly, it behaves
like a rectifier, since it conducts in one direction only-that is, with its anode posi-
tive with respect to the cathode. However, unlike the rectifier, the conduction does
not normally start until a control signal is applied to the third electrode called the
gate. In that respect it resembles a transistor, which also needs a suitable bias on its
third electrode-the base. The difference between the thyristor and the transistor
lies in the fact that when the base drive is removed, the transistor switches off;
when the thyristor gate signal is removed, the anode current continues until it is
reduced to a very low value (practically zero) by the external circuit. On a.c., or
unsmoothed d.c., no problem exists, since the supply waveform naturally reduces to
zero. In d.c. circuits with a smoothed output this is not the case, and other means
of turning the thyristor off have to be adopted.
A simplified arrangement of a line output stage with thyristors is shown in
figure 10.11. Each thyristor (THI and TH2) is connected in parallel with a diode
(Dl and D2, respectively). This is necessary to allow for both the forward and the
reverse flow of current in the scan coils in the manner described in section 10.4. A
line output transistor can conduct in either direction if required; thyristors can pass
current only one way.
The line output thyristor, TH2, is in parallel with the efficiency diode, D2-this
arrangemen t resembles the conventional transistorised stage shown in figure 10.7.
The output pair is connected to a suitable tap on the line output transformer,
which, in turn, feeds the scan coils and proVides other services in the usual manner.
The other thyristor-diode pair (THl, 01) is used to switch off the output devices
and initiates the flyback. It is easier to explain the circuit action if one examines it
at the instant when the pulse from the line oscillator switches the flyback thyristor,
THI on (it actually occurs a couple of microseconds before the flyback is due).
This action connects the tuned circuit formed by Ll and C1 to chassis. C1 was
originally charged up to h.t. when TH2 was conducting; now it discharges via L 1,
THI and the output pair-D2 and TH2. Effectively, this presents a reverse bias across
the output thyristor which switches off, but the discharge of C1 continues via D2. The
change in the flow of current is communicated, via the output transformer, to the
line scan coils and the flyback commences. The duration of the flyback depends on
the tuning of L 1, C 1 , which, together with the inductance of the scan coils, form a
Line Time Base Orcuits 205

ringing circuit. During the second half of the flyback oscillation C1 discharges
through the line output transformer, Dl and L 1 • This, in effect, places a reverse
bias across THI and the flyback thyristor also switches off.
Flyback ringing stops when the beam has reached the left-hand side of the
screen. At that point the e.m.f. in the scan coils reverses and the efficiency diode,
D2, is biased ON. The diode shorts out the tuned circuit L I, C 1 , the scan current
decays gradually and the first half of the scan is produced.
The second half of the line scan takes place when thyristor TH2 switches on.
However, it is not the line oscillator waveform which is applied to the line output
thyristor. Instead a suitable pulse is developed by the input transformer, Tl, when
its current is interrupted at the time of THI switching off. This pulse is then

line
L, C,
output
transformer
and
scan coils
TH1 D1 TH2
from line
oscillator

~
flyback com mutating scan
circuit circuit circuit

Figure 10.11 Principle of thyristor line output stage. The division of functions
between the devices is due both to their inability to conduct in reverse and to the
need for a thyristor 'turn-off pulse. Commutation is the changeover of conduction
from one thyristor circuit to the other

delayed by C2 , R 1 and L2 to allow the efficiency diode to conduct before the out-
put thyristor turns on.
The changeover of conduction from THI to TH2 and back again is called com-
mutation. Since this action is produced by C1 (during flyback) and T1 (during
scan), the tuned circuit, C" L" is called a commutatingcircuit, and T1 is some-
times referred to as the commutating transformer.
A practical thyristor line output circuit might include a number of damping
206 Television Principles and Practice

components connected across the thyristors. This is necessary to prevent either

unwanted ringing or voltage 'spikes', which could turn one of the switching devices
on and upset the timing of the entire sequence.

10.10 Faults in Line Time Base Circuits

Broadly speaking, the main fault symptoms due to a defect in the line time base
may be grouped as follows: incorrect line frequency, insufficient width, horizontal
non·linearity, blank screen.
In fault· finding it is important to establish that the fault is likely to be within
the given stage, since identical effects could be produced by defects elsewhere in the
receiver. For example, lack of line synchronisation could also be due to a weak
signal, sync. separator faults, etc. A blank screen might be traced to a problem in
the power supplies, the video amplifier, and so on.
Whenever valves are used, they are often the first suspect components; however,
they are also easily replaced and consequently eliminated from the enquiry. It will
be noted that ordinary valve testers are not always useful, because the high ampli-
tude pulses present in the circuit cannot be simulated by the tester.
Faulty transistors are easier to discover, as they usually develop either definite
short or open circuit symptoms. On the other hand, it can be both time-consuming
and often undesirable to remove them from the circuit unless it is reasonably
justified.
line output transistors can be accidentally damaged by careless fault-rmding.
Shorting of transistor electrodes with test probes can either apply excessive drive
or remove it at the wrong time interval-the resultant surge could be destructive.
Many manufacturers suggest reducing the h.t. to the line output stage during fault-
finding (especially in mains-operated receivers). This can be done by connecting a
resistor of a suitable value and power rating in series with the h.t. feed or by reduc-
ing the output voltage from the relevant power supply regulator. The same tech-
nique can be adopted in the initial setting-up after the replacement of the output
transistors. The servicing engineer must consult the makers' manual in this respect.
An oscilloscope is extremely useful in fault-rmding in time base circuits, as one
deals most of the time with waveforms which can be easily displayed. However, a
multimeter can also indicate the presence or the absence of a drive. This is due to
the fact that most pulse waveforms are not symmetrical about the horizontal axis
and they have a net d.c. value which the voltmeter indicates. For example, the
drive waveform maintains the output valve in a cut-off state for a considerable
amount of time; hence, the d.c. readings tend to show as a high negative grid bias.
Similarly, in transistor circuits a very low, or perhaps a reverse, base bias can be
indicated in the presence of a pulse waveform.
Test equipment should not be connected directly to the anode of the line out-
put valve or the cathode of the efficiency diode. Flyback pulse voltages of several
kV can damage the instruments! Similarly, it is usually not recommended to mea-
sure the output voltage from the line output transistor. A low capacitance attenua-
Line Time Base Circuits 207

tor probe should be used when connecting an oscilloscope to the base of a line
output transistor.
E.H.T. should be measured by means of a suitable meter with a probe connected
to the final anode of the tube. In all-valve receivers it used to be common practice
to draw a spark with a screwdriver from any desired part of the line output circuit.
Where semiconductors are used, whether in the line output stage or elsewhere, this
method could easily damage them. E.H.T. can often be felt as static electricity in
front of the tube screen; alternatively, a neon tester, slowly brought towards the
e.h. t. lead, might be used as an indicator.
The reader is asked to refer to the notes on I.e. fault-finding which were given
in some detail in section 7.11.
The major types of line time base faults will now be discussed, together with
some general suggestions regarding possible suspect areas.

Incorrect line frequency

This symptom can be classed as lack of perfect line synchronisation. Suspect areas
are: the sync. pulse discriminator or the line oscillator itself. The discriminator can
normally be disabled and the effect of oscillator adjustments noted. The discrimina-
tor relies on the symmetry in component values (capacitors, resistors and forward/
reverse resistances of the diodes), and it also needs both waveforms (synchronising
pulses and feedback) to develop the control signal.

Insufficient width and picture non-linearity

Lack of width can be accompanied by non-linearity of the picture. In general terms,
defects on the left-hand side of the screen are due to the efficiency diode circuit
and problems on the right-hand side are caused by the line output device (valve or
transistor). In a valve line output stage inadequate drive from the oscillator can
reduce the width and so can the faults in the width/e.h.t. stabilising circuit. Exces-
sive loading of the line output transformer by an external circuit, defective wind-
ings of the transformer or flyback tuning capacitors can also be responsible.
Non-linearity towards the centre of the picture can be caused by excessive drive
to the line output valve (faults in width stabilising circuit). If the picture becomes
trapezium-shaped, it is due to a fault in one section of the line scan coils or their
associated components.

Blank screen (no raster)

This fault suggests that there is no e.h.t. (apart from problems in other sections of
the receiver, as mentioned earlier), since in most receivers the fmal anode supply is
linked with horizontal scanning circuits. It is necessary to establish whether the
fault is due to the e.h. t. section itself (overwind, rectifiers, etc.) or the remainder of
the line time base. Some people may hear the high pitched (15 625 Hz) line whistle
208 Television Principles and Practice

which is caused by the vibrations set up in the line output transformer; this could
quickly confirm that the oscillator and most of the output stage are functioning.
Lack of drive to the output stage can be due to the failure of the oscillator or
the coupling components. Where valves are used in the output, a loss of drive pre-
sents serious problems. The absence of the negative-going grid waveform leaves the
valve with zero bias and a very high current is drawn. As a result, the line output
and/or the efficiency diode can be destroyed, or, worse still, the transformer itself
could be damaged. During the investigation of this type of fault the screen grid
supply to the output pentode can be disconnected to cut the valve off.
A short circuit across any of the line output transformer terminals willload the
output stage to such an extent that flyback oscillations can not take place; the
e.h. t. will then be very low or non-existent. Since the transformer supplies a number
of circuits, each one has to be disconnected in turn until the fault has been cleared.
Both the boost capacitor and various tuning capacitors are a common source of
short circuits, because they are subjected to relatively high pulse voltages. Failure of
a number of protective components, in both the line output stage and the driver,
can result in a breakdown of the line output transistor.
The line outp;.!t transformer is the last item on the 'suspect' list. The cost of a
replacement and the work involved must be fully justified by reasonable evidence
of the fault being juside the transformer.

Other faults
(1) 'Ballooning' (picture 'breathing'). The picture expands as the brightness control
is advanced. This is due to the e.h.t. falling rapidly as the demand for beam current
increases, and can be frequently caused by faulty e.h.t. rectifiers or the tripler capa-
citors, as applicable. Similarly, reduced h.t. or boost h.t. can also cause 'ballooning'.
TripIer breakdown may be accompanied by its overheating, flashes on the screen
and unpleasant smell!
(2) Corona discharge and arcing. Corona is in the form of violet 'glow', which can
surround any component or lead carrying very high voltages; arcing means that
flashover takes place to other parts of the circuit (normally to receiver earth).
Usually corona is a warning sign of imminent arcing. High voltage discharge can be
caused by dust or moisture on the suspect components, sharp points on soldered
joints, defective insulation, etc. Arcing can damage semicoruluctor devices and pro-
duce interference on the picture.
(3) Vertical striations. Alternate narrow dark and bright vertical bars especially
noticeable on a blank, darkened raster. These are caused by ringing in the width,
in the linearity or even in the scan coils, owing to faulty damping components.

10.11 Safety Aspects of Line Time Base Operation and Servicing

The line output stage, including the e.h.t. section, is a high power and a high volt-
age circuit. Under fault conditions it can carry large currents which, in turn, could
Line Time Base Circuits 209

cause dangerous overheating. A number of components are designated as 'safety

components'; any replacements and the methods of mounting them in the receiver
must conform to the makers' specifications. Failure to do so could cause the repair
to be deemed dangerous in terms of British legislation. Some of the more obvious
components in this category are: the line output transformer, the boost capacitor,
the screen grid feed resistor, etc.
Valves in the line output stage, including the tube, can emit X-rays; the risk
increases with the operating voltage (usually in excess of 5 kY). The danger is that
such radiation is not immediately noticeable, but its effect accumulates and can be
a serious health risk.
Valves which are subjected to high voltages are mounted within a metal screen,
which should always be in position unless it is absolutely necessary to remove it
for access. The screen not only absorbs the X-ray radiation, but also helps to pre-
vent the spread of r.f. interference and acts as a mechanical barrier to the sections
which operate at high voltages.
11 Picture Tubes and
Associated Circuits

11.1 Basic Principles of Picture Tube Operation

In principle, both monochrome and colour tubes have much in common. Funda-
mentally, they are a form of thermionic valve in which the electrons are made to
strike a special material, known as phosphor, which, in turn, emits light. The
amount of light produced depends upon the number of electrons hitting the
screen material at any time. The flow of electrons gives rise to the tube or beam
cu"ent, which is governed by the magnitude of voltages applied to the various
electrodl's. In a TV tube some of the voltages are kept constant, others can be
controlled QY the viewer (or the service engineer) and yet others are amplified
versions of the actual received signals which produce the picture.
The picture itself is formed by variations in the amount of light emitted from
the screen as the electron beam scans its entire area. The scanning is under the
control of the field and the line time bases. The light output persists for a fraction
of a second after the beam has left any given position on the screen. This persis-
tence, together with the chosen scanning rate, ensures that the eye retains a com-
plete image without excessive flicker.
The colour of the emitted light depends upon the chemical composition of the
phosphor. In monochrome tubes maximum beam current gives rise to white light
output (in practice, it might be 'cool white'-see the discussion in chapter 1). The
screen of a colour tube is coated with three types of phosphors deposited in a
carefully arranged pattern. Each phosphor is capable of emitting light of different
colour-namely red, green or blue. From a normal viewing distance the light out-
puts appear to mix and create an impression of any desired hue, as described in
chapter 1. The entire process is under the control of the decoded and matrixed
colour signals applied to the tube electrodes-which are usually the cathodes.
The electrons have to travel over a considerable distance between the cathode
and the screen, which means that they need a lot of energy. They must also have
enough energy left when they reach the screen to allow some of it to be converted
into light of adequate intensity. For this reason the operating voltages are high;
the voltages increase in value with the size of the screen and, for reasons given
later, they tend to be higher in colour receivers than in monochrome receivers.
A thin layer of aluminium is deposited on the inner surface of the tube screen.
210
Picture Tubes and Associated Circuits 211

This allows the electrons to pass through, while acting as a barrier to the heavier
and destructive ions. The coating also acts as a 'mirror' which reflects the emitted
light forwards and improves the picture brightness.

11.2 The Electron Gun

The part of the tube which is responsible for the emission and acceleration of the
electrons is called the electron gun. As shown in figure Il.1(a), the electron gun
includes the heater, the cathode, the control grid, the first anode (sometimes
referred to as the screen grid or the 'AI '), the second anode, the focus

linearity sleeve

(a) (b)

Figure 11.1 Principle of construction of monochrome tube: (a) arrangement of

electrodes in the electron gun; (b) linearity sleeve is used in some receivers; the
sleeve is pushed under the scan coils over the tube neck

anode and the fourth anode (commonly termed the final anode). The electrodes
are in the form of hollow cylinders to allow the beam to pass through. The electri-
cal supplies are so arranged that the electrode voltage levels increase progressively
with respect to the cathode. There is one exception-the control grid is at a lower
potential with respect to the cathode, the value of that voltage depends upon the
setting of the brightness control and tube design. Typical electrode potentials of
a monochrome tube are: heater, 6.3 V; cathode, 80 V; control grid, 20 V; AI,
300 V; focus, 350 V; final anode, 17 kV. The corresponding voltages of a colour
tube are: heater, 6.3 V; cathode, 150 V; control grid, 40 V; AI, 500 V (average);
focus, 4 kV; final anode, 25 kV.
Complex electrode assemblies are sometimes used, especially in colour tubes,
their aim being primarily to improve the focusing of the electron beam over the
entire area of the screen. There are involved reasons, beyond the scope of this
212 Television Principles and Practice

discussion, why the electrons forming the beam tend to 'spread out', not only
during their straight passage to the screen, but also during the deflecting process.
The degree of such distortion depends on the amount of deflection from the
centre; therefore, in the setting-up procedure optimum focus has to be aimed at.
Because of the photographic similarities in the problems encountered, the electron
gun is sometimes called the electron lens. Early designs had magnetic rather than
electrostatic focusing, using permanent magnets or electromagnets mounted on
the outside of the tube neck.
Most colour tubes have three complete guns housed side by side in the neck of
the tube. Each gun is associated with a particular 'colour' beam, which, in turn,
must always land on the appropriate colour phosphor. If all three do not come
close enough on the screen, the impression of colour mixing is lost, and that leads
to a colour fringing effect. For this reason there is an additional arrangement both
on the inside and on the outside of the neck of the tube for the purpose of converg-
ing the three beams.
If one or more of the three colour beams land on the 'wrong' colour phosphor,
then incorrect colours are produced in patches over certain parts of the screen, as
though some paint had been spilled on the picture. To counteract this effect, known
as lack of colour purity, adjustable magnets are placed on the outside of the tube
neck. Both the purity and the convergence adjustments will be considered separately.
The scan coils are positioned over the neck of the tube and against the flared part;
this coincides with the space immediately following the electron gun and any correc-
tion magnet assemblies.
The electrical connections to the electrodes of the gun, except that to the final
anode, are brought out to the base of the tube. The final anode operates at such a
high voltage that it could create serious insulation problems both outside and inside
the tube. Therefore the e.h.t. supply is made via a special connector on the flared
part of the tube and brought back to the final anode in the gun assembly by means
of a conducting coating of graphite ('Aquadag'). This arrangement has additional
advantages: firstly, together with a similar layer on the outside of the tube, it forms
a reliable e.h.t. capacitor which is used for smoothing the final anode supply (the
capacitance is in the order of 2000 pF); secondly, it prevents internal light reflec-
tions between the screen and the glass; finally, the coating completes the electrical
circuit from the screen of the tube to the anode.

11.3 Arrangements for Correct Picture Geometry

Correct picture geometry implies a rectangular picture with reasonably straight
edges, of a correct width and height, properly centred within the available screen
area and, finally, of a linear, undistorted shape.
The linearity, as well as the height and the width, is governed by the shape and
amplitude of each scanning waveform. This, in turn, is controlled by the time base
circuits and components external to the tube. In some receivers the linearity of the
horizontal scan can also be controlled by means of the so-called linearity sleeve. The
Picture Tubes and Associated Circuits 213

sleeve consists of two thin copper pressings made into loops [see figure 11.1(b)]
and secured onto a tube of an insulating material. The whole assembly is placed on
the tube neck and is partly underneath the scan coils. The position of the linearity
sleeve with respect to the scan coils can be readjusted if absolutely necessary. The
loops work on the principle that the current in the scan coils induces currents in
the linearity sleeve. Consequently, the electron beam is subjected to two magnetic
fields-one produced by the current in the copper loops and another set up by the
scan coils themselves. The resultant distribution of the deflecting magnetic field is
such that a more linear raster can be achieved.
Pincushion distortion is another form of deformation of the picture. As the
name suggests, the edges of the raster, instead of being straight, tend to become
bowed inwards; the picture seems to be overscanned towards the corners (see
figure 11.10). This is particularly pronounced in wide angle, 'flat' screen tubes. The
maximum deflecting force of both the field and line scan coils-coinciding with the
extremities of the raster-causes the overscan. In monochrome receivers pincushion
distortion can be corrected by means of permanent magnets mounted on brackets
close to the scan coils. The magnets can be adjusted in any desired direction until
a satisfactory, straight raster is displayed. The use of freely adjustable magnets close
to colour tubes (at either the front or the back) is not allowed, since it would alter
the relative paths of the three beams and result in lack of colour purity on the
screen. For this reason pincushion distortion in colour receivers becomes part of
scan waveform correction. In 110° tube circuits the arrangement is particularly
complicated, as will be shown .in a later section of this chapter.
The position of the scan coils upon the tube neck can be altered in the majority
of TV tubes. The assembly must be pushed forward so that the raster fIlls the
screen without 'corner cutting'. Where appropriate, the coils can be rotated to ensure
that the picture is not sloping. Final centering of the display in monochrome
receivers is by means of adjustable ring magnets on the neck of the tube and posi-
tioned before the deflecting coils. Again, this method cannot be applied to colour
tubes, so that picture shift takes place by injecting a suitable d.c. into the respective
scan coils. To allow for an up-or-down and left-or-right movement, some form of
polarity reversal of such a current is included.
In many colour receivers the adjustment to the position of the scan coils is also
used to set up colour purity, as explained in chapter 13. There are some colour
tubes which use precision scanning to reduce convergence problems; the scan coils
are then permanently fixed to the tube, no adjustment is available and the tube
must be replaced with its own deflecting coils.

11.4 The Shadowmask Tube-Delta Gun Version

The screen of a colour tube has three phosphor materials-one fer each primary
colour, red, green and blue. There are two principal methods of distributing the
phosphors-either in the form of dots or as vertical stripes. The dot arrangement is
an early but still very popular design. The dots are deposited in a triangular forma-
214 Television Principles and Practice

tion; each group of the three primary colours is known as the triad. Spacings be-
tween the triads are so small that from a normal viewing distance the dot structure
is not evident.
The three electron guns in the neck of the tube are also arranged in a correspond-
ing triangular fashion known as the delta gun formation. The electrons emitted
from each gun are then aimed at their respective colour dots. To ensure the correct
landing positions for each beam, a perforated steel plate, called the shadowmask, is
mounted inside the tube and just in front of the screen. This acts as a masking plate
-only the beam of one particular colour may strike its own phosphor dots; the
other two beams cannot reach the 'wrong' phosphor, because the dot lies in the
'shadow' cast by the mask. This happens because each beam approaches the holes
in the mask from a different position in the gun assembly. The diagram in figure
11.2 illustrates the principle of the shadowmask. The three beams converge upon
the holes in the plate before they reach their appropriate phosphor dots.

---
blue gun
[Link]--- __
-B
_ __ Qgreen

red gun CJ-- - - - - -

---
c:r--- --
green gun
--='-J\ \.J blue

primary colour
phosphor dots

Figure 11.2 Principle of shadowmask. The three colour beams converge upon
the holes in the mask, which ensures that the beams strike their own colour phos-
phor dots

The shadowmask is an obstruction in the path of the beam, and the majority of
the emitted electrons never reach the screen. Those which do must have sufficient
energy to compensate for the missing contribution from the remainder in order to
produce adequate light output. As a result, the total beam current of a colour tube
could reach 1.5 rnA and the anode voltage is on average 25 kV. Both figures are
well above their counterparts for a monochrome tube of the same screen size.
Those electrons which are captured by the mask generate heat in the plate,
causing it to expand. (This power loss is in the region of 15-20 W.) This must be
taken care of in the construction of the tube; otherwise the 'shadow' effect would
alter, giving rise to serious purity errors.
Picture Tubes and Associated Orcuits 215

The shadowmask, being made of steel, can become accidentally magnetised

from domestic electrical appliances, toys, tools, power cables, etc., if these are
placed close to the :TV receiver. If it happens, the plate will deflect the electrons
in a random fashion, producing irregular patches of unwanted colour over the
picture. This can, however, be corrected by either automatic or manual demagnetisa-
tion (degaussing).
The chief disadvantage of the delta gun arrangement is the difficulty in achieving
perfect convergence of the three beams over the whole area of the screen. Conver-
gence implies that the beams land on the phosphor dots of one triad at a time. If
only two beams struck the dots of one triad, while the third beam reached the
phosphor of its correct colour but in a different triad, the three light outputs would
not merge and colour fringing would take place. There is an obvious tendency for
the three beams to diverge as they become deflected by the scan coils. This is caused
by the unequal distances covered by each beam in order to reach their common
destination. For example, if the arrangement of the guns is as shown in figure 11.3,

field dynamic convergence

'--....I~~ coils

'---'~___ line dynamic convergence

coils
internal pole
+--~.-magnetic field
pieces

internal magnetic
movement of beam (radial)
screening

neck of tube

Figure 11.3 Delta gun tube-magnet assembly which provides radial convergence
correction. The three beams are separately pre deflected by their convergence
magnets before entering the scan coils. The direction of the magnetic field is
between the internal pole pieces, which results in a radial movement of the beam.
A separate lateral convergence assembly (not shown here) is used for the blue
beam
216 Television Principles and Practice

the 'blue' beam needs less vertical deflection to reach the top of the screen than
either the 'red' or the 'green'. On the other hand, the 'red' beam requires less hori-
zontal deflection to arrive at the extreme left edge of the picture than either of the
other two. The amount of misconvergence increases with the degree of deflection
from the centre of the screen. The process of correction which maintains the con-
vergence of the beams during their line and field deflection is known as the dynamic
convergence. However, the beams must be converged first at the centre of the
screen before any deflection takes place; this is known as the static convergence.
The principle of convergence is based on predeflecting the beams by means of
special permanent magnets and electromagnets ahead of the magnetic field of the
scan coils. The combined action of the scan and the convergence fields produces a
converged raster. Each 'colour' requires a different amount of correction to ensure
convergence at the top or bottom, and at the left- or right-hand side of the screen.
Usually the adjustment of static convergence is by means of permanent magnets
(although electromagnets operated from a d.c. source are also used), and dynamic
convergence is obtained from electromagnets whose coils are fed with suitably
shaped waveforms. An appropriate magnet assembly is placed on the neck of the
tube and the resultant magnetic fields are guided into the path of each beam by
means of internal pole pieces, as shown in figure 11.3. Since the deflection caused
by each magnet is perpendicular to the lines of magnetic field, it follows that the
resultant beam movement is along the radius of the neck-this is sometimes referred
to as radial convergence. A closer study of the arrangement in figure 11.3 reveals
that the red and the green beams can be moved diagonally, which in effect gives an
up-or-down as well as left-or-right shift. On the other hand, the blue beam can move
either up or down but not sideways. For this reason a separate magnet assembly is
also placed on the tube neck; this time the magnetic field is guided into the neck of
the tube at such an angle as to move the blue beam from side to side. In practice,
the arrangement deflects the blue beam in one direction while shifting the green and
red in the opposite direction until the three colours converge. The control magnet is
known as the blue lateral (as opposed to the blue radial previously discussed).

11.5 The Shadowmask Tube-In-line Version

The convergence process in a delta gun tube is relatively involved. This problem has
been overcome by the introduction of the in-line gun assembly, in which all three
guns are in line along the horizontal diameter of the neck. It can be seen from the
diagram in figure 11.4 that the amount of vertical deflection is the same for all
three beams. Horizontal deflection, however, is symmetrical only for the centre
colour, and it is unequal for the other two beams. In some tubes the green gun
occupies the centre position, in others the red beam is centrally placed. The in-line
arrangement requires less convergence correction, which leads to simpler circuits
and fewer adjustments; it also reduces the overall power losses in both the line and
the field time base circuits.
Picture Tubes and Associated Circuits 217

in-line

9U~Sblue
f green

<?;;;-£
/ --~-~
red _

Figure 11.4 Principle of in-line tube and its mask. The guns are in line across
the tube neck, the shadowmask is slotted and the phosphors form stripes on the
screen

The simplicity of an in-line arrangement has been applied to the precision-in-line

(PIL) tube. All dynamic convergence adjustments have been dispensed with,
although static convergence may be corrected if absolutely necessary. This apparent
simplification is achieved by very close tolerances in the manufacture of the tube,
by the position of the gun and, in particular, by the use of a highly specialised design
of scan coils. The latter produce a magnetic field whose distribution provides the
correct beam deflection. Naturally, the scan coils then become part of the tube to
which they are permanently bonded.
Because of the in-line positioning of the guns, the phosphors on the screen are in
a similar configuration. Instead of dots there are now vertical stripes; the shadow-
mask has vertical slots which effectively make it more 'transparent' to the electron
current, unlike the previous type of mask described in section 11.4. Thus, for
similar values of e.h.t. and beam current, the light output is increased, giving a
brighter picture. Some makers of either type of tube give the phosphor stripes, or
the dots, a black surround, which improves the contrast between the illuminated
and the dark areas of the screen, creating an impression of added 'brightness'.

11.6 The Trinitron Tube

The Trinitron was introduced by the Sony Co. before the in-line shadowmask
version appeared on the market. The similarity between the two types of picture
tube lies in the fact that they both have vertical phosphor stripes. Instead of the
rather more solid shadowmask there is now an aperture grill; the three cathodes of
the electron gun are in line, but beyond the cathodes there is only one set of elec-
trodes common to all three beams. These electrodes are known as the electron lens,
since their effect on the three beams is similar to the action of a complex optical
lens arrangement. The three beams are brought to a common point before they
218 Television Principles and Practice

enter the area of the deflecting coils. In effect, there is now only one apparent
point of origin of the beams, which makes for good focus and easy convergence.
Convergence in the smaller screen Trinitron tube is achieved by applying suitably
shaped voltage waveforms to two pairs of plates which are inside the neck of the
tube. External convergence magnet assemblies are used in the later versions of the
tube.

11.7 Dynamic Convergence Circuits-General Requirements

The requirements of a delta gun tube will now be considered, but these can be
easily adapted to the needs of an in-line tube, where the problems are similar
(although considerably simplified).
Since the guns are situated away from the tube axis, the amount of both the
horizontal and the vertical deflection must be somewhat different for each colour
beam. As the scan cOils set up symmetrical magnetic fields with respect to the axis
of the tube, the convergence correction fields must be made non-symmetrical to
provide the required unequal deflection.
The convergence electromagnets shown in figure 11.3 are supplied with wave-
forms known as parabola and tilted parabola; examples of these are shown in
figure 11.5. The convergence waveforms are derived from the two time bases, be-
cause the correction process must be synchronised with the action of the main scan.
Initially, the time base waveforms are available in the form of either rectangular

,
(a)
I
----t----

I
Ib) ~-----! ~-:-'"
--~----~

I I

(d)
I
I
tilted parabola

(c) I,

Figure 11.5 Basic waveforms used in convergence correction circuits

Picture Tubes and Associated Orcuits 219

pulses or sawtooth. The change from these into a parabola and a tilted parabola
takes place in the convergence control circuits. Various forms of pulse integrating
circuits are used following the principle of integration described in chapter 9. The
reader will recall how a rectangular pulse is changed into sawtooth by an integrating
circuit; if, however, a sawtooth waveform is fed to an integrating circuit it becomes
parabolic in shape [figure 11.S(a)] . A tilted parabola [figure 11.5(c)] is produced
by adding a sawtooth waveform [figure II.S(b)] to a parabola. The available con-
trols vary either the amplitude of the parabola (these are often labelled 'amp'
controls) or the amount of tilt sawtooth to be added (consequently, the latter are
referred to as 'tilt' controls).
The most satisfactory method of setting up the receiver convergence is in con-
junction with a test pattern display of crosshatch. It consists of squares formed by
vertical and horizontal white lines. Any misconvergence shows up as colour fringing
or, more aptly, as colour splitting in a particular area of the screen where the im-
pression of white lines is lost. The correction circuitry is arranged and labelled
according to the defects which could be noticed in the crosshatch display on the
screen. Invariably a suitable, if somewhat exaggerated, diagram of the misconver-
gence accompanies each control to facilitate their identification. We shall return to
the adjustment procedure in chapter 12.
In the delta gun tube the red and the green beams are controlled together for
ease of operation. However, one set of controls-red/green (matrixed}-shifts both
beams diagonally in the same direction-that is, either towards or away from the
axis of the tube neck (see diagram in figure 11.3). The other set of controls-red/
green differential-shifts the red beam diagonally in one direction, while the green
is moved the opposite way. The reason for such an arrangement will be described
in chapter 12. The need for the separate blue lateral and blue radial adjustment was
described in section 11.6.

11.8 Dynamic Field Convergence Circuits-Delta Gun Tubes

The diagram in figure 11.6 shows a dynamic field convergence circuit which is
associated with a transistorised field output stage (similar to the one described in
section 9.6). The return path from the field scan coils is via C 1 and R 1 ; therefore
the sawtooth scan current develops two waveforms: a parabola across C1 (the
large value electrolytic capacitor integrates the input waveform) and the sawtooth
across R l ' The three sets of field convergence coils L JiL2' L3 /L4 and Ls /L6 are
fed with those waveforms via their respective adjustments.
The input to the BLUE coils consists of the parabola from VRl, to which a degree
of sawtooth has been added via VR2 and Rs. The return from Ls and L6 is connec-
ted to the junction of two equal value resistors R2 and R3 ; this establishes a centre
reference point, so that the blue beam can be moved either up or down.
The other two sets of coils, RED and GREEN, are effectively in series; the para-
bola is fed at one end (VR3), the sawtooth at the other end (VR4). The red/green
differential control, VRS, acts as a variable shunt; when adjusted towards its top
220 Television Principles and Practice

RIG
SYMMETRY

feedback
linearity RIG
correction ~!::::::==~DIFF

Figure 11.6 Field dynamic convergence circuit-delta gun tube. Field scan
current develops a parabola across C1 and a sawtooth waveform across R 1 • Both
waveforms are fed in appropriate proportions to the convergence coils (Thorn)

end position, the current through the RED coils is reduced and that through the
GREEN coils is increased, which provides for an opposing diagonal movement of
the two beams.
The various adjustments to the shape of the convergence waveforms tends to
alter their effective d.c. levels, which, in turn, could upset static convergence. Diode
DI, together with the coupling capacitor C2 acts as a d.c. restoring circuit; the para-
bolic waveform is now clamped at a fixed level, resulting in less need for constant
readjustments to the static convergence. A transistor with its base and emitter
strapped to form a diode may sometimes be used in place of D I.
In valve receivers the required parabola can be derived from across the cathode
bypass capacitor of the field output valve, while the sawtooth is provided by a
winding on the field output transformer.
The field scan coils are in two sections which are connected in parallel and
placed on both sides of the tube neck. If there is a lack of symmetry between the
two sections, convergence errors are likely to occur. For this reason a balancing
control-the low value variable resistor, VR6-links the coils as shown in figure
11.6.
Picture Tubes and Associated Orcuits 221

11.9 Dynamic Line Convergence Circuits-Delta Gun Tubes

The principle of operation of the line convergence circuitry is similar to that of its
field counterpart. The initial waveform in the shape of either rectangular or saw·
tooth voltage pulses can be obtained from a separate winding on the line output
transformer, or even from across the line scan coils. A tilted parabola is needed
again; therefore the necessary waveform shaping is provided by the correction
circuit. Use can be made of inductance as an integrating component, because the
required time constants are now relatively short compared with those of the field
scan. If a rectangular voltage pulse is applied to an integrating inductor, the
resultant current has sawtooth waveform; when a sawtooth voltage is applied to
such a coil, the resultant current waveform becomes parabolic. The convergence
coils themselves are often used as integrating components in addition to external
variable inductors.
Tuned circuits can sometimes be employed for convergence correction. The
sinew ave they produce resembles a parabola, which is further shaped by diodes
and resistor-capacitor networks. Since the line frequency is much higher than that
of the field, the component values tend to be more critical and their precise func·
tions difficult to predict without makers' information.
The reader has to appreciate that is is the current waveform through the con-
vergence coils which does the actual correction. Such waveforms are not always
easily obtained by means of an oscilloscope, as the instrument is usually connected
to display voltage waveforms. Current waveforms may be viewed if a special
current probe is connected between the oscilloscope and the leads to the conver-
gence coils.
Figure 11.7 shows an arrangement for a line convergence circuit in a delta gun
tube. Effectively the various convergence coils are in series with the line scan coils,
L 1 and L 2 • The scan current from winding L 15 on the line output transformer
flows through the parallel connected scan coils; balancing between the two sections
is by means of the inductor L3 (see the field counterpart VR6 in figure 11.6). The
current returns via the linearity coil, L 4 , S-correction capacitor, C1 , and another
section of the output transformer-L 16 and L 17 • This section feeds the convergence
correction circuits and their respective coils.
The blue lateral convergence coil, L6 , is supplied with a current whose amplitude
is varied by the blue width adjustment, L 5 (lateral movement of the blue beam
effectively increases the width of the blue raster). Since the correction may be
needed to provide movement from left to right or vice versa, the connections-B 1
and B2 -to the coils can be reversed. In some receivers dynamic blue lateral action
is not necessary and the coils may be disconnected. After all, the blue gun is in the
central position as far as the line scan is concerned, which should not lead to line
convergence errors.
The red, green and blue radial convergence coil circuits are effectively in series
with one another. The adjustments are across the coils to modify the voltage wave-
forms applied to them; this in turn gives rise to the tilted parabola current waveform
222 Television Principles and Practice

to pincushion transductor

L, LIN!::
BALANCE
line

part
of line
output
transformer

BLUE AMP

I BLUE
LsI WIDTH

RIG
DIFF

D1

Figure 11.7 Line dynamic convergence circuit-delta gun tube. The entire
assembly is fed from an auxiliary winding on line output transformer. The
adjustments provide the required waveform shaping. The sense of the Blue Lateral
convergence coil is reversible (at terminals BI-B2). A pincushion transductor is
illustrated in figure 11.11. (Decca)

in coils Ls to L 13 • The blue radial adjustments in figure 11.7 include a tuned circuit
L 7 , C3 ; here it is tuned to twice line frequency and the control is labelled 'blue
shape'.
The red and the green convergence currents are controlled together in a manner
similar to that used with their field counterparts. The red/green differential control,
VR4, produces an opposing flow of currents in the two sets of coils. The two
Picture Tubes and Associated Orcuits 223

matrixed controls are VR3 (red/green amplitude) andL 14 (red/green tilt); these,
as previously explained, affect the two beams equally. Diode Dl in series with
resistor R6 is connected across the red/green coils to clamp the [Link] of the
correction waveform. Consequently, the centre (static) convergence is relatively
unaffected by the dynamic adjustments; in addition, the clamping effect also helps
to make the various dynamic controls less interdependent.
The line and field convergence arrangements described so far are known as
passive circuits-all the energy supplied to them originates from the line and the
field output stages, respectively. These circuits are normally associated with 90 0
tubes; convergence errors in large screen, 110 0 delta gun tubes could be consider-
able, and the correction field must be made stronger. Instead of placing an addi-
tional burden on the time base circuits, separate amplifiers are used to provide the
driving power for the convergence coils. Such an arrangement is called an active
convergence circuit. The initial waveform at a low power level is derived from the
line and field scan circuits and it is then fed to suitable transistor amplifiers. Apart
from the amplifying stages, however, the principles of waveform shaping, etc., are
similar to those used in passive circuits.

11.10 Dynamic Convergence Circuits-In-line Tubes

The in-line positioning of the guns simplifies the convergence requirements-even in
1100 tubes. One of the guns, often the green gun, occupies the central position
along the tube axis; consequently, only two beams need convergence correction with
respect to the centre colour. The necessary circuits are very much simpler and the
total number of adjustments is reduced.
The diagram in figure 11.8 shows the line dynamic convergence circuits associa-
ted with a 1100 , in-line tube. There are two separate convergence cOils-L4 for the

part of

"00
output '~
: L,
transformer I

C;t

Figure 11.8 Line convergence circuit-in-line gun tube (Rank)

224 Television Principles and Practice

red andL s for the blue beam. Pulses from windingL 1 on the line output trans-
former are changed into sinewaves by the tuned circuits C2 ,L 2 and C3 ,L3 ; the
coils are the equivalent of the parabola amplitude controls. The shape of the initial
sinew ave is then altered by the presence of the diode clamp circuitsR 1 , DI and
R 2 , D2 and the parallel networks of VR1, C4 and VR2, C6 ; the variable resistors
are now the equivalents to tilt controls. The waveforms which are applied to the
convergence coils, again, are in the form of a tilted parabola. This requirement is
similar to that for a delta gun tube, since both the red and the blue beams are still
off-centre as far as the line deflection is concerned.
Some manufacturers have Simplified the circuitry further by connecting the red
and the blue line convergence coils in series. Only one amplitude and one tilt
adjustment are needed, because the resultant magnetic fields are arranged in such a
way that one beam moves to the left while the other one shifts to the right and
they converge.
The field dynamic convergence circuit of the same receiver is shown in figure
11.9. First thoughts suggest that an in-line tube should not require any field con-
vergence correction. The three beams are in the same horizontal plane and they
should be deflected in the vertical direction without parting company. Indeed, it is
possible to find receivers which dispense completely with the field convergence. Any

01

from field
driver amp.

RED

BLUE

04 Rs
RED/BLUE OVERLAY

Figure 11.9 Field convergence circuit-in-line gun tube. To reduce the loading
on the field output stage, the convergence correction assembly is supplied from a
separate convergence waveform amplifier. (Rank)
Picture Tubes and Associated Orcuits 225

residual errors can then be corrected by slight repositioning of the convergence coil
assembly or even by readjusting the purity magnets! (This should only be undertaken
where specific instructions are given in the makers' manual.)
The circuit illustrated in figure 11.9 is of interest, since it uses a technique also
found in delta gun receivers. The sawtooth waveform developed in the field time
base is taken from the output driver stage to be amplified separately by the field
convergence amplifier (shown as a block). The a.c. coupled waveform is then split
into both positive and negative half-cycles by the two pairs of diodes D 1, D2 and
D3, D4. Each half of the waveform is responsible for the field deflection between
the centre and either the top or the bottom of the screen. The arrangement makes
the convergence adjustments easier because oflack of interaction between the
controls. The centre-tapped potential dividers,R 1 ,R 2 and R3 ,R4 , form zero
reference points, so that the beam may be moved either upwards or downwards as
required.
The red/blue overlay convergence coils are connected in series, but their magnetic
fields act in opposition to one another to converge the beams at the top and bottom
of the screen. Practical effects of these adjustments are discussed in the next chapter.

11.11 Pincushion Correction

General requirements
Pincushion distortion gives bowed edges to what should be a rectangular picture. It
is due to the effect of scanning a flat, wide angle screen, which tends to produce
'stretching' at the corners of the raster. There is a tendency for the general expan-
sion of the raster in any case, as described in the section on S-correction. Despite
S-correction, there is additional stretching due to the combined effect of the line
and the field deflection which reaches its maximum towards the four corners of
the screen. This is represented diagrammatically, if somewhat exaggerated, in
figure 11.10; outline (a) illustrates the natural tendency towards pincushion distor-
tion. To obtain a rectangular raster (b) the corners would have to be 'pulled-in' and
the top and bottom centres slightly expanded. This is achieved by superimposing
on the normal deflecting field a correction field which, by itself, would produce a
barrel shaped raster (c). The combined effect of (a) and (c) should then be close to
the ideal (b).
As already mentioned, in monochrome receivers the correction field comes
from small magnets mounted on flexible brackets near the tube flare. These are
carefully adjusted to produce straight outlines to the picture.
'Free' magnets are not permissible near colour tubes, because of the danger of
magnetising the mask, or any other steel parts associated with the screen, leading
to a complete loss of colour purity. Therefore pincushion correction is obtained
electronically by modifying the scan waveforms, which in turn alters the deflecting
fields. Correction is needed along the vertical axis of the screen, North-South, to be
226 Television Principles and Practice
N
N

W--t-----+------f--E W-+----~--~~E

lC
(b) ideal raster (c)
(a) distorted raster

screen bottom
line
poe'od , ~

.....----field p e r i o d - - - - - - ...,,,,-:--------i~~1

screen
centre

modulated field scan

screen
(d)
top

screen centre
screen top screen bottom

individual line
1
modulating envelope
scan waveform from field time base
(e)

Figure 11.10 Pincushion distortion and correction waveforms: (a) distortion

shows up as stretching of the raster towards screen corners and is accompanied by
inward bowing of the verticals and the horizontals; (b) ideal, rectangular raster;
(c) correction applied to both time bases aims to produce magnetic field which,
if acting alone, would set up outward bowing raster; (d) field sawtooth waveform
modulated by line parabolic waveform (N-S correction); (e) line sawtooth wave-
form modulated by field parabola (E-W correction)
Picture Tubes and Associated Orcuits 227

applied to the field time base, and along the horizontal, East-West, fed to the line
time base.
The outlines (a) and (c) in figure 11.10 indicate that the amount of correction
along N-S and E-W depends upon the position of the beam at any particular time.
For example, the field time base must increase its deflection as the line scan pro-
gresses towards the centre of the screen, and must reduce it near the edges. Con-
versely, the line deflection is at its maximum as the field scan nears the centre, but
it is reduced in the top and bottom parts of the picture. It is, therefore, necessary
to 'inform' one time base of the action of the other. This information is exchanged
by modulating the field deflection current with a parabolic waveform at line
frequency (N-S correction), and, conversely, by modulating the line scan current
waveform with a parabola at field frequency (E-W correction). Such modulated
waveforms are shown in figure 11.1O(d) (for clarity only a few lines per field,
instead of 312t, are indicated).
In figure 11.10(d) the basic field sawtooth (broken line) is corrected at line
frequency by a small parabolic waveform. The direction of the line parabola is
reversed after the centre of the screen, because the bowing in the picture occurs in
the opposite sense in the two halves of the scan. The amount of correction is nil
at the centre of the screen, and it increases progressively towards the top and
bottom. The individual parabolas show that maximum correction takes place at
the centre of each picture line in order to produce the barrel shaped deflection
from figure 11.10(c).
The E-W correction modulates the basic line scan sawtooth waveform at field
rate. As a result of this, the amplitude of the current in the line coils follows a
parabolic outline which is indicated by the broken line in figure 11.1 O(e). At the
centre of the screen, along the E-W axis, the line deflection is a maximum, pro-
gressively reaching its minimum in the top and bottom parts of the picture, as
suggested in figure 11.l0(c).
There are two basic methods of pincushion correction used in colour TV -by
means of a special device known as a transductor, or by employing separate modu-
lating circuits using diodes and transistor amplifiers. Some manufacturers may even
combine both methods in one receiver.

Passive correction circuits

The principle of construction of a pincushion correction transductor is shown in
figure 11.11(a). At first glance it looks like a transformer, but the method of its
operation also implies a variable inductor, thus giving rise to the name transductor.
Its purpose is to modulate the two scan waveforms in the manner described above.
The winding on the centre limb is known as the control winding and it is connected
usually in series with the field scan coils. The two outermost windings are called
the load windings and are then in parallel with the line scan coils. These connections
apply when the transductor is used as the only method of correction, which is
normally found in 90° tube circuits. As the amplitude of the field scan current
228 Television Principles and Practice

to line
scan coils

to field scan coils

via pinchushion controls
(a) (c)

from field _ _ _--_-'"'~f\....._--~........

scan coils

to line
scan coils

return to
field output ___- 4 - -......- -......~--~...!......J
stage
(b)

Figure 11.11 Pincushion correction by means of a transductor: (a) principle of

construction of a transductor; (b) typical circuit diagram; ( c) alternative circuit
symbol for a transductor

through L 1 increases, the magnetic flux in the core rises towards saturation. Con-
sequently, the inductance of L2 is reduced during one half of the field scan and
that of L3 during the second half of the scan. (If the core is already magnetically
saturated, it cannot accept any further increase in its flux and the effective induct-
ance of a coil wound on this core is reduced. This principle was also exploited in
the line linearity coil described in section 10.6.) Since the two load windings are
connected across the line scan coils, they divert some of the line deflecting current
away from the coils, which, in turn, reduces the line scan itself. Maximum current
is diverted when the inductance of either L2 or L3 'is a minimum-corresponding to
the extremities of the field scan. This process generates the waveform shown in
figure 11.1 O(e), which satisfies the requirements of E-W correction.
The N-S correction takes place as a result of the field scan current in L 1 being
modulated by the line waveform from either L2 or L3 . The two load windings are
wound on their respective sections of the core in such a way that when saturation
occurs in one coil, the other one functions normally. Modulation is achieved by
transformer action between the non-saturated winding and Ll . In effect, this
induces e.m.f.s of a suitable polarity and waveform which modify the field scan
current as shown in figure 1 [Link](e).
The diagram in figure 11.11 (b) shows how such a transductor can be incorpora-
Picture Tubes and Associated Circuits 229

ted in the correction circuit. The control winding, L 1 , consists of two parallel
sections, L 1a and LIb' The available pincushion correction adjustments are: phase
control (L 4 ), which governs the position along the raster where control takes place
(along the N-S axis); and amplitude control (VR1), which governs the amount of
correction actually applied.

Active co"ection circuits

Colour receivers which feature 110 0 tubes use active pincushion correction circuits,
sometimes including a transductor as well. Owing to the relative complexity of the
whole arrangement, partial block diagrams are used to illustrate the general principles;
these are shown in figures 11.12 and 11.13.
In figure 11.12 the N-S correction is by means of the trans duct or connected in
series with the field scan coils in the manner described above. The modulating line
pulses are fed in from the line scan coils, which, again, are in parallel with the trans-
ductor windings.

T2

N-S correction

field output field output transductor

driver stage amplifier

field scan
coils

keystone raster correction

and width and modulator ~-----'
amplifier drive

E-W correction

Figure 11.12 Active pincushion correction arrangement-circuit A. This type of

circuit is used in conjunction with a 1100 tube where large errors have to be
corrected. N-S correction is by means of a transductor, but E-W correction uses
a diode modulator (D 1, D2). A number of additional features are provided by the
arrangement (see text)
230 Television Principles and Practice
N-S
amplitude
V R1 ' - - - - - - - - ,

from line
output transformer

to line
scan coils

Figure 11.13 Active pincushion correction arrangement-circuit B. E-W correction

circuit is similar to that in figure 11.12. N-S correction no longer uses a transduc-
tor; instead a separate modulator amplifier provides the modulation of field scan
waveform by line pulses

In the E-W correction circuit the modulating field waveform is obtained from
the voltages developed by the return field scan current across C4 and R 1. The wave-
form across C4 is a parabola (the capacitor integrates!), while a sawtooth is pro-
duced across R 1 . The parabola is fed to a shaping circuit and then amplified before
being used to modulate the line scan. This, in turn, takes place via the modulating
transformer Tl: the correction current is driven into winding L3 and, because of
transformer action, it modifies the line deflection current that flows in L2 . The path
of the driving current from L3 is completed via D2 to chassis. This diode, together
with Dl, forms the diode modulator, which is connected to windingL4 on the line
Picture Tubes and Associated Qrcuits 231

output transformer (T2). Both diodes conduct during the line scan period, thus
clamping the top terminal of L3 to earth. At the same time, D1 and D2 rectify the
output from L4 , which is smoothed by C" to generate one of the receiver l.t. sup-
plies (l.t.1). In some designs one of the diodes operates as the efficiency diode.
The variations in the current induced in L3 (and L 1, for that matter) by the
modulator drive are seen by the entire line output stage as changes in the effective
inductance of the scan circuit. As this could interfere with flyback tuning and the
generation of the e.h.t. (see chapter 10), an additional capacitor-C3 -is brought
into the circuit during line flyback. Its value is carefully chosen so that in conjunc-
tion with LIthe flyback tuning remains unchanged. (Dl and D2 are reverse biased
by the retrace pulse, leaving only C3 in circuit.)
The circuit also includes keystone distortion correction. If uncorrected, the
picture could be somewhat trapezium shaped, as the sides of the raster are not
parallel. The improvement is achieved by the addition of variable sawtooth from
across R 1 (via VR2) to the field parabola.
Keystone and pincushion correction are forms of dynamic width adjustment,
and it is logical to introduce overall width control to the picture through the
modulator circuit (control VR3).
The circuit in figure 11.13 presents an alternative arrangement. E-W pincushion
correction is almost identical in principle with that in the previous example. T1 is
the E-W modulating transformer, T2 represents part of the line output transformer,
and for ease of comparison the component numbering in the diode modulator
circuit is the same as the equivalent in figure 11.12.
The N-S correction circuit in figure 11.13 does not use a transductor. Opposite
polarity line pulses are fed to a bridge circuit given by D3, D4 and VR2 (balance
control). When the bridge is balanced, there is no output from the junction of D3
and D4. However, a variable amount of field sawtooth is introduced to the slider of
VR2, which upsets the symmetry of the bridge. The resultant output is then shaped,
amplified and, fmally, fed to the N-S modulating transformer, T3. Since the second-
ary winding of T3 is in series with the field scan coils, the line frequency correction
waveform is impressed upon the field deflecting current.
In some receivers the N-S modulation waveform is fed to one of the field time
base amplifiers as though it were part of linearity control arrangement.

11.12 Grey Scale Tracking

Grey scale tracking applies to the correct adjustment of colour tube voltages.
These settings ensure that the operating conditions of the three guns result in a
black and white (plus grey!) display when no colour signal is applied. The indivi-
dual guns, although housed in the same glass envelope, are likely to have different
characteristics-namely equal beam currents could be produced only by unequal
voltages on the electrodes of each gun. At the same time, the sensitivity of phos-
phors to the beam current depends on the chemical composition of the material.
For practical reasons some of the voltages are common to all three beams, so
232 Television Principles and Practice

that any balancing between the guns is achieved by trimming the potentials at the
three cathodes and at the first anodes (AI). If the voltage levels are wrong, then
the guns are not driven together into cut-off, giving some colour tinting in what
should be a black, or dark grey, area of the picture. Similarly, if the peak beam
currents are incorrectly balanced, a coloured display will appear instead of white.
Correctly adjusted guns will track over the entire range of picture brightness from
white to black. Methods of adjustment are described in the next chapter, since
they are part of colour receiver setting-up procedure.
The circuit diagrams of video output stages given in chapter 6 showed the
necessary provisions for varying the drives to the tube cathodes. The supply to the
Al electrodes can be derived from the boost voltage in a valve line output stage, or
from a suitable winding on the line output transformer. The supply must be recti-
fied, smoothed and adequately decoupled to ensure freedom from often unexplained
colour tinting. The adjustments are based on preset potentiometers in the Al supply
lines; in a few receivers there is also a tint control (in addition to saturation control),
which can bias the beams towards a particular colour to suit the viewers' taste.

11.13 Automatic Degaussing

One of the disadvantages of the steel shadowmask is that it can become accidentally
magnetised, and this would cause serious purity errors. The irregular patches of
unwanted colour which would then appear on the screen cannot be corrected by
ordinary purity adjustments. In severe cases the entire picture is covered with un-
related colour patterns. Magnetisation can affect not only the mask, but also other
steel parts associated with the tube-for example, the metal shield surrounding the
cone of the tube, the rimband which reinforces the junction of the screen with the
flared portion, etc. This' effect can be produced by some domestic electrical
appliances brought too close to the receiver, magnetised tools, electric toys or
current carrying cables, and even by the changes in relation to the earth's magnetic
field. The latter can occur if the receiver is moved from its original position in the
room.
Severe cases require manual demagnetisation, or degaussing by the service
engineer-a method will be described in the next chapter. In most instances the
amount of accidental magnetisation is slight, and each receiver has a circuit which
automatically applies a degaussing field around the tube flare and the sides when-
ever the set is switched on from cold.
Two circuit diagrams are shown in figure 11.14. The degaussing coils-L 1 in
circuit (a) or L2 in circuit (b)-are placed around the flared portion of the tube to
include the front edge of the screen and the magnetic shield which surrounds the
cone of the tube. In circuit (a), when the receiver is first switched on, thermistor
Rl is cold and has a low resistance. The voltage across the VDR is then high and its
resistance is also low. A large a .c. current (about 1.5 A) flows through the coil;
since this current raises both the temperature and the resistance of the thermistor,
the voltage applied to the VDR reduces. The combined effect of the increasing
 Picture Tubes and Associated Circuits 233

p.t.c.

receiver
a.c.
mains
supply

--/0>---___ --.1

(a) (b)

Figure 11.14 Automatic degaussing circuits

resistance of R 1 and that of the VDR reduces the amplitude of the a.c. current; the
alternating magnetic field set up by the coils gradually decays to demagnetise the
tube. The resistance of the VDR is finally very high, and any remaining current
through Rl is diverted into R 2 . In this circuit R2 is necessary to maintain the
temperature of the thermistor and its high resistance.
In circuit ll.14(b) the VDR has been replaced by another thermistor, R4 ;
actually R3 and R4 form a single unit, as one resistor maintains the temperature of
the other. The principle of operation is similar to that of the previous arrangement.
Initially, there is a high demagnetising current due to a low resistance of both R3
and R4 ; as they warm up, however, the current reduces until it is finally diverted
into Rs to maintain the temperature of R 3 . Since R4 is physically close to R 3 , its
temperature and resistance also remain high. The entire process of automatic
degaussing is like the erasing of magnetic tapes, which are similarly subjected to
decaying magnetic fields.
The disadvantage of the circuits described is that if the coils were open-circuited,
a large current could flow through the shunt resistors (R2 or R s ), creating a risk of
overheating. Therefore a dual thermistor can be used to dispense with the parallel
resistor-an open circuit in the coils would simply result in no current!

11.14 Tube Safety Features; Precautions when Handling C.R.T.

The tube operates under vacuum, and because of its large surface area it is subjected
to aJarge compression force. The glass must be strong enough to withstand this in
normal circumstances. Many tubes have a steel band which reinforces the edge of
the screen and the cone. This is to ensure that in the event of a fracture the result-
ant implosion will be contained, to prevent glass being scattered in all directions.
Very high voltages are applied to the tube electrodes-the focus and the final
anode, in particular-and they can lead to internal flashovers between the various
electrodes. Such a discharge, although momentary, is accompanied by a flow of
peak current of several hundred amperes. The flashover potentials could be fed to
234 Television Principles and Practice

the amplifiers supplying the particular electrodes and destroy some of the com-
ponents. To prevent this happening, the electrodes are connected to spark gaps
which usually form part of the tube base panel. The width of the gap is critical
and its condition is important for correct operation and safety. The gaps are re-
turned to the main earth of the receiver via a separate connection to prevent the
discharge current flowing through any other section of the chassis and inducing
destructive overvoltages.
The danger of flashovers is reduced if the ell.t. is maintained at its correct
level, which often means that the h. t. supply to the line output stage is as re-
commended by the makers.
The level of e.h.t. applied to the tube is such that it can be responsible for the
production of X-rays. The latest colour tubes are said to be free from dangerous
radiation for up to 30 kV. It would be inadvisable, however, to operate a receiver
while working close to the tube for unduly long periods of time. The effect of
X-rays is cumulative and is not immediately obvious!
The tube Aquadag coatings_form a very efficient capacitor, which in dry room
conditions can store its high voltage charge for a considerable length of time. This
is not sufficient to cause any direct danger, but it might give an electrical shock,
leading to an involuntary reflex which could cause an accident. For that reason the
e.h.t. must be discharged from the tube connector to chassis before one works on
the tube itself.
The tube rimband might also acquire a high electrostatic charge, which could
interfere with the main electron beam or give shocks if the edge of the screen was
touched. It is therefore connected to chassis via a parallel combination of a high
value resistor and a capacitor. A direct connection to the chassis in mains-operated
receivers is unacceptable, since it could bring a dangerous mains voltage to the
screen. The components must be suitably rated and considered to be a safety
feature replaceable only with the makers' recommended type.
The magnetic shield around the tube, where fitted, and the external Aquadag
coating must be connected to the receiver chassis. A spring-loaded contact is used
to afford the necessary connection-its condition must be good, to ensure freedom
from interference or even flashovers.
When handling the tube, great care must be taken to prevent damage. No undue
strain on the glass is allowed when either removing or refitting it in the receiver.
The tube must not be carried by its neck alone, which could fracture under its
own weight. The face of the tube could be easily scratched-if it is necessary to
rest the screen on any surface, it must be placed on a suitable protective mat.
12 Receiver Setting-up
Procedure

12.1 Fundamentals of Receiver Setting-up Procedure

The purpose of setting up a receiver is to ensure that it is capable of displaying a
picture according to its design specifications and to the high standards of broad-
casting. The TV receiver must also operate within the margins of safety prescribed
by the manufacturers. The need for setting-up arises either as part of the installa-
tion checks on a new receiver or as the adjustments after certain repairs or as a
result of receiver ageing. The procedure is more involved-and more critical-in
colour receivers than in transistorised monochrome sets. In turn, those receivers
which use a delta gun tube require a greater amount of effort to produce an accept-
able picture, since their convergence is more complex than that of the in-line, or
precision-in-line, tube.
One can consider both colour and monochrome receivers together, since it is
initially essential to establish a good monochrome picture in either case.
Because of the much higher voltage and power levels found in colour receivers,
the presets associated with power supplies (for example, h.t. and/or e.h.t., protec-
tion features, etc.) must be. adjusted strictly in accordance with the makers' instruc-
tions. The colour tube itself presents different requirements from its monochrome
counterpart in order to display a good black and white picture.
The list of checks in section 12.2 summarises typical setting-up procedure for a
colour receiver, but this can be easily adapted to suit the requirements of a mono-
chrome receiver. The actual order of operations must be confirmed with the service
manual, since their sequence is influenced by any interaction between the adjust-
ments. In particular, any presets which affect the operation of the time bases
(linearity, amplitude, e.h.t., etc.) must be checked before the purity and convergence
adjustments are carried out because the correction waveforms are derived from the
main scan waveforms.

12.2 Installation Checks on a Colour Receiver

Some of the following checks are best carried out in the workshop, while others
have to be done, or repeated, in the viewer's home.

235
236 Television Principles and Practice

(1) Check for best positioning of the set in the room. This is possible with tactful
consultation with the customer, ideally no strong direct light (natural or artificial)
should fall on the screen; the receiver should be away from large ferrous objects
or electrical equipment, as they might introduce magnetic fields which affect the
shadowmask. The position must be reasonably permanent, since moving the
receiver about can cause a change in the effect of the earth magnetic field. At the
same time, the set must not be close to a source of heat or dampness.
(2) Check mains connection-ensure correct mains voltage tapping and the polarity
of the live/neutral leads where applicable.
(3) Switch on the receiver and assess the overall picture quality-checking for the
freedom from noise, ghosting, etc., which could be caused by an inadequate aerial
installation, incorrect tuning, excessive or insufficient signal strength; ensure precise
line and field synchronisation, etc. Allow at least 10 min for the receiver to warm
up before any circuit adjustments are to be carried out. A static display, such as a
test card, is best suited for many checks.
(4) Check the receiver h.t. level and adjust if a suitable preset is provided in the
power supply regulator; in some receivers it is the e.h.t. which must be measured.
(5) Check the brightness and/or contrast control-ideally with the aid of a test
card; ensure that peak white and all shades of grey are clearly reproduced (ignoring
at this point any colour casts). Often the two controls have associated presets in the
beam limiter, black level clamp, video amplifier drive, a.g.c. and others which need
careful setting strictly to the makers' instructions.
(6) Check the height, width, linearity and centering of the picture; the correct
dimensions of the required display are given by the test card (see section 12.9).
In some receivers no separate width adjustment is available; in these cases
the correct e.h.t. level should automatically give an adequate line scan amplitUde.
On the other hand, if the width and the horizontal linearity had to be altered by
their respective controls, it might be necessary to recheck the e.h.t.
(7) Check picture focus on bright details near the central part of the screen. Exact
adjustment may have to be performed at the end of the entire setting-up procedure
in order to arrive at an optimum which maintains good focus over the full range of
brightness control and gives minimum moire effect. The latter is in the form of
irregular dark line pattern (somewhat reminiscent of a 'fmgerprint' pattern) parti-
cularly noticeable on a blank raster.
(8) Check picture purity-if necessary, degauss the receiver manually and then
adjust the purity; the more typical procedures will be described separately in
section 12.3 and 12.4.
(9) Check static convergence and readjust the purity if the convergence had to be
altered-see section 12.5.
(10) Check dynamic convergence-see section 12.6.
(11) Check for pincushion distortion-see section 12.7.
(12) Recheck picture height and width. The convergence adjustments tend to alter
the time base loading and affect the scan waveforms; iflarge convergence errors have
to be corrected, it may be necessary to repeat the sequence from that point. In a
Receiver Setting-up Procedure 237

really badly rnisadjusted receiver the procedure might be repeated three times; on
each occasion the errors are reduced and fmally optimum settings are obtained.
(13) Check grey scale tracking-see section 12.8.
(14) Check saturation (colour) control for correct flesh tones, and trim the bright-
ness and contrast controls if required.

12.3 Manual Degaussing

The auto-degaussing circuit of a colour receiver (section 11.13) can remove all but the
most severe magnetisation of the shadowmask and the steel parts associated with the
tube. If random purity errors persist on the screen, it may be necessary to de magnet-
ise the tube using a portable coil of a suitable design.
The coil is connected to the a.c. mains and, with its push button ON-OFF switch
depressed, it is moved very slowly over the top, bottom and sides of the receiver at
the front of the cabinet. Next, in a circular motion, the coil is moved over the face
of the tube and then slowly brought away from the screen as if following a spiral
path until the coil is about 2.5 m (8 ft) from the receiver. The whole procedure
should take approximately one minute; at the end, tum the degaussing coil face
downwards and release the ON-OFF switch.
The following precautions must be observed to ensure complete and safe
demagnetisation:
(1) The coil must not be brought close to the back of the cabinet, as it would
de magnetise the various permanent magnets mounted on the neck of the tube.
(2) The coil must be moved away from the receiver to simulate a decaying magnetic
field; the speCified distance ensures that the field reaching the receiver at the
moment of switch-off is practically nil; as an added precaution the coil is first
turned face downwards. In a workshop it is essential to check that the coil is at
least 2.5 m from any colour receiver prior to switching off; otherwise, the field
from the coil could induce magnetisation.
(3) The coil has to pass a relatively heavy current; therefore it must not be switched
on for longer than about two minutes at a time, to prevent it overheating.
NOTE: The receiver may be left switched either on or off during the operation.

12.4 Purity Adjustments

Bad colour purity produces irregular shaped random patches of colour which are
most noticeable on a blank, white raster and they persist despite prior degaussing.
Purity adjustments ensure that the given colour beam can only strike its own
phosphor dots, or stripes, over the entire screen area. This is achieved by the
shadowmask, provided that the angle of approach between each beam and the
mask is correct. The approach angle depends upon the position of the scan coils
along the neck of the tube and on the paths of the beams in relation to the tube
axis before they even reach the deflecting coils. The scan coils of all colour tubes
238 Television Principles and Practice

except the precision-in-line type (PIL) can be moved along the neck axis. The paths
of the electron beams can be modified by means of ring magnets placed on the neck
of the tube.
The more typical procedure is now summarised:

(1) The receiver displays an unmodulated (no picture) raster.

(2) Switch off the blue and the green guns of the tube, leaving only a plain red
raster-a single colour display helps to observe even slight purity errors. (Most
receivers have suitable gun switches which usually disconnect the supplies to the
first anodes; there are also signal generators whose outputs include plain, single
colour rasters.)
(3) Slacken the wing nuts which secure the scan coils to their housing and slide the
assembly as far back as possible-initially, this will result in a very impure raster,
especially around the edges, and possibly with 'shaded' corners.
(4) Adjust the purity magnets until a pure red area is displayed approximately in
the centre of the screen only, ignoring the raster impurity elsewhere. The ring
magnets have tabs which, apart from facilitating the adjustments, act as position
indicators. It may be necessary to rotate the rings to bring the tabs closer together
or spread them further apart; alternatively the magnets are turned as a pair in either
direction on the tube neck.
(5) Slide the scan coils forward, noting that the centre red area will now expand;
the correct position of the deflecting coils is reached when a pure red raster com-
pletely fills the screen.
(6) Switch off the red gun and switch on the green gun, and check the purity of the
raster; repeat for the blue raster. If the purity of either of these two is not accept-
able, a slight readjustment of the scan coils should be sufficient. Failing that, slide
the scan coils back again and readjust the ring magnets. In some rare cases a com-
promise setting will have to be adopted for the best overall purity of all three rasters.
(7) Lock the scan coil securing nuts.
NOTES: The action of the purity magnets shifts the three beams from their original
positions; hence, the picture centering controls may have to be readjusted. The
purity magnets also affect static convergence (see section 12.5); conversely, the
static convergence adjustments may alter the colour purity, because of somewhat
similar effects on the beams in each case. Consequently, it is often recommended to
check and adjust the static convergence before and after 'purifying' the display. It
has been reported in a few rare instances that damage to the tube resulted if the scan
coils were left for too long in their rearmost position.
PIL tubes have fixed deflecting coils, and overall purity can be obtained by adjust-
ing the ring magnets. In this type of tube the purity and the convergence magnets
form a group of rings on the neck which are clamped in position by a locking ring in
front of the assembly. Often there is an identification line marked on the magnets-
when all of them are in their optimum positions, the individual markers should
form a continuous line across the assembly. Normally, the purity magnets occupy
the rearmost position in the arrangement.
Receiver Setting-up Procedure 239

12.5 Static Convergence Adjustments

Errors in static convergence give coloured outline ('fringing') to picture detail in
the central area of the screen. This is especially noticeable on a monochrome dis-
play. The term 'static convergence' suggests that the beams converge at the centre
of the screen-the landing area if the beams remained static or undeflected.
The ideal display for this adjustment is either a crosshatch or a dot pattern;
alternatively, test cards can be used. Converged crosshatch consists of intersecting
white vertical and horizontal lines, and is produced by a specially designed signal
generator. Such an instrument can also have a dot pattern facility which is often
recommended as being more suitable for static convergence adjustments; a con-
verged display consists, then, of rows of white dots across the screen. The impres-
sion of white in either case can only be obtained if the beams land on their respec-
tive phosphors within one triad (or group of stiipes). Misconverged beams hit the
correct colour phosphors, but they are too far apart to create satisfactory colour
mixing from a normal viewing distance.
In most receivers static adjustments are carried out by means of permanent
magnets which are associated with the convergence assembly mounted on the neck
of the tube. In a few receivers a suitable d.c. supply is used for this purpose.
Delta gun tubes use a radial convergence arrangement for each beam, together
with a lateral adjustment for the blue. It will be recalled from chapter 11 that in
the delta gun tube the radial convergence magnets move the red and the green beams
diagonally across the screen, but the blue beam is shifted vertically up or down; the
lateral magnet moves the blue beam 'sideways'. The procedure for static convergence
adjustments is as follows:

(1) Apply crosshatch or another suitable signal to the receiver.

(2) Switch off the blue gun.
(3) Converge the red and green, horizontal and vertical lines in the centre of the
screen, which should give yellow crosshatch; ignore any misconvergence away from
the middle.
(4) Switch the blue gun on and converge the blue lines with the yellow to obtain
white at the centre; both the radial and lateral controls are used as required.
Ordinary in-line tubes often have the static convergence magnets so arranged
that the beams are moved diagonally across the screen in a manner similar to that
with the delta gun system. The reason for this is illustrated in plate 1 (opposite
p. 292), where part of a crosshatch display is shown. The solid lines indicate the
beam positions; the green and the blue are close together, and the red is not
converged. To converge the red lines with the other colours, the horizontals would
have to be shifted upwards and the verticals moved to the left until their correct
locations, as indicated by the broken lines, were reached. Only one diagonal move-
ment of the beam is now needed to achieve this. The same comment applies to the
blue beam should any correction be required. The green gun usually occupies the
central position, and it is not affected by the adjustments. Some tubes have the
240 Television Principles and Practice

red gun in the middle, so that the other colours are converged with respect to it.
Precision-in-line tubes have four ring magnets to adjust static convergence. The
rings form part of the assembly with the purity magnets already described. The
front pair, next to the locking ring, is adjusted first, followed by the second pair.
Whichever beam is in the centre, it will not be affected by the magnets. The follow-
ing instructions are given for a tube with the red gun in the middle.
(1) Apply a crosshatch signal to the receiver.
(2) Switch off the red gun.
(3) Slacken the magnet locking ring.
(4) Observing the centre of the display, adjust the front pair of magnets: separate
the identifying tabs slightly by rotating the magnets in opposite directions, and then
rotate them togeiher, to reduce the amount of separation between the blue and the
green lines of the display.
(5) Adjust one magnet with respect to the other, to converge the blue and green
into cyan. If necessary, repeat step (4) for the best results.
(6) Switch the red gun on and, using the method outlined in steps (4) and (5),
a.!ljust the second pair of ring magnets to converge the cyan with the red, to obtain
white.
(7) Recheck the blue and green convergence and adjust the front pair of rings again,
because of possible interaction between the magnets.
(8) Carefully tighten the locking ring so as not to disturb the adjustments.
The somewhat different method of convergence adjustment of the PIL tube is
due to the fact that four-pole and six-pole ring magnets are used. The arrangement
gives a greater freedom of movement to the beams (and the magnet~!) than is pro-
vided by the two-pole magnets found elsewhere. Identifying markers can be provided
on the rings to denote their correct position -see section 12.4.

12.6 Dynamic Convergence Adjustments

Errors in dynamic convergence cause colour fringing away from the centre of the
screen which usually tends to get worse towards the edges of the screen. The
diagram in figure 12.1 indicates the areas on the screen where maximum mis-
convergence may occur. The centre is affected by static convergence as already
explained. The top and bottom of the display are influenced by the field dynamic
convergence since it corresponds to maximum field scan. Similarly, the left- and
right-hand sides are governed by the line dynamic convergence as the errors coincide
with the extremities of the line scan.
To correct dynamic convergence errors, a cross-hatch display is again used,
although test cards may be acceptable. The receiver controls are labelled according
to the effect they have on the lines of the crosshatch pattern. Some manufacturers
provide 'flow-charts' to show the sequence of adjustments together with colour
diagrams which indicate errors to be corrected.
A delta gun tube needs a large number of adjustments; the various combinations
are presented diagrannnatically in figure 12.2 and their effects on a crosshatch are
Receiver Setting-up Procedure 241

field convergence
errors
static
convergence
error

line
convergence
errors

Figure 12.1 The areas on a TV screen identify source of convergence errors.

Correction is applied while concentrating on the errors along the vertical (field)
and the horizontal (line) axes of the picture. Details of adjustments are given in
plates 2 and 4

blue lateral

amp
~ ~amp

'I
RIG RIG RIG [Link] shape blue
matrixed symmetry differential (line convergence - radial

tilt/ '-" t only) ~ tilt

~d convergence ~

Figure 12.2 Typical dynamic convergence arrangements for a delta gun tube

listed in plate 2. They are not difficult to memorise if one studies this list in con-
junction with figure 12.1. Differential controls are used to converge the red and
green horizontal lines, while the matrixed adjustments bring together the corres-
ponding vertical lines of a crosshatch. The diagram in plate 3 illustrates the need
for the two types of controls. Part of a misconverged crosshatch display is represen-
ted here by the lines of individual phosphor dots. The red/green matrixed controls
move the two beams diagonally in the same direction, and the differential adjust-
ments cause a shift in opposite directions. For example, if the matrixed controls
242 Television Principles and Practice

were to be adjusted to move the beams upwards, the red and green vertical lines of
the crosshatch would be brought together, while the horizontal lines would also
move upwards, but their mutual relationship would not be affected. Conversely, a
differential adjustment which takes the red horizontal'line downwards shifts the
green line upwards and they converge; at the same time, the verticals will effectively
move to the left, with their relative position otherwise unchanged.
If the complete sequence of dynamic convergence adjustments is to be carried
out, then:

(1) Apply crosshatch signal to the receiver.

(2) Switch off the blue gun.
(3) Carry out the red-green adjustments in the sequence indicated by the numbers
in the table in plate 2-field convergence first, followed by the line controls.
(4) Switch the blue gun on and converge with the yellow to obtain white.
NOTES: (a) There may be some loss of static convergence during the above proce-
dure; it is therefore recommended to check and adjust this as necessary between the
dynamic adjustments.
(b) The respective symmetry and differential controls should be adjusted in conjunc-
tion with each other.
(c) Large initial errors can only be corrected by repeating the entire sequence.
(d) The 'tilt' and 'amp' adjustments tend to affect each other; the degree of inter-
action can be reduced by operating both controls together.
(e) In some cases it is very difficult to obtain perfect convergence over the entire
screen area; therefore it may be necessary to settle for compromise settings. As
much of the central area of the display as possible should then be converged satis-
factorily, since the eye tends to concentrate on this part of the picture.

In-line tubes require very little dynamic convergence; the adjustments are usually
straightforward, as listed in the table in plate 4. In fact, this list is comparatively
long, since many makers may omit certain controls altogether. The gun which
occupies the centre position will not be affected by the adjustments (usually the
green, sometimes the red). The following sequence is recommended by one
manufacturer (Rank):

(1) Apply crosshatch signal to the receiver.

(2) Adjust the line symmetry (balance) control to correct red/blue crossover of
central horizontal lines; if difficulties are experienced, the convergence coils may
be rotated slightly or the individual coil assemblies can be tilted.
(3) Carry out line convergence adjustments (see plate 4).
(4) Carry out field convergence adjustments (see plate 4).

12.7 Pincushion Distortion Adjustments

Pincushion errors show up as bowed verticals and horizontals, especially near the
edges of the screen. This could be particularly important in large screen receivers,
 Receiver Setting-up Procedure 243

or if 1100 tubes are used. Because of the effects produced by these errors, they are
best observed on a crosshatch display.
Where a simple transductor is used, the adjustments aim at straightening the
crosshatch pattern. In a more involved arrangement the E-W keystone and the
E-W pincushion controls are adjusted for straight verticals at the sides. These are
followed by the N-S pincushion amplitude and N-S phase, to achieve straight hori-
zontallines at the top and bottom of the picture. Effects of various pincushion
correction adjustments on the shape of the picture are shown in figure 12.3.

[JOtJ
+. I
• •

-O--O-_D_
N-S balance N-S amplitude N-S phase

E-Wamplitude E-W keystone

Figure 12.3 Effects of various pincushion correction adjustments on the shape of

the picture. Arrows indicate the direction and the position of the applied correction

12.8 Grey Scale Adjustments

Incorrect grey scale is indicated by colour 'tint' or 'cast', especially noticeable on

a monochrome display. Such unwanted colour may appear over the whole area of
the screen, or it could be associated with either the brighter (highlights) or the
darker (lowlights) parts of the picture. Correct grey scale is necessary, not only
because of its effect on a black and white programme, but also because an impres-
sion of wrong colours could easily be produced if the balance between the three
guns were upset. It has already been stressed that before investigating a 'wrong
colour' type of fault, the servicing engineer must ensure correct grey scale tracking.
The adjustments are in two groups:
(1) Highlights-to control the amount of video drive (whether R, G, B or luminance)
to each of the three cathodes of the tube.
(2) Lowlights-to control the beam cut-off voltage for each gun; this is governed by
the first anode (AI), sometimes referred to as the screen grid (G2), potential.
244 Television Principles and Practice

The procedure is best carried out with the aid of the standard colour bar display
which has the colour information completely removed; it is then known as the grey
scale pattern. Some signal generators have the facility to produce the luminance
waveform only; otherwise the colour control has to be turned down or the receiver
detuned slightly until colour drops out. Test cards are a suitable alternative. The
adjustments should be carried out in subdued lighting, as the changes in the tube
output are then easily noticed.
Correctly adjusted video drives should produce white light output which corres-
ponds to the standard known as llluminant D 6500. The figure refers to the equi-
valent temperature to which a black object would have to be heated to reach the
required 'shade' of white. Ideally, a reference source ofD 6500 should be used,
since the broadcasting standards are also based on it. The reference source consists
of a fluorescent tube which provides the desired light output, e.g. Tropical Daylight
(Atlas Lamps). Sometimes a suitable grey scale fllter is fitted over the tube to give
not only white, but also a few shades of grey and black. The necessary adjustments
are made while holding the tube against the receiver screen.
If the grey scale is grossly out of adjustment, then it is advisable to consult
makers' instructions before any setting up is carried out. In some receivers a number
of conditions have to be fulfilled prior to such adjustments. For example, it may be
necessary to set the grid-cathode voltages to a prescribed level, or a switch needs to
be operated which turns off the field time base and applies certain fixed voltages to
the tube electrodes etc.
For delta gun and ordinary in-line tubes the recommended procedure of
readjusting the grey scale (thus disregarding any special conditions mentioned above)
is summarised as follows:

(1) Apply a grey scale pattern to the receiver.

(2) Adjust the contrast and brightness control to ensure equal brightness change
between each step of the grey scale.
(3) Observe the black and the dark grey portion of the display ('lowlights') for any
sign of coloration; the appropriate Al control(s) have to be turned down to reduce
the unwanted tint.
(4) Observe the white and the light grey portion of the display ('highlights') for any
sign of coloration and adjust the video drive controls accordingly. Some receivers
have only two adjustments, with the cathode of the third gun (usually the red)
being at a fixed potential.

Precision-in-line tubes have a common Al potential applied to all three guns.

Therefore the cut-off point and the maximum video drive are set at the video output
stages for each gun in turn. The lowlights are controlled by altering the d.c. bias of
the individual output transistors; the highlights are adjusted by varying the gain of
the respective video output amplifiers.
Receiver Setting-up Procedure 245

12.9 Test Cards

The purpose of test cards is to provide a stationary TV display which can be used
for the assessment of the receiver performance (both colour and monochrome) and
for setting-up. The latter application has its limitations; more specialist display
patterns may be preferred in some cases, as outlined in the preceding sections.
Test cards are designed jointly by the broadcasting companies and the manu·
facturing industries. The display is revised periodically to include new techniques,
extend its usefulness, etc. The chronological order of the appearance of test cards
is indicated by the appropriate letter. Currently in the UK two types are broadcast:
Test Card F and Test Card G.
Test card F
Test card F is shown in plate 5; it is televised from a picture seen by the studio
camera except for the colour bars, which come from a suitable generator. The
purpose of each section of the display will now be explained.
(1) Standard colour bars at the top-the first 24 lines of each frame are modulated
with the colour bar signal to facilitate decoder fault fmding (chapter 7), both by
viewing the bars on the screen and by waveform examination with the aid of an
oscilloscope. In the latter case it may be necessary to trigger the oscilloscope time
base with the field flyback pulse. The visibility of the colour bars depends upon
the receiver height adjustment-see item (5) below.
(2) Coloured castellations at the bottom and on the left- and right-hand side of the
picture, together with the colour bars at the top, indicate the performance of the
4.43 MHz subcarrier oscillator and the associated control circuit ([Link]). If the
top bars are correct, then the decoder is not affected by the field synchronising
sequence during which the colour burst is not transmitted. A temporary lack of the
burst signal should not upset the a.c.c. (variations in saturation) or the oscillator
(variations in saturation or 'unlocked' colour). The castellations at the bottom show
whether the oscillator regained full synchronisation at the end of the picture. The
castellations on the left (red in top half, blue at the bottom) introduce colour at
the beginning of the picture line, which could upset the oscillator if the burst gating
was incorrect (allowing picture content through the gate). Such a fault would affect
the part of the centre colour picture level with the coloured castellations. By com-
parison, the alternate black and white pattern has no chrominance content and the
oscillator should remain unaffected. The right-hand side yellow castellations indi-
cate the oscillator stability at the end of the picture line.
The end-of·line castellations-both the black and white and the colour-test the
performance of the sync. separator, which must not allow the video (and the
chroma) information to affect its action. A fault in this respect would show as
'wavy' verticals level with the bright sections of the pattern.
(3) The background grid pattern-this forms a crosshatch display; some of the
white lines have a black outline, to make the assessment of receiver dynamic con-
vergence easier. Static convergence is indicated by the white cross in the centre
246 Television Principles and Practice

('noughts and crosses' on the blackboard). Adjustments, however, ought to be

carried out with the aid of a proper crosshatch generator.
(4) The colour picture in the centre-includes areas of flesh tones and bright
colours to assess the overall quality of colour reproduction. Any defects in the
decoder will clearly reappear in the centre picture. Sound/chroma beat pattern,
Hanover blinds, subcarrier dot interference, ghosting and other effects can be
easily observed. The surrounding circle, together with the background crosshatch
can be used to check picture linearity.
(5) White 'arrowheads'in the centre of the edge castellations (top castellations and
the top 'arrowhead' are almost completely obscured by the electronically inserted
colour bars) indicate the limits of the picture. This allows the receiver height, width
and centering controls to be set up where applicable. Modern TV receivers use
picture tubes of the same aspect ratio (width/height) as that of the camera tube at
the transmitter (4:3). Therefore the display should be set to within the limits of
the points of the 'arrowheads' (the top colour bars to be inside the picture). Older
receivers employ an aspect ratio of 5:4; the height could then be set as above, while
the width is adjusted until a linear picture is obtained.
(6) Grey scale rectangles in the column to the left of the centre circle-the six small
rectangles of varying brightness can be used to check the grey scale, contrast and
brightness. The difference in brightness between adjacent rectangles should be
approximately constant. Inside the bottom and the top rectangle there is a brighter
spot. Too much brightness causes 'white crushing'; not enough, 'black crushing' (it
can also be caused by faulty video amplifier biasing). This appears when the top or
the bottom spot, respectively, merges with its surrounding. The top rectangle in
the grey scale does not correspond to peak white; therefore the pattern may not be
suitable for the adjustment of the 'highlights'.
(7) Frequency gratings in the column to the right of the centre circle-six rectangles
with vertical black and white stripes of varying thickness. These are produced by a
square wave signal which turns the beam alternately on and off. The size of the
stripe depends on the frequency of the signal (see the chequerboard concept in
chapter 2); hence, the display shows the frequency response of the video amplifier
and indirectly the bandwidth of the i.f. strip. The order of the video frequencies
represented from the top to the bottom rectangle is: 1.5 MHz, 2.5 MHz, 3.5 MHz,
4 MHz, 4.5 MHz, 5.25 MHz. In a colour receiver the 4.5 MHz pattern will be very
blurred because of the chroma notch fIlter in the luminance amplifier. The 4 MHz
and 5.25 MHz gratings might also be difficult to resolve, owing to a bluish colouring
that could be displayed. This is known as cross colour and it is caused by the high
frequency content of the luminance signal which lies within the passband of the
chrominance amplifiers (the action of the demodulators gives rise to spurious
colour).
(8) Black rectangle within the white rectangle above the centre circle gives indica-
tion of the low frequency response of the video amplifier (and to some extent the
alignment of the i.f. vision carrier). Poor response produces streaking at the right-
hand edges of the rectangles.
 Ope rating Field convergence adjustments Operating Line convergence adjustments
sequence errors viewed along vertical axis of screen sequence errOrs viewed along horizontal axis of screen

4 RIG matrixed to converge

RIG matrixed to verticals at left and right
1
straighten verticals 'T[3H"
- - amp
m]" RIG differential to
5
straighten horizontals

RIG differential to
[8]
2 converge top and
I bottom horizontals RIG line symme try to
6 remove crossover of
~ I horizontals
t
BJ t
1 I
.; Blue (radial) to straighten
....
RIG fie ld symmetry to 8 centre horizontal (amp) and
remove crossover of ~ remove their crossover (tilt)
3
ve rticals and converge t t
horizonta ls as in 2 above

Blue (radial) shape to remove

rn 9 any irregularities along
horizontal lines

7
Blue (rad ia l) to converge
lOP and bottom
horizonta ls
IA I Blue lateral to converge
10 vertica ls at left anti
~ [jJ right

Plate 2 Dynamic convergence adjustments-delta gun tube; the recommended

sequence of the complete procedure is indicated by the number to the left of each
diagram.
 =====t:t:t:====================::t=====}-
- - - - r - - - - - - - - - - - - - - co nverged bl u e and
green horizontals
I
I
I direction of
I movement
of beam ,
I
II
I
I
I
t
---- misconverged red horizontal
I and vertical
I
converged blue and
green verticals

Plate 1 Effect of static convergence adjustment-in-line gun tube. Static correction

fields act diagonally-as in delta gun tube. This ensures convergence of both vertical
and horizontal lines of crosshatch display. Solid red lines represent misconvergence;
these must be moved to take up the converged position-broken red lines.

direction of ___ / ~ /G matrixed convergence

beam • •
movement
• •
I.• • ••• • • •
•
RIG ,.- ••• • ••
differential
""- • •
• ••• • • • • • •
convergence

• •
• •
• •
Plate 3 Dynamic convergence in a delta gun tube_ The 'lines' of crosshatch consist
of rows of illuminated phosphor dots; the movement of the beams aims at striking
rows of dots which are closer together. Owing to the radial (diagonal) shift of the
RIG beams (see figure 113), matrixed adjustments converge the 'verticals', and
differential adjustments converge the 'horizontals' of the display.
 Plate 5 Test Card F.

Plate 6 Test Card G.

 Line convergence adjustments Field convergence adjustments
(if included in receiver)

[2l
X U~ 'Ymm~" [""",,[
~
Red top and bottom to
1 to remove crossover of 4 converge horizontals to
red and blue horizontals yellow

L[j ' ' -"" ",",

~
Blue top and bottom to
[.m""..d'
2 -- - inductor) to converge 5 converge horizontals
right-hand side verticals to white

3
[
-- -
',H'" ,~, [""
] potentiometer) to converge
left-hand side verticals
6

DO Top and bottom overlay

to straighten centre
verticals

Plate 4 Dymanic convergence adjustments-in-line gun tube. The recommended

sequence is indicated by the number to the left of each diagram. Apart from the
electrical controls shown, some movement of the convergence assembly may be
needed to straighten red and blue 'horizontals'. If required, the purity magnets
are carefully adjusted to eliminate crossover of red and blue 'verticals'; ensure the
rings shift the raster up or down-not sideways, or purity will be affected. Note: The
green beam is central in this example, and is not influenced by the adjustments.

Plates 5 and 6 (photographed from TV screens) are

reproduced by courtesy of the BBe.
Receiver Setting-up Procedure 247

(9) Diagonally striped areas originating from each corner of the display-used to
check the uniformity of picture focusing. Good focus is most important in the
centre of the picture, but it should also extend towards the edges. Most of the
striped area should be well focused, although in wide angle colour tubes some form
of compromise may be necessary.

Test card G
This pattern is transmitted from a special signal generator. It is based on a design by
Philips, but certain modifications have been introduced by the BBC. Other broad-
casting authorities may use either the original design (type PM 5544) or their own
version of it. Many facilities are similar to those offered by test card F; therefore
in the following description cross-reference will be made as appropriate. A photo-
graph of test card G is given in plate 6.
(1) Black and white castellation at the sides-check on the effect of picture content
on the performance of line synchronisation and the synchronising pulse separator.
A fault here produces wavy verticals level with the white blocks.
Picture width and centering are adjusted to ensure that the side castellations are
just visible with the aspect ratio 4:3, and excluded if the ratio is 5:4 (see item 5, test
card F).
The stability of the receiver black level clamp is judged by comparing the black
(and/ or the white) at the beginning of the line and at the end.
(2) Top and bottom castellations-check the performance of the field synchronising
circuit by introducing a pulsed video waveform. These are also used to centre the
picture and adjust the height to include the two borders.
Sometimes the top castellations may be replaced by standard colour bars, as
item (1), test card F.
(3) Background crosshatch pattern-check the dynamic convergence; static conver-
gence is judged by the intersecting lines at the centre of the picture.
The horizontal lines are used to check receiver interlace-if faulty, the centre
line would be of a different thickness from that of the others.
The second and third squares of the crosshatch down from the top on the left-
hand side, together with the neighbouring pair of castellations, are made up of
unswitched (R - Y). The combined action of the PAL delay line and the PAL
switch should cause no chroma output-the pattern appears grey ('colourless'
information). Faults in the delay line adjustments would show up as Hanover
blinds in that small area or in the subsequent coloured pattern (see item 4 below).
Defects in burst gating (at the beginning of the line) would cause either variable
saturation or unsynchronised colour later on the same level as the test squares.
(4) Coloured 'brackets' on the left- and right-hand sides of the centre circle-these
coloured areas are used to further check the performance of the decoder. The left-
hand side vertical stripe of the 'bracket' is produced by a chrominance signal for
which (B - Y) = 0 and (G - Y) = 0; however, the top half corresponds to negative
248 Television Principles and Practice

(R - y), which gives turquoise colour, and in the lower half is the positive (R - Y)
-purple. The smaller sections at the top and bottom next to the (R - Y) signals
consist of (B - Y) and (R - Y) of such polarities and proportions that the matrixed
(G - Y) = o. Two colours are then possible, since the constituent (R - Y) and
(B - Y) can change their polarities (bluish at the top and reddish brown at the
bottom).
The vertical columns of the right-hand side 'brackets' consist of the negative
(B - y), top half; positive (B - y), lower half. The small (G - Y) = 0 areas are as
before. This part of the test card can be used to check the operation of the
synchronous demodulators and the (G - Y) matrix. When viewed on an oscilloscope
(triggered at line rate), the decoder waveform will consist of sections corresponding
to the colour information displayed across the TV screen. For example, the (B - Y)
demodulator should have no output waveform in the two left-hand side sections
immediately after the line pulse [no (B - Y) in the 'colourless' information on
the left, nor in the ± (R - Y) which follows] . There should be an ouput in the last
section of the waveform, though, which is due to the ± (B - Y) towards the end
of the picture. Following the above reasoning, both the (R - Y) demodulator and
the (G - Y) matrix can be checked.
(5) The centre circle-for picture linearity checks.
(6) Large black rectangle at the top-low frequency response (see item 8, test card
F).
(7) Thin, short vertical line below and to the left of the low frequency rectangle-
check for 'ghosting'.
(8) Row of alternate dark grey-white rectangles-checks for transient response:
freedom from ringing or other distortion primarily in the video amplifiers.
(9) Standard colour bars-check overall colour quality of the receiver.
(10) Frequency gratings-row of six rectangles with vertical stripes: the frequency
range is the same as in test card F.
(11) Grey scale rectangles below the frequency gratings-the full range from black
to peak white is available; hence, complete adjustments can be carried out if
required.
(12) Coloured area at the bottom of the circle, red on a yellow background-check
the luminance delay line circuitry; if correct, the red block should line up exactly
with the centre square of the crosshatch immediately below.
13 Television Aerials and
Systems

13.1 Properties of V.H.F. and D.H.F. Signals

As we have seen, the bandspread of one TV channel can be 8 MHz or more, depend-
ing on national standards. In order to accommodate a number of channels of such
large bandwidth, it is necessary to use high carrier frequencies, in both the v.h.f. and
u.h.f. regions; microwave transmissions may be adopted in the future. The propaga-
tion of electromagnetic waves is very severely affected by the order of their fre-
quency. The higher the frequency the more pronounced is the 'line of sight' dis-
tribution of the signal. That is, at the higher frequencies the receiver installation
ought to 'see' the transmitting aerial. The process of reflecting the signals at fre-
quencies above 40 MHz from the upper layers of the atmosphere is no longer
reliable for normal communication purposes. It will also be appreciated that the
signal strength will be considerably reduced as the reception area becomes more
distant. In countries where the population density is low it is necessary for the
transmitter to cover a large land area; in these cases v.h.f. transmission is particu-
larly useful, because it provides increased coverage. In the UK, with its large popu-
lation centres, often only 40 miles apart, u.h.f. transmissions offer better facilities
in allocating the available channels without adversely affecting the reception in
the neighbouring areas. Additionally, since very little man-made interference is
generated at u.h.f. frequencies, transmissions in this frequency band result in better
quality reception.
The reception area in the true 'line-of-sight' region is governed not only by the
earth's curvature, but also by the terrain. In fact, TV signals can be received beyond
the horizon, owing to a small amount of 'bending' (refraction and diffraction) in
the line of the radiated signal. This effect is specially noticeable at frequencies in
Band I. The usable reception area can be further extended by raising the trans-
mitting aerial. Whenever possible, the primary transmitters are on high ground and
have very tall masts. The positioning of relay or 'fill-in' stations is governed by
other factors; in fact, it may be desirable to restrict their range, to prevent inter-
ference with neighbouring transmitters. The signal coverage is also influenced by
the radiated power of the transmitter, which may be expressed in watts or kilowatts.
Although the signals at such high frequencies are not normally reflected from
the ionosphere, they can be reflected from the surface of the earth, from hills, tall
249
250 Television Principles and Practice

buildings -and other obstacles. When this happens, the receiving aerial will get at
least two signals-the direct and the indirect wave. The indirect, or reflected, wave
has to cover a longer distance and arrives at the receiver after the direct signal. The
resulting time delay produces a second image on the screen-known as a ghost
image; it is possible to receive a number of reflected signals in this way, which result
in several ghost images. The latter must not be confused with the symptoms of
instability in the receiver amplifiers. This manifests itself as a series of equally spaced
outlines, while 'ghosting' produces irregularly spaced images.
Aircraft passing in the vicinity of the receiver can give rise to a fluttering picture
(aircraft flutter), which is due to the reflected signal from the aeroplane. As the air-
craft moves, so the reflected and direct signals go in and out of phase with each
other, because the length of the indirect path alters. The rate of picture flutter is
about two or three times per second. Some improvement is obtained by using a
highly directional aerial, and receivers with fast-acting a.g.c. are less prone to this
effect.
The signal strength at any particular location is expressed in mV/m or p.V/m; it
gives the voltage induced in an aerial of effective length of 1 m. Broadcasting
authorities issue maps of the transmitter service areas showing signal strength
'contour lines'. From such information it is possible to anticipate the effectiveness
of a particular aerial design before its installation. In some cases it is necessary to
use a portable field strength meter in order to help choose and position the aerial
for best reception.

13.2 Fundamentals of the Receiving Aerial; the Dipole

The receiving aerial must capture some of the transmitted energy and feed it to the
tuner via the cable. In this respect, the design of the aerial must match the remain-
der of the installation to achieve maximum signal transfer-that is, the various
impedances involved ought to be equal: the impedance of the aerial should have
the same value as that of the feeder cable, which, in tum, should be equal to the
input impedance of the first r.f. amplifier in the tuner.
The aerial should be directional in order to discriminate against any unwanted
signals (e.g. reflections) and it must have adequate gain to produce sufficient
noise-free output from the available signal strength.
V.H.F. and u.h.f. receiving aerials are based on a half-wave dipole, which in its
simplest form consists of two metal rods mounted on a suitable support. The
length of each rod is equal to one-quarter (approximately) of the wavelength to be
received; thus the overall dimensions of the dipole approximate half the wavelength
of the signal [see figure 13.2(a)] . Electrically, such an arrangement behaves like a
resonant circuit, which makes it frequency-selective. The rods themselves offer an
inductance (similarly to a length of any conductor); the necessary capacitance is
distributed along these elements-in this respect, it resembles the action of stray
capacitance.
Television Aerials and Systems 251

The transmitting aerial is also based on the dipole and is fed with energy from
the transmitter amplifier. The flow of current in the elements gives rise to an
electromagnetic field and an electric field which simultaneously radiate outwards
from the aerial. The two fields are radiated at right angles to each other. The
electric field can be arranged to spread in the required direction in either the verti-
calor the horizontal plane, depending upon the orientation of the transmitter
dipole. Therefore it is said that the transmitted signal is either vertically or horizont-
ally polarised.
When the two fields reach the receiving aerial, the position of its dipole must be
such that, for a maximum signal pick-up, the elements are in the plane of the arriv-
ing electric field. This principle gives rise to either vertically or horizontally polarised
aerials. Normally, in the UK the main u.h.f. transmitters employ horizontal polarisa-
tion; signals from relay stations are usually vertically polarised, to prevent mutual
interaction. The effect of horizontal polarisation makes the dipole somewhat direc-
tional. The receiving aerial is capable of picking up equal amounts of signal from
the 'front' or from 'behind' (that is, the 'broadside' facing the transmitter); radia-
tion approaching from the 'sides' will not be readily accepted. In a vertically
polarised system the dipole receives signals from all points and the aerial is no
longer directional.
The ability to develop different voltage outputs, depending upon the direction
of the incoming signal, is often illustrated by the polar diagram of the aerial. Gener-
ally, it consists of a loop, or a number of loops, drawn in relation to the common
point, known as the pole, and placed upon the sketch of the aerial itself. In prac-
tice, the junction between the cable and the aerial marks the pole of the diagram;
the distance from there to any point on the curve(s) represents the output developed
by the signal arriving from that particular direction. Therefore the curve becomes a
form of directional pattern of the aerial. This principle is illustrated in figure 13.1;
a polar diagram for a dipole is shown when vertical polarisation is used in (a) and
horizontal polarisation in (b). The direction of each arrow represents the possible
line of signal pick-up; the length corresponds to the output which would be fed to
the cable. Often the manufacturer quotes the outputs in relation to the possible
maximum, so that they are expressed as ratios or in dB. As the strongest signal is
always received from the 'straight ahead' direction, those arriving from elsewhere
are weaker and the figures given are negative dB.
Vertical polarisation gives a circular polar diagram, which confirms that the
dipole has no directional properties. As a result of horizontal polarisation there are
two loops in the form of a figure of eight. It indicates that the signals arriving from
the front, or from the back of the dipole, can develop equal outputs. Those from
the sides are relatively ineffective.
The polar diagram shows the angle of acceptance of the aerial-that is, the angle
between the directions of the received signals when the output is reduced by 3 dB
(that is, 0.707 of maximum sensitivity). As shown in figure 13.1(b), this angle (0:)
is approximately 800 for a simple horizontal dipole (±40° from the maximum).
The aerial must have a certain bandwidth to ensure relatively constant response
252 Television Principles and Practice

90 0

polar diagram

1800~-------------Jf------------'~0°

270 0
(a)

maximum
reception
900

1300 500
"" /
/

1500 " /

direction of
1800 _ _ -",!"'~'illllp. ___ minimum
reception

polar
diagram

2700
maximum
reception

(b)

Figure 13.1 Polar diagrams of a simple dipole aerial: (a) vertical polarisation-
equal pickup from all directions; (b) horizontal polarisation-the dipole now has
some directional properties; angle a is the acceptance angle
Television Aerials and Systems 253

,,/4 ,,/4 C~_----, )

~ /-~~ner
It I
to tuner
It I
to tuner
/,,/4
(a) (b) (c)

length 0.511\
length 0.48" length 0.46"

..!!.
oj
C.
"C

cross arm maximum

sensitivity
insulated
cable
j~.mction

r--0.24,,-+0.13,,-1
(d) (e)

back
- - axis of array
lobe

side I
lobes I

(f)

Figure 13.2 Examples of receiving aerials: (a) dipole; (b) folded dipole; (c) V-
shaped dipole; (d) three-element Yagi array; (e) polar diagram of a three-element
array; (D polar diagram of a nine-element array. Note: A(lambda) is the wavelength
of the received signal:
300
A= -------m
frequency (MHz)
254 Television Principles and Practice

to the frequencies associated with one channel, or even a group of channels. The
bandwidth of each u.h.f. channel in the UK is 8 MHz, but a single aerial installation
must be able to receive at least four programmes, which in turn are separated from
one another by gaps of unused channels. Therefore the effective frequency range
corresponds to a group of 11 channels, which require a bandwidth of 11 x 8 =
88 MHz. It appears that a simple dipole cannot fulfil this, especially at the lower
frequency end. V.H.F. aerials are usually designed for a particular channel and a
dipole may then be useful.
An important property of an aerial is its impedance, which is the ratio of the
induced voltage to the resultant current. The need for impedance matching has
already been stressed if maximum signal transfer to the receiver is to be achieved.
Serious mismatch in the complete installation can lead to a loss of signal and signal
reflections which would appear as 'ghosting'. The impedance of a simple dipole is
73 n at the frequency of resonance; this figure is often referred to as being nomin·
ally 70 nor 75 n. The dipole is thus easily matched to a coaxial cable downlead
whose impedance is also approximately 75 n.
Another form of dipole is the folded dipole, which could be constructed by
folding back a metal rod on itself so as to form a very tight loop, as shown in
figure 13 .2(b); the overall dimensions are still based on the half-wave principle.
Some of its properties are similar to those of the simple dipole, except that the
impedance is now 300 n and the bandwidth is also increased. The impedance may
be modified further by making the 'top' and the 'bottom' parts of the assembly
from rods of different cross-sectional area or shape; such an arrangement is useful
in the design of multielement aerials. Because of the impedance mismatch, the
folded dipole alone is not suitable for connecting to the coaxial cable, but it can be
used directly with a twin lead parallel line, which is favoured in some countries.

13.3 Multielement Aerials

The simple dipole aerial has a relatively low signal output, except in areas of high
signal strength. Even there, its lack of directional properties-that is, its inability to
discriminate against unwanted transmissions-can make it unsuitable. Finally, the
available bandwidth of a dipole is restricted. Some improvement can be made by
bending the arms of the dipole to produce a V-shaped arrangement. This modifica-
tion improves the directional properties and widens the frequency response.
A second element can be mounted behind the dipole to form a reflector. If the
distance between the two is carefully chosen, the aerial will produce a greater out-
put than the dipole and have better directional properties by rejecting signals from
the back. The gain of an aerial is expressed in dB and it is based on the signal out-
put of a particular array with respect to the output of a single dipole. Thus the gain
of an H-type arrangement can be about 5 dB-that is, approximately 1.8 times more
voltage than from a dipole alone. There is also a reduction in the signal pick-up
from the unwanted direction. The ratio of the maximum signal developed from the
front to that from the backward direction is known as the front-to-back ratio, and is
Television Aenals and Systems 255

normally expressed in dB. The corresponding figure for the H-type aerial is up to
10 dB.
The reflector i~ of a slightly greater length than the dipole, but it also forms a
tuned circuit at the required frequency. The signal induced in the reflector is then
reradiated towards the dipole, to increase the resultant output. At the same time,
the somewhat involved phase relationships between the currents in the two elements
will cause partial cancellation of the unwanted pick-up.
Additional elements can be placed in front of the dipole, which are then called
the directors. Their dimensions and mutual spacings are critical and depend upon
the frequency of the channels to be received. The directors also pick up the signal
energy, only to reradiate it ultimately towards the dipole. The gain increases with
the number of elements in the array. There is a practical limit, however, due
either to the physical size, especially in Bands I and III, or due to electrical factors
which cause reduced impedance and bandwidth. The advantages of a multielement
aerial of the type described, also known as a Yagi array, are high gain and good
directional properties. A diagram showing the typical dimensions of a three-element
array and its polar diagram are in figure 13 .2(d) and (e), respectively.
The polar diagram of an aerial with many directors consists of a number of
lobes; the main one is along the axis of the array which indicates maximum sensitiv-
ity. An example of this is shown in figure 13 .2(t). The angle of acceptance tends to
become narrower as the number of elements increases. The resultant reduction in
the width of the main lobe makes the positioning of the aerial more critical, and a
few degrees 'off beam' might produce an unusable signal. The existence of the side
lobes in the polar diagrams must be taken into consideration. Either unwanted
signals may be received from those directions, or the required signal, but of in-
adequate strength, could be produced. The front-to-back ratio of a Vagi array is
good, being approximately 30 dB; the gain of a u.h.f. 10-element aerial can be
typically 12 dB.

13.4 Special Aerial Designs

The disadvantage of a simple Vagi array (consisting usually of a folded dipole and
straight rods as both the reflector and the directors) are: insufficient gain in extreme
fringe areas and relatively narrow bandwidth to cover reception on different bands
or different channel groups. It has not been practicable to eliminate both disadvant-
ages with one design. Aerial design is based both on theoretical considerations and
on the practical experience of manufacturers. The number of elements in an array
does not in itself guarantee good performance.
To improve the front-to-back ratio, the reflector may be in the form of a grid,
which may either be flat or parabolic.
The improvement in gain would require many additional elements. At u.h.f. the
economical optimum for a Vagi array is 18 elements; to increase the gain by another
3 dB, it would be necessary to double the size of the aerial. In fact, stacked arrange-
ments are sometimes used consisting of two arrays side by side (or stacked in the
256 Television Principles and Practice

vertical direction), their dipoles interconnected and joined to the common down-
lead. The additional array increases the overall gain by 3 dB, and also helps to
remove some of the signal reflections from the ground.
V .H.F. aerials, because of their greater physical dimensions, produce a greater
output at a given signal strength. The design of some u.h.f. aerials also attempts
to present a greater effective length of the elements to the incoming signals. This
is done without increasing the overall size of the aerial and still maintaining the
critical dimensions of the rods related to the wavelength. Such designs use X-shaped
dipoles and directors; the dipole may be cut for full instead of half wavelength.
One manufacturer quotes a gain of21 dB for an array based on this principle, with
an aerial consisting of 21 directors, dipole and a reflector, the angle of acceptance
being only about 12°.
A wide bandwidth aerial is usually designed at the expense of gain; hence, it is
suitable where the signal strength is high. A log-periodic array is very suitable to
cover a very wide bandwidth, which can extend from v.h.f. to u.h.f. An ordinary,
Vagi-type, array has a limited frequency range since its electrical properties change
drastically at frequencies considerably different from the design figure. As the dia-
gram of a log-periodic array in figure 13.3 shows, the length of the individual rods

,
" ,,
,,
,,
" ,
to _
" "-
'>0 _from
tuner / transmitter

/1
./
/
./
./

/
I,r./
./
./
./
/
/
./

Figure 13.3 Arrangement of elements in a simple log-periodic aerial array

increases with the distance from the apparent apex (point 0 in the diagram) of the
aerial. The opposite elements along the beam are electrically connected. The proper-
ties of a log-periodic aerial repeat themselves periodically as the frequency increases
along a logarithmic scale. For example, if the array was designed to have an imped-
ance of 70 n and a gain of 5 dB at 50 MHz, it will also possess the same specifica-
tion at 100 MHz, 200 MHz, 400 MHz and 800 MHz. In appearance, the array has
Television Aerials and Systems 257

small elements towards the front and they become larger towards the back. The
low gain comes from the fact that at any particular frequency only a few of the
elements are actively resonating, giving a relatively low contribution to the overall
output.
For the reception of Band I and III a combined aerial can be used with perhaps
only one element for the lower frequency; the remainder are designed for the
higher frequency range. If a large gain is needed, however, two separate aerials are
necessary, but in order to simplify the installation, only one cable is provided. The
two arrays feed via a diplexer, which is a splitting network to ensure that the signals
produced by one aerial are not affected by the presence of the other one. An
example of a diplexer circuit is shown in figure 13.4. It consists of two T network
fIlters: L 1 , C1 , L2 pass signals of Band I while offering a high impedance to Band III,
and C2 , L 3 , C3 pass the frequencies of Band III and block Band I signals.

L, L2 c2 c3
~~~~~~~~~~~~~

Band 1 to Band 3
aerial
c, tuner aerial

Figure 13.4 Diplexer circuit

13.5 Aerial Feeder Cables

The cable is an important part of the installation, as its properties may affect the
picture quality. In the UK and in many other countries the feeder is of a coaxial
type, because it offers several advantages over a parallel two-wire line. The outer
braiding, or screen, prevents the cable radiating its signal to interfere with other
signals, it also ensures that little interfering signal will be picked up. The impedance
of a correctly installed coaxial cable is unaffected by adverse weather conditions,
pollution, etc. The value of this impedance is 75 n, which should make it suitable
for direct connection to a simple dipole. Unfortunately, at the lower v.h.f. range
there can be problems, because of the current distribution in the cable, which does
not produce a balanced electric field. A balun (balanced to unbalanced) transformer
can be used to achieve suitable matching. Such a device may not necessarily resemble
a more conventional type of transformer, as it can be in the form of a short sleeve
surrounding the cable.
The presence of the cable introduces some reduction in the magnitude of the
signal reaching the tuner, which can become significant at high frequencies. To keep
the losses down, the conductors can be of a relatively large cross-sectional area; very
258 Television Principles and Practice

low loss cables have a solid centre conductor instead of a stranded type. Signal
attenuation can be also due to the losses in the insulation between the core and the
sheath; at frequencies up to 300 MHz the insulation could be made of either solid
or foam polythene. For u.h.f. application the insulation is 'semi-air'-namely the
polythene is extruded so as to form a large air cell structure. The plastic is needed
only to support the inner conductor.
When installing such cables, it is necessary to avoid large physical stresses, because
either the inner conductor or the polythene dielectric can be easily damaged-often
without any obvious external signs. Therefore one must not apply forces likely to
cause stretching, leave a long piece of cable hanging without any support, etc. In-
sulation must be removed very carefully, to avoid nicking the centre core, which
could cause a fracture at some later time. A suitable clamp should be used at the
aerial end, to prevent water entering the cellular structure of the polythene-any
moisture here will cause very severe loss in the signal strength. Kinks in the cable
can compress the inside so much as to cause a short circuit between the outer and
the inner conductors; also, such localised irregularities can lead to signal reflections
which cause ringing (short length of cable) or ghosting Oong cable run). When the
cable has to be bent, the radius must be at least five times the cable diameter. Where
the cable is stapled to the walls of the building, it must be done carefully so as not
to compress the cable; again, irrespective of any obvious mechanical damage, the
irregularities produced in the aerial downlead can cause signal problems, as already
described. An interesting point to note is that signal reduction could occur when the
cable run is longer than 5 m and the staples are fixed at regular intervals which
happen to be equal to half the wavelength of the received channel frequency (see
the wavelength formula in figure 13.2).
Signal loss in cables can be expressed in dB per 100 m run at a given frequency
(e.g. for the cable when it is used at 50 MHz, 200 MHz or 800 MHz); these figures
are available from the supplier and they must be considered to ensure adequate
signal feed to the tuner.

13.6 Choice of Aerial Installation

The choice of the type of aerial installation is governed by the following factors:
signal strength in the given reception area, possible source of signal reflection and
obstruction, ·presence of interference and the performance of the receiver itself.
A simple dipole should only be considered where the signal strength is
high and there is freedom from interference, reflections, etc. Directional aerials
which involve multielement arrays must be used in all other circumstances. V.H.F.
arrays have relatively large dimensions, particularly on Band I; hence, the number
of elements must be restricted (say three on Band I and six on Band III). Special
attention would have to be paid to the mechanical strength of the supporting mast
and its attachment to the building. The effect of the additional stress on the chim-
ney stack or any other structural part must not be overlooked.
Mechanical problems are reduced when installing u.h.f. aerials. There are usually
Television Aerials and Systems 259

6-, 10-, 13- or 18-e1ement arrays apart from the special types previously mentioned.
A high gain aerial may be necessary in an area troubled by signal 'ghosting' or inter-
ference; the narrow angle of acceptance of this type of aerial prevents unwanted
signals developing any significant output. An aerial with fewer elements has a wider
acceptance angle, in which case it can be so positioned that the wanted signal
arrives at an angle to the line of maximum sensitivity. It may then be possible to
have the 'gap' between the lobes of the polar diagram to face the direction of the
unwanted signal. Such a method of installation may be useful when there is only
one major source of interference or reflection. In practice, the array is rotated
until the 'ghost' disappears and a satisfactory picture remains.
The principal source of signal reflection could often be judged from the
measurement of the distance between the main display and the 'ghost' on the TV
screen. The electron beam takes 5211S to scan one line of picture. If the screen is
40 cm wide (approximately 19 in tube), then the beam travels 40/52 =0.77 cm
per IJ.S. At the same time, the speed of propagation of the TV signal from the trans-
mitter is 300000 km/s (speed of light!) or 0.3 km/lJ.s. Equating these two velocities,
it can be seen that for approximately every 8 mm of 'ghost distance' the radio
waves must have travelled 300 m longer than the direct path. (Thus 1 in. or
25 mm, spacing gives approximately 1 km, or 0.6 mile, extra distance.)
A high gain aerial used in an area of good signal strength can develop an output
which might overload the receiver tuner and iJ. stages. The a.g.c. has only a limited
ability to reduce the receiver gain. In such circumstances an attenuator may have to
be fitted between the tuner aerial socket and the cable. The design principles of such
a circuit are discussed in section 13.7.
The positioning of the aerial array is important. Only in exceptionally favourable
circumstances can the so-called 'set-top' dipole be at all satisfactory. The picture
quality is likely to be affected by interference signals, or even by movement of
people in the room. This type of aerial is normally provided with small portable,
battery receivers. A loft-mounted aerial will be an improvement on this type, and it
may be a necessity if circumstances do not allow external installation. Signal reflec-
tion, or even signal absorption, can be caused by metal water tanks or pipework in
the vicinity of the aerial. Because of the obstruction introduced by the walls of
the building, the wanted signal may be attenuated, allowing interference signals to
predominate. A highly directional array might be needed to overcome those
problems.
An external aerial is the most satisfactory arrangement. The height of the mast
might have to be adjusted, especially in fringe areas, to minimise the effect of
obstructions in the signal path.
A very low loss cable has to be used where long runs are involved, especially
for the reception of the upper number channels (Band V).
Generally, colour receivers require a higher signal strength than their monochrome
counterparts. An installation which is only just satisfactory for black and white
reception must be improved before an even comparable picture quality in colour can
be obtained.
260 Television Principles and Practice

Satisfactory reception of Teletext requires a very efficient aerial installation. Any

tendency to ghosting can cause either a loss of characters or incorrect characters in
the displayed 'page'.
Aerial signal preamplifiers may sometimes be recommended to improve the
picture quality. It should be considered as the last measure in the attempt to boost
the signal. Modern TV tuners are of optimum design, and an indifferent preamplifier
may not warrant the expense. The best design array of maximum gain, followed by a
low loss cable, are the first considerations, since any signal gain is not accompanied
by additional noise. It must be remembered that electronic noise will be produced
by both the preamplifier and the tuner. For this reason an amplifier, if it has to be
used, ought to be installed close to the aerial on top of the mast. It then handles a
relatively noise-free signal and boosts the output prior to the inevitable attenuation
in the cable.
U.H.F. aerial arrays are made to receive a group of channels; these can be identi-
fied by group letters and sometimes colours: Group A (channels 21-34, coding red),
Group B (39-51, yellow), Group C/D (48-68, green), Group E (39-68, brown).

13.7 Aerial Attenuators

Attenuators may be necessary in exceptional circumstances; these reduce the
signal strength fed to the tuner to prevent overloading of the receiver circuits. Over-
loading would manifest itself as an overcontrasted picture, possibly with grey crnsh-
ing or white crnshing (see test cards). Often synchronisation is affected, giving rise
to 'wavy' verticals; sometimes instability can be induced in certain parts of the
receiver. Also, severe interference patterns might be displayed (cross-modulation).
Attenuators are basically resistive potential dividers, but their most important
property is that their effective impedance matches the impedance of the remainder
of the installation. Where a coaxial cable is used, this must be nominally 75 n.
There are two alternative circuits which can be used. Diagram 13.5(a) shows a T
network, and 13 .6(b) shows a 1T network. The values of the resistors required are
given by the following formulae

from to
aerial receiver

(a) (b)

Figure 13.5 Aerial signal attenuators: (a) T network; (b) 1T network

Television Aerials and Systems 261

A -1
T network Rl = R - - (2 required)
A +1
R _ 2A
2 -
A2 -1
A2 -1
1T network R =R--
1 2A

R2 =RA-+-1 (2 required)
A - 1
In the above expressions R is the impedance of the cable (typically 7S Q) and
A is the numerical attenuation ratio (VIN !VOUT)'
Since the attenuation needed is often expressed in dB, it must be converted into
a numerical ratio. For example, when attenuation is 6 dB,A = 2; for 20 dB,A = 10;
etc. Of course, the nearest preferred values of resistors are chosen after the calcula-
tions.

13.8 TV Signal Distribution Systems

It may be necessary to feed several receivers from a common aerial installation. The
signal is split into a number of outlets in such a way as to present to the cable a
matching impedance of, say, 7S Q.

75 n connected if
r-D-
R
: 1
I
receiver
unplugged

from
aerial

I
I ___ to other
L
outlets

Figure 13.6 Resistive signal splitter

The method of distribution shown in figure 13.6 ensures the necessary imped-
ance match, provided the equal value resistors, R, have been calculated as follows
n-1
R=--xZ
n+1
262 Television Principles and Practice

where n is the number of outlets, Z is the impedance of the cable (75 Q) and R is
chosen to the nearest preferred value. Should one of the receivers be disconnected
from the socket outlet, a 75 Q resistor ought to be connected in its place, to pre-
serve the original value of impedance; an open-ended outlet can give rise to signal
reflections. The resistors in the arrangement will dissipate some of the signal energy
which an aerial array may not be able to supply; therefore an amplifier will be
necessary to feed the system, if a satisfactory picture is to be maintained.
Extensive distribution systems have proportionally higher signal losses, including
large attenuation in the cable network itself. It can be more economical to convert
the received u.h.f. signal from the aerial into v.h.f. for distribution rurposes, as the
effect of signal loss in the cable is considerably reduced at the lower frequencies.
The receivers in this scheme can be equipped with either v.h.f. or integrated (v.h.f.
and u.h.f.) tuners. A block diagram of an arrangement based on this system is shown
in figure 13.7.

u.h.f. aerial

u.h.f. u.h.f./v.h.f.
amp. converter

splitter

to
individual
receivers

Figure 13.7 Block diagram of a distribution network with v.h.f. conversion

Alternatively, if the more popular u.h.f.-only tuners are used, the signal has to be
converted back from v.h.f. to the higher frequency by converter/amplifiers situated
ahead of final subsections of the system. The 'up-' or 'down-frequency' converters
are basically mixer/oscillator circuits which also include suitable amplifiers.
The systems described in this section can be expanded relatively easily, provided
the amount of amplification is sufficient at the 'strategic' points in the network.
14 Receiver Power Supplies

14.1 Power Supply Requirements

The power requirements of the various sections of TV receivers are at differing
voltage levels, depending on the function of the circuit to be supplied. In fully
transistorised as well as hybrid (valve and transistors) receivers relatively low voltages
are supplied to the circuits which handle small amplitude signals. Large amplitude
waveforms can only be accommodated if the amplifier is connected to a suitable
high voltage feed. Valves, by virtue of their principle of operation, need relatively
high anode (and screen where applicable) potentials, irrespective of their circuit
functions. The voltage levels used on the picture tube electrodes vary from low
voltage on the heater to the extra-high tension on the fmal anode.
The various supply lines are often distinguished from one another by abbrevia-
tions: Lt. (low tension), h.t. (high tension) or e.h.t. (extra-high tension). Since
several feeds at slightly different voltages may have to be provided within a given
category, an identifying number has to be included, say 1.t.1 or 1.t.2 etc.
The following list presents the various requirements of either a hybrid or a fully
transistorised TV receiver as applicable.
a.c. supply: valve heaters, including the picture tube (0.3 A and at least 6.3 V per
valve), degaussing circuit in colour receiver.
d.c. supply: l.t. (between 10 V and 40 V approximately) to the tuner, Lf. strip,
colour decoder, sound, video drive, transistorised time base oscillators, often 'solid
state' field output;
h.t. (between 50 V and 300 V approximately), all valve anode supplies, line output,
video output.
boosted h.t. (between 350 V and 1000 V approximately), field charging circuit in
valve oscillators, Al supplies to the picture tube, focus voltage to monochrome
tube.
e.h.t. (between 6 kVand 25 ky), final anode of the picture tube, focus voltage
to colour tube.
Some of the supply lines have to be stabilised to ensure a constant voltage feed
to a particular circuit.
Total consumption can be from approximately IS W for a small screen battery
263
264 Television Principles and Practice

portable to, say, 150 W for a large screen, hybrid, colour TV.
The design of the power supply circuitry must ensure safe operation of the
receiver under normal conditions as well as under fault conditions. To this end, a
number of protective features are incorporated which will be discussed later in this
chapter.

14.2 Heater Supplies

Television receivers employing valves are designed so that the heaters are all connec-
ted in series with one, another, each carrying a current of 0.3 A r .m.s. The basic
circuit is shown in figure 14.1; the power consumed by the circuit is constant and,
for a mains voltage of 240 V, is 240 x 0.3 =72 W. The reasons for diodes 01 and
02 shown in insets (i) and (li), respectively, are given later. The voltage requirement
of the heater chain is dependent on the number of heaters in series and also on the
p.d. across each heater. Consequently, a dropper resistor must be included in series
with the heater chain to absorb the remaining voltage. In practice, the dropper
resistor R 1 is in series with the thermistor R2 shown in figure 14.1. The name
'thermistor' is a contraction of thermally dependent resistor, and is a device, in the
case considered, whose resistance reduces with temperature. This fact is indicated
by the _to in association with the circuit symbol; alternatively, the abbreviation
n.t.c. (negative temperature coefficient or resistance) may be written against the
resistor. When the heater chain is first switched on, the temperature and resistance
of the heaters are low; in the absence of any method of limiting, the heater current
flow would initially be very large, introducing a possible risk of a fJIament failure.
At switch-on the high resistance of the thermistor restricts the heater current to a
small value. As the device warms up, its resistance falls and the current increases to
its normal value of 0.3 A.
Capacitors C1 and C2 in figure 14.1 are for decoupling purposes. This prevents
high frequency signals developed in one valve, or in a group of valves, from being
transmitted to another section of the receiver via the heater chain. The c.r.t. heater
is placed towards the chassis end of the arrangement so as to reduce the heater-to-
cathode potential; this minimises the risk of a heater-to-cathode breakdown of the
expensive tube.
In fully transistorised receivers the tube heater is usually supplied from a suitable
winding on the mains transformer or the line output transformer.
Hybrid TV receivers employ a relatively small number of valves, and the power
dissipated by the dropper resistor in figure 14.1 would be much greater than in an
all-valve receiver. The reason for this is that the smaller number of valves drop a
smaller voltage, while still taking a current of 0.3 A; consequently, the p.d. across
the dropper resistor would have to be increased. To reduce the power consumed by
the dropper resistor, diode 01 shown in inset (i) in figure 14.1 can be connected in
series with the heaters. In some cases this diode is shunted by a capacitor of a few
nanofarads for voltage surge limiting purposes. Owing to the rectifying action of 01,
the power consumed by the heater circuit is halved from the previously quoted
Receiver Power Supplies 265

valve heaters r-----------l

I I
I
I
I
1
1
I
1

I
I
I
I
I
1
I
---0 I
N I

I-------~~:--~~::
I
I
R.
1
I
I
I
smoothed
d.c. I
I
supply 1
I
I
I
I
I
I
1
I
L _________________ JI
(iii)

Figure 14.1 Valve heater supply circuit. Alternative (i) shows a diode dropper
(Dl). In (ii) diode D2 protects the heaters against effects ofDl short circuiting.
Arrangement (iii) can be used to extend the heater chain to provide a low voltage
d.c. supply

72 W to 36 W. Readers should note that there is no other advantage to be gained by

using a d.c. to supply the heater chain. However, the available d.c. can sometimes be
used to provide an Lt. feed to another section of the receiver, as shown in inset (iii),
figure l4.l. The voltage developed across R3 is smoothed by the circuit consisting
of C3 ,R 4 and C4 ; the resultant d.c. is then directed to the appropriate circuits.
In diode-fed heater chains there is a risk that the series diode may develop a short
circuit, with no apparent ill-effects until one or more heaters burn out. One method
used to protect against this type of fault is the use of an additional diode, D2,
connected across the supply as shown in inset (ii) in figure 14.1. Under normal
operation D2 is reverse biased and does not carry current; if Dl develops a short
circuit, diode D2 is forward biased in alternate half-cycles of the a.c. supply and
applies a short circuit to the heater supply. This either causes the cut-out to operate
(when fitted) or blows the main fuse. Alternatively, the d.c. supply shown in
266 Television Principles and Practice

figure 14.1 may be used to power either the sync. separator or the field oscillator;
shorted 01 will then make the picture unviewable. A similar arrangement, although
tapped at a higher voltage level in the heater chain, can supply the screen of the
field output valve. Again, a diode fault will cause field roll which could not be
corrected by the viewer.
To be strictly accurate, voltage and current measurements taken in diode-fed
valve heater chains should be made by means of either a moving iron or a hot wire
instrument, since these give a true r.m.s. indication. A mean reading instrument,
such as an Avometer, gives an indication which is lower than the true reading; an
r .m.s. current of 0.3 A gives an indication of about 0.19 A on the d.c. range of a
mean value reading instrument, and an r.m.s. voltage of 6.3 V gives a reading of
about 4 V on the d.c. voltage range of this type of instrument.

14.3 Unregulated D.C. Supplies

A method of obtaining a low voltage d.c. supply was shown in figure 14.1. The
polarity of the output can be positive with respect to the chassis in the circuit as
shown, or it can be reversed simply by reversing the connections of diode 01.
In the case of mains-only receivers, the a.c. mains can be rectified directly and
smoothed to give a d.c. supply of about 260-290 V. A typical half-wave circuit is
shown in figure 14.2. The input is taken from the mains via fIlter network C 1 , L 1 ,
C2 , which is employed to minimise the possibility of mains-borne interference from
affecting the operation of the receiver. In some circuits only a single capacitor
connected across the supply is used. Series resistor R 1 and thermistor R2 are inclu-
ded in figure 14.2 to reduce the effects of surges; rectifier diode 01 is protected
from the effects of mains voltage transients by capacitor C3 • Capacitor C4 acts as a
reservoir capacitor, the voltage across it being smoothed by fIlter L 2 , Cs to provide
the main h.t. supply, HTl. Additional h.t. supplies, HT2, HT3, HT4, are derived
from HTl via RC smoothing and decoupling circuits.
Electrolytic capacitors in power supplies, as in other parts of TV circuits, do not
provide satisfactory decoupling at radio frequencies. Adequate decoupling is obtained
by shunting them with polystyrene or polycarbonate capacitors of relatively low
capacitance-for example C6a in figure 14.2.
Low voltage supplies for transistor stages are frequently obtained from full-wave
bridge rectifier circuits of the type in figure 14.3. Resistor R 1 is included to reduce
the peak value of the charging current drawn by capacitor C 1 , and typically has a
value of a few ohms. The value of the smoothing capacitor C1 lies typically in the
range 1000-3000 fJF. The circuit shown can be designed to provide any desired
value of current or voltage. A negative output potential is obtained if the chassis
connection is transferred to point X and the output is taken from point Y. Alter-
natively, if the chassis connection is transferred to the centre point, S, of the
secondary winding, then a positive potential relative to the chassis is developed at
point X, and a negative potential relative to the chassis simultaneously appears at
point Y.
Receiver Power Supplies 267

L ~
C>-<rjo-l----l-_JY'tV\-H JY......,'"""'4_-_-- +270
H ......"""'............. h.t.l
V
I

rr
I
I
I
240 i
"-' I
I
I
I
0--0""0
N

200
~.......- - - -.......- - - - - h.t.2
+210V

220
1---+_ _.......~------h.t.3
+200V

5.6k
~_--h.t.4

Figure 14.2 Half-wave rectifier circuit which provides a number of [Link] in

the receiver

R,
x
I -.....- .......~A

a,c'~11
sup~

L-------__ to other circuits

Figure 14.3 Full-wave bridge rectifier circuit

268 Television Principles and Practice
01

'~II
Rl

supply
----__.p (+)
I--......

02

03
r - - -.......- - - - - N ( - )

Figure 14.4 Biphase rectifier circuit (Dt, D2) for the main positive Lt. rail and an
auxiliary half-wave circuit (03) to provide a negative supply

A bridge rectifier can also be connected directly across the mains to produce a
full-wave output which is easier to smooth than a half-wave supply. The individual
diodes are shunted by suitable surge diverting capacitors. Either the positive or the
negative d.c. terminal of the bridge is connected to chassis; consequently, the neutral
of the a.c. supply is isolated from the receiver earth. This introduces a safety prob-
lem when servicing such a receiver. The reader is urged to study the safety aspects
of TV maintenance discussed in section 14.11.
Other types of mains-derived Lt. power supplies are shown in figure 14.4.
Diodes Dl and D2 form a biphase rectifier circuit giving a positive supply, P,
relative to the chassis. A centre tap transformer is now necessary, but the rectified
output is equivalent to a full-wave arrangement. Diode D3 is a half-wave rectifier
circuit which provides a negative supply, N, relative to the chassis.

14.4 Simple StabDised Supply Circuits

The simplest form of circuit used to provide a regulated supply voltage is shown in
figure 14.5. It consists of a resistor,R 1 , in series with a Zener diode, the output
voltage being taken across the Zener diode. Capacitors C1 (electrolytic) and C2
Qow value, say ceramic type) are connected across the Zener diode to provide
additional smoothing and high frequency decoupling. The supply voltage, VI, is
always greater than the value of the Zener diode breakdown voltage, V2 ; the excess,
(VI - V2 ), is dropped across the current limiting resistQr,R 1 • The diode works in
its reverse breakdown mode with the cathode connected to the positive terminal of
the supply. The output voltage remains relatively constant; it is equal to the break-
down potential of the Zener diode, and has a value which can be selected out of a
wide range from a few volts upwards.
Receiver Power Supplies 269

Alternatively, in some circuits the Zener diode is replaced by the voltage depen-
dent resistor (v.d.r.; also known as a varistor) shown in inset (i) in figure 14.5. This
arrangement is used where high values of voltage are involved, as, for example, in
valve circuits needing a stabilised anode voltage supply (e.g. the feed from the
boosted [Link] to the field charging capacitor).
A disadvantage of a Zener diode stabiliser is that the output voltage fluctuates
if the load current is subject to considerable variations or if the diode temperature
changes. At the same time, the power losses in the series resistor (Rl in figure 14.5)
and in the diode itself tend to be considerable when relatively high current and/or
high voltage loads are supplied.
The tuning voltage for varicap diodes must be particularly stable, to prevent
tuner drift. Many manufacturers use an integrated circuit regulator which has only
two external connections, so that it replaces a Zener diode. The I.C. also requires a
limiting resistor and a suitable decoupling capacitor; this alternative is shown in
inset (ii), figure 14.5.

+0-----1 ......~_ .......- - - - - - - - - - - - o , +

I
I
I
I
I
unregulated I
I regulated
supply o
C,. supply output
V,
I
v2
I
I
I
I
I
~----~~~~---------~I-

Figure 14.5 Simple voltage stabiliser circuit. The main circuit shows a Zener diode
regulator; alternative (i) uses a VDR, and in (ii) a stabiliser I.C. replaces the diode

A type of arrangement frequently used to provide a stabilised low voltage supply

is based on the series regulator circuit. In these circuits a tran~istor of a suitable
power rating is connected in series with the load; its base voltage is controlled by
a Zener diode. In more complex regulators an error amplifier is used (for example,
see figure 14.8). A simple series regulator is shown in figure 14.6. The bias chain for
the transistor is formed by the Zener diode stabiliser circuit, R 1 , D 1 (described
previously). The load current is supplied by the series regulating transistor TR1; in
this circuit the diode current is reasonably constant, so that the base voltage remains
constant. The output voltage is approximately (Vz - 0.6 V) with a silicon transistor,
where Vz is the breakdown voltage of the Zener diode. High frequency decoupling
and low frequency smoothing is again provided by capacitors C1 and C2 , respec-
tively. The circuit in figure 14.6 is more economical than the one shown in figure
14.5, especially when relatively high and fluctuating load currents are supplied; the
270 Television Principles and Practice

TRl

~"
...
e-
::l

C2 ::l
0
"C
2l
co
:;
'"~

Figure 14.6 Simple series regulator circuit

series
regulator
transistor

t
error
unregulated input amp. regula ted output

J
voltage
reference ~ comparator

1
- feedback
and manual
adjustment
circuit

Figure 14.7 Block diagram of a series regulator with feedback

only major power dissipation is in the transistor across which the excess voltage,
(V2 - VI), must be dropped.
The circuit in figure 14.6 is an 'open-loop' type, because there is no provision to
ensure that the output voltage stays constant with load current fluctuations, tem-
perature variations and component tolerances. The block diagram in figure 14.7
shows the arrangement for a closed-loop regulator, in which the base current of the
series transistor is altered to compensate for any changes in the output voltage. A
sample of the regulator output voltage is fed to a comparator circuit which compares
it with a reference voltage from a Zener diode. The resulting difference in potential
is amplified and is applied to the base of the series regulator transistor to adjust its
conductivity. In the circuits discussed so far the output voltage was rigidly tied to
the breakdown voltage of the Zener diode. In the present arrangement the level of
the regulated potential can be manually preset within a fairly wide range if desired.
A practical circuit is shown in figure 14.8. The familiar Zener diode arrangement
of R2 and Dl supplies the voltage reference which maintains the emitter voltage of
 Receiver Power Supplies 271

I
I

,___ /L_
I
I

+
rectified and )r..-----<!o---------<!o-----~ stabilised
smoothed supply o-----.----<r--~< + 25 V supply
R,

Figure 14.8 Series transistor regulator circuit with feedback

TR2 constant at, say, 6 V. The potential divider formed by R 3 , R4 and VRI across
the output senses the value of the regulated voltage, and a suitable proportion of the
output voltage is fed to the base of TR2. This transistor compares its base voltage
with the emitter reference potential and produces an output to correct the error.
The output from TR2 is fed to the series regulator transistor, TRl, to adjust its
conductivity accordingly. For example, if the output voltage from the stabiliser
rises for some reason, the increased potential at the slider of VRI is communicated
to the base of TR2, which increases its conductivity; consequently, the collector
voltage of TR2 falls and reduces the base voltage of the regulator TRI. The reduced
base potential ofTRI causes the emitter voltage to fall (emitter follower action),
thus restoring the output to its normal value. Variable resistor VRI is used to preset
the required level of the output voltage.
The ability of the regulator to oppose any changes in its output voltage can be
used to perform 'electronic smoothing'. Residual mains ripple on the stabilised
supply line is fed viaRs, C 1 to the base ofTR2, where it appears as a fluctuating
output. The circuit takes a corrective action to improve considerably the overall
power supply ftltering. In effect, it looks as though the capacitance of C1 has been
multiplied by the regulator.
In some versions of the circuit in figure 14.8 transistor TRI is replaced by the
Darlington connected pair of transistors shown in inset (i) in the diagram. This con-
figuration causes the current gain of the combination to be approximately equal to
272 Television Principles and Practice

the product of the gains ofTR3 and TR4. Such an arrangement is used in high
current regulators, since a relatively small current in the control circuit can cope
with large power levels in the output. Occasionally, even a third transistor may be
connected in a similar manner.
There are many variations on the basic design of the circuit discussed above.
For example, the series transistor can supply an output from the collector instead
of the emitter. In this case a low value resistor is connected between the emitter
and the collector, to provide a supply voltage to the control circuit which is placed
on the output side of the regulator. The resistor also shares the load current with
the transistor, reducing the power rating of the device. The comparator and error
amplifier circuits can also be more complex than the basic version.
It is possible to connect the regulating transistor across the output instead of
being in series with the load. A diagram of one form of shunt voltage stabiliser is
given in figure 14.9. The output voltage is equal to the sum of the potential across
the Zener diode Dl and the base-emitter voltag:: ofTR2. Since both voltages
remain relatively constant, the output is stabilised. Should the supply voltage tend
to rise, the emitter potential of TR2 would increase, which, in turn, would also
raise the base voltage of TRI. The increased conduction of TRI causes more
current to be taken from the unregulated source, and the additional p.d. across Rl
restores the output to its stabilised level.

R,

unregulated
d.c. supply stabilised output

01

Figure 14.9 Shunt voltage stabiliser circuit

14.5 Switched-mode Power Supplies

A feature of the regulating circuits described in section 14.4 is that a large amount
of power is dissipated in the principal components. Greater economy and reduced
heating effects can be achieved if an electronic switch is used instead of a con-
tinuously conducting transistor. The switch conducts intermittently; it can either
be a transistor or a thyristor, but when the device is ON the voltage across it is
nearly zero and in its OFF state the cu"ent through it is zero. Therefore the power
dissipation in the electronic switch is very low. The output from the circuit is in the
form of pulses which can be fed to a transformer to be stepped either up or down
and give various voltage levels. Alternatively, the pulses pass through a conventional
smoothing ftlter, to produce a steady d.c. The mean value of the output voltage is
Receiver Power Supplies 273

controlled by the ON/OFF ratio, or mark/space ratio of the electronic switch. The
switching rate depends upon the actual type of circuit used. A thyristor-fed supply
operates at 50 Hz or at 100 Hz, which is governed by the half-wave or full-wave
rectified a.c. mains pulses. Other switched-mode power supplies are controlled by
the line time base oscillator, which runs at 15625 Hz. Finally, a separate power
supply oscillator may be used whose normal operating frequency is around 30 kHz.
The higher frequencies make subsequent smoothing easier, and should a transformer
be required at all, its physical size is very much less compared with one operating at
50 Hz. The reader will already have noticed how the line output transformer is used
to provide a number of supply lines to the receiver. This trend is justified by the fact
that the line output stage behaves like a switched-mode circuit except that in its
simplest form it does not provide stabilised outputs. However, a few receivers
utilise a relatively complex circuit where the line output transistor is also used as
a power supply regulator.

14.6 Thyristor Regulator Circuits

The thyristor, also known as an SCR (silicon controlled rectifier), is a semiconduc-
tor device having three electrodes-namely the anode, the cathode and the gate [see
figure 14.10(a)]. When the anode of the thyristor is negative with respect to its
cathode, the device is non-conducting and cannot be triggered into conduction by a
gate signal. In this respect it is similar to a reverse biased diode. It differs from a
conventional diode in that, when the gate voltage is zero, the thyristor remains non-
conducting even if the anode potential is positive with respect to the cathode. It is
triggered into conduction, however, when a positive potential (which may be either
a steady voltage or a pulse of a few microseconds duration) is applied to the gate
electrode at an instant of time when the anode is positive with respect to the cathode.
Once conducting, the thyristor remains in this state until its anode current is reduced
to zero (or to a very low value). When conducting, the forward voltage drop across
the device is about 1 V.
A somewhat similar device, known as a silicon controlled switch (SCS), is also
used in power supply circuits [see figure 14.l0(b)]. The reader will already have
come across an SCS in the field time base oscillator in chapter 9. In addition to the
cathode gate, this device has an anode gate; the latter is normally used for triggering
the SCS into conduction. A negative potential (steady d.c. or a pulse) with respect
to the anode must be applied to the anode gate to switch the device on. Once
switched on, an SCS behaves in a similar manner to a thyristor. An SCS is sometimes
used to provide trigger pulses for the regulator thyristor in full-wave circuits; it is
also found in power supply protection schemes (to be discussed in a later section).
Both the thyristor and the SCS have four layers of semiconductor materials in a
p-n-p-n arrangement. This structure may be considered to be equivalent to two
transistors-see figure 14.10(c). TRI is a p-n-p transistor whose base and collector
are connected to the n-p-n transistor, TR2. The interconnections arise from the
fact that the two middle p- and n-regions are shared between TRI and TR2. The
 3
274 Television Principles and Practice

f
anode

anode
anOde~
gate
(cathode) gate cathode
gate

cathode
cathode
(a) (b)
anode

(anode) gate O----.--f....

(cathode)
TR2 ,-+~---Q gate

cathode
(c)

Figure 14.10 Thyristor circuit symbols: (a) alternative thyristor (SCR) symbols;
(b) silicon-controlled switch (SCS); (c) two-transistor analogy of a p-n-p-n switch-
ing device

behaviour of the thyristor can now be studied by considering the operation of the
two-transistor analogy. With zero signal on either gate, the device does not conduct,
because of lack of base bias. Applying either a positive voltage to the cathode gate
or a negative one to the anode gate causes the appropriate transistor to conduct,
which, in turn, biases the other transistor through the interconnections between
them. The pair rapidly drive each other into saturation, and the voltage across the
thyristor falls to a very low value. The two-transistor arrangement is used in many
applications instead of a thyristor or an SCS. This was shown in the various field
time base oscillator circuits, but it can also be found in power supply regulators to
provide thyristor trigger pulses.
The voltage regulating effect of a thyristor circuit is shown in figure 14.11. The
input to the regulator may be a.c. from the mains, since the device can provide half-
wave rectification; alternatively, the thyristor may be supplied from a half-wave or
a full-wave rectifier. At time t 1 a trigger pulse is fed to the gate of the thyristor,
which causes the device to switch on to recharge a smoothing capacitor connected
across the output. The thyristor automatically turns off whenever the supply volt-
age falls below the capacitor voltage. The energy fed into the smoothing circuit is
proportional to the shaded area in figure 14.11; this, in turn, is an indication of the
value of the output voltage. The smoothing capacitor maintains the d.c. supply
between the pulses. If the trigger pulse is applied later during the input sinewave-
Receiver Power Supplies 275

mains waveform output voltage across reservoir capacitor

~/l,\
/ I \
I I
I '
\
i
"0
\
\
\ I
I
/

>
\ I

~ \ I I time
. \ I I
conduction \ I I
period of \ I I
thyristor \ /
\ / I
I \ / I
'\~/ I
I I
:J I I
C.
I I

l l
Q)
Cl
Cl
'S

Figure 14.11 Effect of thyristor switching on power supply output voltage. At

time t 1 , thyristor turns on early in the mains cycle-high output voltage; at t2
thyristor switches later-reduced output voltage

for example, at t2 -less energy will be transferred to the output and the regulated
voltage will fall correspondingly. Therefore a trigger pulse timing circuit is necessary
to fire the thyristor at a predetermined moment dependent upon the level of the
stabilised potential.
A typical circuit is shown in figure 14.12. The regulator thyristor is TH1,
which supplies the smoothing circuit comprising R 10 ,C4 and C 5 • Thyristor firing
pulses are obtained by discharging C2 through the diac DA1 (capacitor C2 is
charged via R4 from the a.c. line); R 7, Rs ,C3 and R9 are pulse limiting and shaping
components. The diac-a bidirectional breakdown diode-is a semiconductor device
which is non-cQnducting until the voltage across it reaches a certain value known as
the breakdown voltage. When this is reached, the diac is rapidly driven into conduc-
tion and the potential across it falls to a low value. In this respect, the diac is
another form of electronic switch; the type used in TV has a breakdown voltage of
about 30 V.
The voltage across C2 is thus allowed to build up to 30 V when DAI fires, and
the energy from the capacitor triggers the thyristor. The charging rate of C2 is con-
trolled by TRl, which shunts the capacitor. If the transistor conducts heavily, it
takes longer to reach the diac breakdown voltage and the thyristor trigger pulse
occurs later in the mains cycle, with consequent reduction in the output voltage.
Variable resistor VRI alters the bias on the base of TRI, which affects the stabilised
output in the manner described above. The feedback resistor:R s , senses the regula-
ted d.c. level and influences the conduction of TR1 accordingly.
A 'slow-start' circuit is often used in the thyristor regulator. A typical arrange-
276 Television Principles and Practice

d.c. feedback resistor

L/
~ o~--~~~~~----r-~--~H-------~~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

~__~---4--+-__-+--~~~~----~
N

Figure 14.12 Thyristor voltage regulator circuit. The thyristor and its trigger
pulse generator can be fed either directly from the mains or via a rectifier [D3-
alternative (0] . A slow start circuit-arrangement (ii)-is often used to reduce the
initial current taken from the supply

ment is shown in inset (ii) in figure 14.12. When the receiver is first switched on,
the large value filter capacitors have to be charged, which causes the output voltage
to be low. The regulator would try to trigger the thyristor early in the mains cycle
to increase the voltage; this, in turn, could overload the device, or blow the fuse or
even damage the capacitor itself. The slow-start circuit connected in the emitter of
TRI makes the transistor conduct heavily to retard the firing pulses and allows the
h.t. voltage to rise gradually. Transistor TR2 is connected across Zener diode Dl in
the emitter circuit of TRI. Normally TR2 is non-conducting, because of the lack of
base bias, but when the receiver is first switched on, C7 begins to charge. The charg-
ing current flows through the base-emitter junction of TR2 and the transistor con-
ducts. This' causes TR2 to bypass the Zener diode, Dl, causing the emitter voltage
of TRI to fall, so that this transistor conducts heavily; the net result is that the
voltage across C2 takes longer to reach the trigger voltage of DAI. Once C7 is fully
charged, the voltage across Dl is restored to its steady value and the entire regulator
Receiver Power Supplies 277

operates normally. When the receiver is switched off, C7 discharges viaR 13 and D2
ready for the next slow start.
In the circuit in figure 14.12 the trigger pulse generator operates from the mains,
since the thyristor itself acts as a rectifier. In some designs a diode [D3 in inset (i)] is
in series with the thyristor, to share the reverse voltage when TH2 is turned off.
The switching action of the thyristor can cause severe interference; this is ftltered
out by the action of a low value inductor, L 1 , and capacitor, C1 •
The overvoltage protection circuit associated with TH2 will be discussed in
section 14.10.

14.7 Series Transistor Switched-mode Regulator Circuits

A transistor can be connected in series with the load in a manner similar to the
thyristor. The input to the regulator is a rectified and partly smoothed d.c. which
is interrupted, or 'chopped', by the switching transistor, as explained in section
14.5. Because of the action of the device, the arrangement is sometimes called a
chopper circuit.
A block diagram of a power supply based on the chopper principle is shown in
figure 14.13. The supply is obtained via a tap on the autotransformer, Tl, and is
rectified by diode D1 and smoothed by capacitor C2 • Transistor TR1 operates as a
switch, its base current being switched between zero and a high value which is
sufficient to saturate the transistor. When the base current is zero, the 'switch'is
OPEN or OFF, and no energy is supplied to the load. When base current is applied,

reference
V amplifier voltage
and
comparator

L
time base

to field
a.c.
D2 time base
mains T1 2.2/J and sound
section

line time base

pulses

Figure 14.13 Simplified diagram of a switched-mode power supply regulator-

using a series transistor (so called chopper circuit). Note the values of [Link]
capacitors: C2 (ripple 50 Hz), C3 and C4 (ripple at line frequency)
278 Television Principles and Practice

it drives the transistor into saturation so that the collector-emitter voltage falls to
a very low value (typically 200 mY), and the switch is CLOSED or ON. The
chopper operates at line time base frequency; thus its output is in the form of
pulses at 15 625 Hz which are fed to the smoothing circuit consisting of inductor
Ll and capacitor C3 • The reader will notice the difference in capacitance value
between smoothing capacitor C2 (at 50 Hz) and C3 (at line frequency). When the
chopper is OFF, energy stored in the inductor discharges into the load to maintain
the load voltage. Diode D2 provides a return path for the inductive current during
this period of operation.
The base switching current for the chopper is provided by the monos table multi-
vibrator coupled to the transistor via transformer T2. To regulate the output volt-
age, the ratio between the ON time and the OFF time of transistor TRI is auto-
matically adjusted; that is, the mark/space ratio of the output of the multivibrator
is made variable. A monostable multivibrator is a type of switching circuit which has
to be triggered by an external pulse for the output to change its state from OFF to
ON. The ON state is then maintained for a period of time governed by the circuit
internal time constant, after which the output returns to OFF again (in which it
remains until retriggered).
In the circuit in figure 14.13 the monostable ON/OFF ratio is controlled by
altering the discharge rate of a capacitor. In tum, this discharge rate is regulated by
voltage Vat the output of an amplifier/comparator circuit. The latter compares a
reference voltage derived from a Zener diode with the chopper stabilised output.
The resultant effect is that the ON time of TRI is increased by the multivibrator in
order to raise the regulator output voltage.
The circuit also includes a slow-start arrangement plus overcurrent and over-
voltage protection; the latter features are discussed later in this chapter.

14.8 Shunt Switched-mode Power Supplies

An alternative arrangement to the series regulator described in section 14.7 is a
switched-mode transistor which, together with a suitable transformer, is connected
across the rectified mains supply. The transformer presents an inductive load where
energy is stored when the switch is ON. The energy is transferred to some of the
receiver circuits during the ON periods and to others during _the OFF periods; the
latter is due to the storage effect of the transformer. Rectifiers are connected to the
transformer windings, their polarities being arranged so that conduction takes place
during either, the positive or the negative half-cycles of the transformer e.m.f. The
system resembles the action of a line output stage; indeed, many designers combine
the functions of a switched-mode power supply with the line time base of their
receivers. The reader will recall from section 10.5 that the switching waveform of a
conventional line output transistor has to be carefully timed, not only to prevent
overloading the device, but also to ensure a linear picture. The transistor used in the
switched mode regulator is fed with a line frequency waveform, but the mark/space
ratio is variable according to the generated voltage levels. Satisfactory operation of
Receiver Power Supplies 279

D2
r-_--IM--.---_~-----I.t. supplies
fuse (via L.O.P.T.)
o E::J 0

a.c.
mains to L.O.P.T.

o 0

L 3 *+ _ _ _ _ _}
~~~~n~o~~ises,
connections to
D2 and D3
from
line
driver

Figure 14.14 Shunt switched-mode power supply with the associated line output
stage (GEC). The receiver is mains/battery-operated, but the regulator circuit (TRl)
is not used during battery operation

a combined arrangement can be achieved by a simple isolating diode which isolates

the regulator from the rest of the line output stage during the conduction of the
efficiency diode.
The circuit in figure 14.14 shows the principle of operation of one type of shunt
switched-mode regulator. The regulator transistor is TR1, whose collector is fed
from the rectified and smoothed mains supply via current limiting resistor R 7 . The
transistor is switched on by line flyback pulses fed to its base from winding LIon
the line output transformer. Therefore, during the regulator ON state, current is
fed to the line output stage via TRI and D4. The energy thus stored is supplied into
 h.t.l T2
from L.O.P.T.
h.t.l flyback pulse
rectifier

common ~ stage

low
fllQrK./:;~i:lC~

~ I ratio outl
I.C.l
r I
TDA2640
--J I II
driver
amp.
start overvoltage
attempt and lC'w
input voltage protection
counter U I II
overcu rrent
comparator common
protection

Rs

Figure 14.15 Shunt switched-mode power supply controlled by an I.C. The regulator (TRl) operates in a manner similar to a
line output transistor. Note: Regulator 'COMMON' is not connected directly to chassis (it is decoupled Via Cs ). Instead regulator
[Link] is earthed. Output rails (I.t.1, h.L2) are conventionally arranged. Therefore servicing this type of circuit must be done with
care. (Mullard)
Receiver Power Supplies 281

the various receiver circuits, including 1.t. rails, during line scan. When the regulator
transistor is switched off, isolating diode D4 is reverse biased by the back [Link].
from the transformer winding, L 3 • (The typical line output arrangement consisting
of the line output transistor, TR2, efficiency diode, D6, and flyback tuning capaci-
tor, Cs , should be readily recognisable.) Although TRI is switched on by flyback
pulses, the actual period of conduction during the 12 IlS time interval is governed
by the charge/discharge cycle associated with C 7 and VRI. This adjusts itself auto-
matically, since the magnitude of the flyback pulse is a measure of the state of
conduction of the line output transistor just prior to the flyback. In turn, the
current at the end of line scan depends on how much of the initial energy has been
dissipated by the receiver power requirements. A large amplitude pulse caused by
excessive current flow charges capacitor C 7 in the base circuit of TRI to a higher
voltage. The increased capacitor potential maintains D5 reverse biased for slightly
longer when the next flyback pulse arrives, and the point of switching TRion is
delayed; consequently less energy is fed into the line output stage. The rate of dis-
charging C 7 is governed by the setting of resistor VRI. Since the amplitude of the
flyback is also an indication of the receiver e.h.t. and the resultant picture width,
VRI can be set to provide correct picture geometry (see chapter 13). At the same
time, the regulator controls the e.h.t., h.t. and 1.t. supplies. The circuit action
resembles that of the width stabiliser found in a valve line output stage (section
10.6).
The regulator transistor TRI can operate only when there are flyback pulses,
which means that the line oscillator, the driver and the output transistor must be
energised independently. This is achieved by means of the starting diode D2, which
supplies a reduced voltage to the circuits concerned until the line transformer
gradually develops its correct output.
In the circuit shown in figure 14.14 the regulator circuit is only operative when
the receiver is used on the a.c. mains. During battery operation stabilisation is
achieved by various Zener diode stabilisers (not shown in this diagram). The level
of the 12 V supply is raised by the boost circuit formed by D3 and Cs . This circuit
is connected across one winding on the line output transformer; boost diode D3
rectifies during the scan period and the resultant voltage developed across Cs is
added to the battery potential. It will be noted that in transistorised receivers the
level of boost voltage is far less than in their valve counterparts.
A different approach to the design of a switched mode power supply is shown
in figure 14.15; this circuit is used in a colour receiver whose power requirements
are greater than those of a monochrome set. The a.c. mains supply is rectified by a
full-wave bridge rectifier and partly smoothed (neither shown in the diagram) to
become h.t.l. This is fed to the regulator transistor TRI via transformer T2. Tran-
sistor TRI is switched ON and OFF by the signal applied to its base from the driver
stage, whose action is, in turn, controlled by the integrated circuit, I.e.1. Alternate
switching ofTRl converts the d.c. input from h.t.1 into a.c. by the action of the
transformer. The two outputs are rectified and smoothed to provide 1.t.1 and h.t.2,
which between them supply all the low voltage and the high voltage requirements
of the receiver.
282 Television Principles and Practice

The circuitry associated with the regulator transistor closely resembles the line
output stage. The reader is asked to turn to section 10.5 to recall the operational
problems connected with high voltage power transistors used for switching an
inductive load. The driver amplifier stage, represented by a block in figure 14.15,
is very similar to the corresponding arrangement in the transistorised line time base
(see section 10.3). The driver output is coupled to the base of TRI by transformer
Tl and the pulse speed-up network R 10, C lO • When the regulator transistor is
driven ON, the collector current is limited by the choke, L 1 , in its collector circuit.
When TRI is switched OFF by the base waveform, the entire output circuit would
tend to burst into self-oscillations (ringing) as well as to induce a large back e.m.f.
at the collector. The rise in collector voltage is prevented by the protection network
C 11 , D4 and R 12 connected across the transistor; diode D3 and choke L 1 also
assist with voltage limiting. Ringing at TRI collector is damped by the respective
resistors R 12 and R 14 in the protection circuit.
The control circuit of figure 14.15 is now discussed. The regulator switching
waveform is generated by the oscillator inside I.C.1 in figure 14.15. The free-
running frequency is governed by the values of C3 and R2 connected between
pins 3, 4, 5. This design ensures that the system is self-starting, unlike in other
circuits, where the line oscillator must first be made operative. However, once the
line time base becomes fully operational, the power supply oscillator is synchronised
with the line frequency by means of the flyback pulses applied to pin 2. The mark/
space ratio of the oscillator waveform is modified in the modulator block according
to the receiver requirements, as explained in section 14.5. The suitably 'modulated'
square wave is amplified by the output stage before it is applied to the driver ampli-
fier at pin 6.
The mark/space ratio is affected principally by two features: the power demand
placed on the regulator circuit of TRl, and the various protection arrangements
incorporated in the overall design.
The control of power requirements is now considered. The amplitude of the
pulses induced in the windings of the regulator transformer depends upon the
magnitude of the current flow in the windings. Therefore a separate coil, L s , is
wound on T2 to feed back a signal proportional to the regulator output. The initial
pulses are rectified and smoothed, and a fraction of the resultant voltage (given by
the potential divider R7 ,R s , VR2) is then fed back to I.C.1 at pin 10. The com-
parator circuit compares the actual output with a reference voltage developed by
Zener diode D2 (applied at pin 9). An error signal is developed by the comparator
and is communicated to the mark/space ratio modulator block. Residual mains
hum can also be reduced by the action of the comparator, which now affords
electronic smoothing. The level of a.c. ripple on h.t.! rail is sampled by Cs , R6 and
fed back to pin 10, where it would appear as an unwanted supply voltage variation
to be corrected accordingly. The two resistor-capacitor networks,R 9 , C9 in the
sampling circuit and R 4 , C7 between pins 9 and 10, govern the speed with which
the regulator responds to any changes in demand while preventing the whole system
becoming unstable.
The arrangement in figure 14.15 includes a slow-start feature, the need for
Receiver Power Supplies 283

which was explained in section 14.6. Capacitor C6 at pin 13 charges slowly, to

affect the mark/space ratio via the slow-start block inside the I.C. After the
receiver has been switched off, C6 discharges via R3 ready for the next slow-start.
A stabilised power supply to I.C.1 (at pin 1) is obtained from h.t.1 via the
Zener regulator R 1, Dl with its associated smoothing/decoupling capacitors C1 , Cz ·
The comparator reference voltage is also derived from this point and fed to pin 9
viaRs.
The integrated circuit includes a very comprehensive power supply protection
scheme, which will be discussed in the next section.
The power supply regulator COMMON connection is isolated from the receiver
chassis by capacitor C s , since it is the positive terminal of h.t.1 supply line (which
is at chassis potential). If a bridge rectifier is used, then the chassis is at half mains
voltage. Care must be exercised in servicing circuits of this type to prevent a short
circuit between the regulator COMMON and the receiver CHASSIS.

14.9 Overcurrent Protection

The TV receiver must be protected against any harmful effects caused by internal
faults. These could render the receiver unsafe (risk of fire, X-rays, electric shock)
or cause the original breakdown to induce more damage, which would be costly to
repair. Various methods of electrical protection are considered below. The general
aspects of safety are discussed in section 14.10.
Overcurrent protection can be provided simply by means of correctly rated
anti-surge fuses, or by means of thermal cut-outs, drop-offresistors and fusible
resistors, or by means of magnetically operated circuit breakers. Fuses provide
protection for normal operating conditions, and also permit switch-on surges to
pass without blowing. In a thermal cut-out the heat produced by the current
causes a bimetallic strip to bend, and if an excessive current is flowing, it opens
a set of contacts which must be reset manually. In the case of a drop-off resistor
(see R 1 in figure 14.14) the heat developed in the resistor is transmitted to a
soldered joint; when the current rating of the device is exceeded, the solder melts
and the circuit is broken. Electromagnetic circuit breakers are a form of relay
which opens a set of contacts when the current through the coil rises above a cer-
tain value; the circuit breaker has to be reset manually.
The above-mentioned protective devices can be installed in the main a.c. supply
line, or in the various 1.t. and h.t. rails. They can be used even in individual sub-
circuits-for example, in the screen feed to the line output valve or in the line out-
put transistor, etc. These methods of protection have one major disadvantage-their
slow speed of operation and lack of sensitivity may not be acceptable in many semi-
conductor circuits. Semiconductor p-n junctions can be easily damaged before
protection starts to operate; on the other hand, very sensitive devices may not allow
normal supply surges to occur without tripping out. A number of receivers use
electronic overcurrent protection, which can be incorporated into a power supply
regulator circuit.
The diagram in figure 14.16 shows a constant current regulator. Zener diode Dl
284 Television Principles and Practice

output

Figure 14.16 Constant current regulator

is connected across the series combination of the base--emitter junction of the

regulator transistor TRI and the two resistors R 1 , VRI in the emitter feed. The
emitter current produces a voltage drop across R 1 and VRI. However, the emitter
current, IE, is limited to the value which causes the potential difference across the
two resistors to be equal to the Zener diode breakdown voltage, Vz (neglecting the
effect of VBE ). That is
h = VZ /(R 1 + VRl)
This type of regulator can be used to develop a constant output voltage across a
fixed load resistor. The arrangement may be found in a stabiliser supplying the Al
controls of a colour receiver.
In other circuits the overcurrent is sensed by measuring the p.d. produced by
the current when it flows in a low value resistor. The resultant voltage biases on a
protection transistor, which, in turn, switches off, or at least reduces the conduction
of the regulator device directly or indirectly. An example of this arrangement is
given in figure 14.15; the I.C. includes overcurrent protection by sensing the p.d.
developed across R 13 (I n) in the emitter of the regulator transistor, TRI. The
control voltage is fed via preset VRI to I.C.1 at pin 12. The internal protection
circuit reduces the mark/space ratio to zero, and the drive waveform is removed
for a period given by the discharge of C6 at pin 13. The regulator automatically
starts up again via the slow-start circuit; the procedure may be repeated a number
of times, as governed by the restart counter and its external capacitor, C4 , at pin
15. If the overload persists, the regulator locks out. A low value capacitor, C 16 , is
connected between pins 11 and 12 to prevent protection operating in the event of
very short duration surges, which are thus diverted from the control circuit.
Another form of overcurrent protection was shown in figure 10.8, section 10.5.

14.10 Overvoltage Protection

Overvoltage protection is necessary, since an excessive voltage may cause damage,
particularly to the line output stage or to integrated circuits and other semiconduc-
tor devices. In addition. excessive voltages will be reflected in an increased e.h.t.;
for example, a rise of 5 V in theh.t. supply can 'boost' the e.h.t. by 600 V. E.H.T.
Receiver Power Supplies 285

levels above the design figure for a given receiver cause frequent tube flashovers
leading to a possible, and often unaccountable, breakdown of various transistors.
Excessive e.h.t: also increases the risk of X-ray radiation from the tube, especially
in colour TV, with its already high value of anode voltage (say 25 kV or more). The
tube heater voltage can also rise if it is fed from a faulty supply line, thus reducing
the life of the c.r.t.
One method of protecting against an overvoltage is to employ a network which
applies a short circuit to the power supply when the voltage exceeds a predetermined
value. That is, an electronic crowbar is applied across the supply terminals; this
usually blows the fuse or operates a cut-out (this form of protection is known as
crowbar protection). A simple circuit of this type of arrangement is shown in
figure 14.17. The breakdown voltage of Zener diode D 1 is chosen so that it does

from mains
f-_---<> h.t.
rectifier

C1

Figure 14.17 Simple electronic crowbar protection

not normally conduct; neither does thyristor THI. If the regulator output voltage
becomes excessive, say owing to a fault in the device or its control circuit, diode
Dl conducts and applies a positive trigger voltage to THI. The thyristor switches
on and causes a large current to flow from the power supply, and the receiver cut-
out operates. To prevent 'nuisance' tripping, capacitor C 1 absorbs momentary over-
voltages produced by tube flashovers.
Another form of overvoltage protection was shown in figure 14.12. Thyristor
TH2 is normally non-conducting. If the stabilised output rises above its nominal
value, the voltage developed across VR2 increases above the breakdown potential
of diac DA2 and the thyristor is triggered into conduction. TH2 short circuits the
regulator timing capacitor C2 , and removes the firing pulses from the regulator
thyristor, THI.
The I.C. controlled power supply in figure 14.15 includes an alternative method
of overvoltage protection. The magnitude of the line flyback pulses is detected by
the flyback pulse rectifier and fed to pin 8 of I.C.l. The overvoltage protection
block acts as a comparator which compares the potential at pin 8 with the reference
voltage at pin 9. At a predetermined level the protection circuit reduces the drive
waveform to zero and the sequence of events is the same as that already described
in the operation of the overcurrent protection for this circuit. The .dvantage of
286 Television Principles and Practice

using the line pulses to operate this system (instead of the more obvious source as
supplied to pin 10) lies in the fact that the protection can also guard against a
number offaults in the line output stage. I.C.l includes a form of back-up protec-
tion against a loss of control voltage at pin 10. Should this happen, there would
be a tendency for the comparator to induce maximum mark/space ratio and a
danger of overvoltage. The 'low feedback voltage protection' block overrides the
comparator and maintains a very low mark/space ratio in the drive waveform.
Some manufacturers use an overvoltage sensing circuit to shut down the line
oscillator to protect the most vulnerable section of the receiver-namely the line
output stage.

14.11 Power Supply Fault-finding and Safety Considerations

Statistically it appears that power supply faults are the most common source of
breakdown. It is essential to establish which sections, if any, of the receiver are
functioning, to see whether the 'dead' circuits share their supply rails. Where
applicable, valve and tube heaters should be energised .. Supplies fed from the line
output stage would be immediately affected in the event of a fault in the line time
base. Various protection features can also affect one circuit in the event of a fault
in another.
Fault-finding in unregulated power supplies may require only the use of a volt-
meter to check for the presence of a.c. input and of the various d.c. rail voltages.
Usual continuity checks will then confirm the position of the faulty component.
Regulated power supplies which employ feedback present additional problems.
The effect of feedback on the behaviour of the stabiliser circuit should be checked
in order to decide whether the fault is in the control circuit or the regulating device
itself. Varying the appropriate adjustment will disclose whether the control sub-
circuit responds in the manner expected.
Switched-mode power supplies can be treated in a similar manner to line time
base circuits. The presence of waveforms can be detected with the aid of an oscillo-
scope. Unless a low capacitance probe is used, the reader should not test for the
drive waveform on the base of the regulator transistor; the input capacitance of the
oscilloscope can affect the timing of the switching waveform and cause damage to
the output transistor. Again, the effect of feedback on the behaviour of the circuit
can be observed. It must not be overlooked that a number of regulating circuits
require an auxiliary supply to energise the control circuit.
Any protective features of the power supply circuit must not be immobilised for
whatever reason, as damage to many components may result.
When one carries out repairs on power supplies and associated circuits, the
receiver must first be switched off and unplugged from its mains socket. It is always
advisable to discharge the electrolytic capacitors via a resistor of about 1 kn.
Before actual fault-finding anywhere in the receiver, it must be remembered that
the chassis may be live at full mains potential, especially in the receivers which do
not have mains transformers in the power supplies. It is necessary to ensure that the
Receiver Power Supplies 287
mains connections are correct to both the receiver and the plug. The standard colour
code used in flexible supply leads is: brown for the live and blue for the neutral.
Where a bridge rectifier is used in the mains power supply without an isolating trans-
former, the chassis is always at half the mains potential with respect to the 'true'
earth. Any test equipment (oscilloscope, signal generator, etc.) used when the set is
connected to the mains should be supplied via an isolating transformer. Otherwise,
the test equipment must not be earthed to avoid possible damage to the circuits-
but the risk of shock still remains!
Workbenches in TV repair workshops should also be supplied via isolating trans-
formers, whose rating ought to be in the region of 500 VA per bench. This allows
the TV receiver and other test equipment to be connected without causing any
reduction in the supply voltage to the receiver. If the transformer is inadequately
rated, it may also produce erratic operation of a thyristor regulator in a TV set.
The design and servicing of TV receivers in the UK must conform, by law, to
strict safety standards. The quality of manufacture is governed by BS 415: 1972,
and the certificate of approval is given under the B.E.A.B. scheme (British Electro-
technical Approvals Board) to the appropriate make of receiver. The conditions of
the approval certificate must not be altered in the process of receiver servicing. This
applies to the important safety components, which have passed stringent tests
before being adopted in the manufacture. These parts are distinguished from others
in service diagrams by suitable markings. The two common methods of identifica-
tion are shown in figure 14.l8. In (a) the shading over R, and the warning sign in

(a) (b)

Figure 14.18 Identification markings for safety components. In (a) R, is a safety

component, and in (b) C, is so designated. (Either symbol may be used in circuit
diagrams)

the triangle denote that the resistor is a safety component; in (b) the letter S in a
'diamond' outline means that C, is a special component. Any replacement must
comply with the same specifications as the original component; the method of
fixing may also be critical-for example, it may be necessary to ensure correct
clearances or adequate ventilation or to use low melting point solder to replace a
drop-off resistor.
288 Television Principles and Practice

14.12 Mains/Battery Operation

Mains/battery receivers are capable of operating in one of several modes, including:
mains operation, battery operation and recharging.
Receivers either may have an internal battery or may be supplied from an extern-
ally connected source, such as a 12 V car battery. If an internal battery is used, care
should be taken in maintaining the electrolyte level in the cells. Also, if the receiver
is inadequately ventilated, there is a danger of an explosion due to the accumulation
of gases if a lead-acid type uf battery is used.
Some receivers include facilities for /loating charge or trickle charge when the
receiver is connected to the mains supply. A supply circuit with this feature is
shown in figure 14.19. Diode D3 is connected between the positive terminal and
chassis, to protect against an accidental reversal of the battery connections. In this
event, the diode would become a short circuit across the supply and cause the fuse
to blow.

battery

L~J II
a.c. mains!
L, +

N~ '-+.....---'
I
I
I
I
I
I
ION/OFF
: switch
I
I c,
I
I
I
I
I
I _____ _
L..

Figure 14.19 A built-in battery charging circuit in a mains/battery-operated

receiver. Note: Diode D3 guards against the effects of incorrect polarity of the
battery. This type of protection applies to most battery-operated receivers
Index

Al electrodes 211 Aspect ratio 14,246

Additive mixing 3 Attenuator, aerial 260-1
Aerial, acceptance angle 251, 252 Automatic chrominance control (a.c.c.)
amplifier 260 112,113,130
attenuator 260-1 Automatic phase control (a.p.c.) 130
bandwidth 254
cable 257-8 (B'-Y') signal see Colour difference
dipole 250-4 signal
directivity 250 Back porch 25-6
front-to-back ratio 254 Bandwidth see also Sidebands
gron 254,255,256 aerial 254
group coding 260 chrominance 18,85
impedance 254 colour difference signal 19, 85
isolating circuit 46-7 i.f. 58,60,61 i 246
log-periodic 256 luminance 85
matching 250 TV channel 17
polar diagram 251,252,253, video 12-17,79,246
255,259 Beam limiting 90,96-9
polarisation 251, 252 Beat pattern see Interference signal
splitter 261-2 Bistable 134
Yagi array 253,255 Black level, clamp 80-3,124-5,247,
A.F.C. (automatic frequency control) see also D.C. level restoration
50,69-72 in video waveform 25-6, 78,
A.G.C. (automatic gron control), 80-1
delayed 107-8 Blanking, burst 114
fault-finding 108-9 flyback 90,95-6,170,176,195
forward 106-7 Boost h.t. 194,200,263,281, see also
gated (keyed) 77, 104-5 Efficiency diode
Lf. 107-8 Brightness control 82,83,90,91-3
mean level 101-2 Burst signal see also Sub carrier
peak level 102-4 regeneration
reverse 106-7 amplifier 126,127
tuner 48,49,107-8 blanking 113,114
Alignment see I.F. amplifiers discriminator 127, 128
Aquadag coating 212,234 formation of 23,32
289
290 Index

Burst signal (cont). Colour control see Saturation

gating 126,127 Colour difference drive 89,124-5
phase 23 Colour difference signal see also
Chrominance signal
Cable, aerial 257-8 amplifiers 124-5
Camera, arrangements in colour TV (B'-Y') 10,29,30
11-12 bandwidth 19
tube 6-7 demodulation 119-21,113
Casco de amplifier 64 de-weighting 89,122
Cathode modulation of c.r.t. 74, 88, formation of 10, 11
122-3 (G'-Y') 11,29,30,89,121
Cathode ray tube (c.r.t.) 42-3 (R'-Y') 10,28,29
colour 210,211 waveforms see Colour bar signal
convergence see Convergence weighting of 22,32,33
correction Colour killer 87,113,132,137
delta gun 213-16 Colour tubes see Cathode ray tube
electrodes 211 Colours, complementary 4
grey scale see Grey scale tracking mixing of 2-5
in-line gun 216-17 primary 4
linearity (of picture) 212, 213, Contrast control 93-5, 102, 103, 114
s"ee also Field time base, Line Convergence correction 212,215-16,
time base 218-19
monochrome 210, 211 adjustments 236, 239-40,
pincushion distortion see Pin- 240-2
cushion correction circuit diagrams 219-25
precision in-line (PIL) 217,238, delta gun tube 215, 219-23,
244 240-2
purity 212,237-8 differential 219,222,241
replacement of 213,234 dynamic 216,219-25,240-2,
safety precautions 233-4 247
shadowmask 214 field 219-23,224-5,240-1
Trinitron 217-18 'in-line'tube 223-5,239,242
voltages 79,125,211 lateral 216,221
waveforms see Colour bar signal line 221-3,224,240-1
Ceramic filter 146, 149 matrixed 219,241
Chrominance signal (chroma), amplifiers precision in-line (PIL) tube 240
112-19 radial 216,221
bandwidth 19,112 static 216,220,239-40,241,
delay line 116-19 245,247
formation of 20-2,32-3 waveforms 218-19
modulation of 120 Cross-hatch signal 219,245,247
phase errors 22-3,114-16 Cross-over distortion 150-1, 175
phasor diagram 31,32,115 Crowbar circuit 285
U and V signals 21-2,31,32, Crystal oscillator 129
116
waveforms see Colour bar signal Darlington pair 271
Clamp circuit see Black level clamp D.C. amplifiers 79-80,151-2
Colour bar signal 27,245,248 D.C. level see also Black level
waveforms 27,28-33,122-3 of convergence waveform 223
Index 291

restoration of 80-1,90 of 194,201-4

of video waveform 78-9 overwind 201,202,203
Decoder, colour 40-1,110-42 rectifiers 201, 202
De-emphasis 147-8,150 tuning 203-4
Degaussing, automatic 232-3 Electronic smoothing 271
manual 237 Encoder, colour 20-2
Delay line, chrominance 116-19 Equalising pulses 34,35,160
luminance 85-6, 248
Demodulator see Detector Fault-finding, in a.f.c. 72
Desaturated colours 5,30,123-4 a.g.c. 108-9
Detector see also Discriminator colour decoder 140-2,245,246,
a.f.c. 70-2 247
a.g.c. 102-3 d.c. amplifiers 108
chrominance 112 field time base 179-80
coincidence 148-50 I.C.s 142-3
colour difference 119-21 Lf. strip 65
diode 74-5 line time base 206-9
Foster-Seeley 70, 71 luminance rtages 86
Lf. 50 power supplies 286
phase 128,182-5 sync. separator 157,178-9,245
quadrature 77, 148-50 tuner 57
ratio 146-7 video amplifiers 108
slope 148 Field time base 42,164-70,171-6,
sound 146-50 176-80
suppressed carrier 20, 119-21 amplifiers 168,175-6, 176-8
synchronous 75-9, 119-21 fault-finding 179-80
video 39,73-9,98 linearity 161-4, 170, 175, 180
Diac 275 oscillators 165-7,168,171-4
Dichroic mirrors (prisms) 11 switched mode 178
Differentiating circuits 160-1 synchronisation 34-5, 159-60,
Diode, attenuator 113 168,173
detector 74-5 waveforms 162,169-70,174,
dropper 265 175
mixer 50,74 Fifth harmonic tuning 203
modulator 230-1 Flashover, tube 198,233
rectifier 266-8 Flyback, blanking 90,95-6, 170, 176
Diode split winding 202 tuning see line time base
Diplexer 257 Flywheel synchronisation 181-5
Dipole 250-4 Frame see Field time base
Discriminator, burst 127-8 Frame frequency 15
line sync. pulse 181 Frequency interleaving 18-19
Driveramplifier 99,119,151,175, Front porch 25-6
190
(G'-Y') signal see Colour difference
Efficiency diode see also Boost h.t. signal
semiconductor 193,195,197, Gating see A.G.C., Burst
204 Ghost images 248,250,254,258,259
valve 200 Grid modulation of [Link]. 89,125
E.H.T.(extra high tension), generation Grey scale tracking 90,231-2,243-4
292 Index

Hanover bars 115,119,133,142,247 output transformer 194-5,

Highlights, picture 243 199-201
Horizontal deflectiDn see line time stabilisation 198-9
base synchronisation 34-5, 161,
Hue 1,5 181-5
waveforms 162,185-6,190
I.C.s (integrated circuits), a.f.c. 71, 77 width control 194,197,198,
a.g.c. 77 199,231,236,281
colour decoder 136-40 linearity see Cathode ray tube, Field
fault-finding 142-3 time base, line time base
field time base 176-8 linearity sleeve 212-13
line time base 185,189 L.O.P.T. see line time base, output
luminance 92, 137-40 transformer
power supplies 48,269,280-3, Luminance signal 5
285 amplifiers 85-7, 100, see also
sound stages 149-50 Video amplifiers
sync. separator 177 bandwidth 85,87
tuner control 48,54,55,56,57 delay line 85-6
video stages 75-7,92,137-40 formation of 11, 28
Ident signal 129, 130, 131, see also waveforms see Colour bar signal
PAL switch
I.F. amplifiers 39, 58-69 Matrix see Convergence correction,
alignment 66-9 Encoder, (G'-Y') signal, R', G', B'
response shaping 58-65 matrix
Illuminant D 244 Mixer stage 50, 52-3, 74
Integrating circuits 157-60,191 Moire pattern 236
Intercarrier sound see Sound stages Multivibrator 165-7, 171-4
Interference signals 59,60,74, 75,
105, 107, 145, 148, 181, 277, see
also Noise signal Negative feedback 63, 84, 151, 153,
Interlace filter 164, 168 164,170,175
Interlaced scanning 15,180,247 Negative modulation 8-9, 73
Noise signal 9, 37, see also Interference
Keystone distortion 231 signal
N.T.C. see Thermistor
Lecher bars 47,49,50,53 N.T.S.C. colour system 23
Umiting amplifier 69, 145
line frequency 26 Oscillator, blocking 166,167,189
line time base 41, 181-209, see also crystal 129
E.H.T. field see Field time base
driver amplifier 190 Hartley 186-8
fault-finding 206-9 line see Line time base
flyback tuning 191-2,193,194, multivibrator 165-7,171-4
195,200 subcarrier see Sub carrier
linearity 161-4,194,197 regeneration
oscillators 177, 185-9 tuner 50
output stage 192-5,195-8, Overcurrent protection 283-4
198-201 Overvoltage protection 284-6
Index 293

PAL colour system 22-3, 114 colours

PAL-D 116 control 95,113-14,137
PALoS 116,119,137 S-correction 162-4,193,194,197,200
PAL switch 133-6, 140, 142 S.C.S. (silicon controlled switch) 173,
Parabola waveform 170-5,218-19, 174,175,273
227,230 SECAM system 24
Peak white 8,25,26,122 Shadowmask 214,217
Phase errors 22-3, 24, 114-17 Shift, picture 213
Phasor diagrams 31 , 32, 115 Sidebands 15-17, 19, 59
Picture tube see Cathode ray tube Silicon controlled rectifier (SCR)
Pincushion distortion 213,225-7 see Thyristor
colour TV 225-31, 243 Simple PAL see PAL-S
monochrome TV 213 Sound muting 57
correction transductor 227-9 Sound system 41
Positive modulation 8-9,73 amplifiers 145, 150-3
Pre-emphasis 147, 148 de-emphasis 147-8,150
Primary colours 4 detection 146-50
Protection circuits see Beam limiting, intercarrier principle 36, 144
Overcurrent protection, Overvoltage pre-emphasis 147, 148
protection, Voltage regulators Spark gaps 234
P.T.C. see Thermistor Stabilisation of e.h.t. 198-9,203,
Purity correction 212,237-8 279,284
height 170
Quadrature modulation 20,120 power supplies see Voltage
Quiescent current 150,152 regulators
width 198-9,281
Raster 41 Stability, amplifier 61, 63
Raster correction see Convergence Striations 198,208
correction, Pincushion distortion Sub carrier , colour, burst see Burst
Reactance stage 129, 186 signal
Rectifier circuits 194,265-8 choice of frequency 19
Reference oscillator 125-30 dot pattern 19, 60
Reflected signal see Ghost signal filter, notch 86-7
Remote control 55-7 90° phase shift 134, 140
R', G', B' amplifiers 82,83,87-9 regeneration of 125-30,137
R', G', B' drive 87-9,100,111,122-24 suppression of 19-20
R', G', B' matrix 38,40,88,122,140 Surface acoustic wave filter 64
Ringing 65,189,190,198,282, see Swinging burst 23
also Stability Switch-off spot suppression 91-3
Switched mode field time base 178
Safety 69, 209, 283-7 Switched mode power supplies 272-83
aerial installation 46- 7 Synchronising pulses 7, 25, 34-5,
cathode ray tube 233-4 157-61
components 209, 287 Synchronous demodulation 75-9,
legal implications 44, 209, 287 119-21
power supplies 44 Sync. pulse separator 41, 154-7, 176,
X-rays 209,234 178-9
Saturation 5, see also Desaturated noise gated 155-6
294 Index

Teietext 36,79,260 Video detector 39, 73-9

Test cards 245-8 diode circuit 74-5
Thermistor 170,175,233,264 filtering 75,77
Third harmonic tuning 203 polarity of 73
Thyristor 273 synchronous 75-9
line time base 204 Video recorder 185
power supplies 273-7 Video signal, black level of 25-6, 78,
Tilt control 219 80-2
Time constant 158,162 [Link] 78, 79
Tint control 24, 232 formation of 7,8-9,13-14,
Transductor 227-9 32-3
Triad 214 polarity of 73
Trinitron 217-18 time/amplitude relationship 25
TripIer 202 Vision buzz 145, 148
Tuner 37 Vision carrier, channel allocation 17,
a.f.c. 50 58,59
a.g.c. 48,49,107-8 negative modulation 8-9
diode switching 52-3 positive modulation 8-9
electronic see varicap Vision Lf. 58
fault-finding 57 Voltage multipliers 201-2
integrated 50-3 Voltage regulators, chopper circuit
mecha'nical 47-8,50 277-8
mixer 50,52-3 I.C. 48,269,280-3
oscillator 50,52-3,58 line output 278-80
remote control 55-7 protection circuits 275-6,282,
r.f. amplifier 46-7, 48-50 283-6,288
touch control 53-5 series transistor 269-71, 277-8,
u.h.f. 46-50 284
varicap 48-50 shunt transistor 272, 278-83
v.h.f. 50-3 switched-mode 272-83
thyristor 273-7
U and V signals see Chrominance signal Zener diode 268-9

Varicap diode 48 Waveforms see Colour bar signal, Con-

VDR (voltage dependent resistor) 91, vergence correction, Field time base,
93,168,198,199,269 Line time base, Pincushion distortion
Vertical deflection see Field time base Weighting factors 22, 33
Vestigial sideband transmission 17, 59 Wideband amplifiers 61
Video amplifiers 39,79-85,99-101, Width control 194, 197, 198, 199, 231,
see also Luminance amp., R', G', B' 236,281
amp. Wobbulator 67
bandwidth 79
brightness control 82,83,90, X-rays 209,234
91-3 X-tal see Crystal oscillator
contrast control 93-5
d.c. amplification 79-83 Y-signal see Luminance signal
fault-finding 108
frequency response 60,74,79, Zener diode 268-70,272,276,284,
83-5,90,246 285