Television How It Workes Rider 1948

TELEVISION
HOW IT WORKS
%i
JOHN F. RIDER PUBLISHER, INC.

480 Canal Street New York 13, N. Y.
TELEVISION
HOW IT WORKS
JOHN F. RIDER PUBLISHER, INC.

480 Canal Street New York 13, N. Y.
Copyright 1948 by
JOHN F. RIDER
First Printing, June, 1948

Second Printing, August, 1948
Third Printing, October, 1948
All rights reserved including that of translation into the

Scandinavian and other foreign languages
Printed in the United States of America

TABLE OF CONTENTS
CHAPTER 1. CHAPTER 6.
GENERAL ASPECTS OF THE TELEVISION THE VIDEO I -F AND DETECTOR SECTION . 92
SYSTEM 1
The Video I -F System-92. Bandwidth Require-
Comparison with Sound Broadcasting-2. Scan- ments -93. Overcoupled I -F Transformers-94.
ning -2. Number of Elements Required -3. Need Resistance Loading -94. Typical Overcoupled Cir-
for Scanning -4. THE CAMERA AND PICTURE cuits Employing Resistance Loading -95. Stagger-
TUBES-6. The Picture Tube -6. The Camera
Tube-9. SCANNING AND SYNCHRONIZA-
-, Tuned I -F Transformers -97. The Need for Sound
and Video I -F Traps -100. Practical I -F Traps -
TION-12. The Scanning Pattern -12. Flicker and
Hum on the Raster-13. Scanning Waveform
13. Over-all View of a Television System -14.
- 101. Contrast Control-105. Automatic Gain Con-
trol (age) -107. Video I -F Amplifier Tubes-108.
VIDEO ,DETECTOR -109. Phasing of the Picture
THE TELEVISION SIGNAL -15. The Video Sig- Signal -109. Detector and Amplifier Action -111.
nal-16. Signal and Sync Pulses -17. Standard Detector Loading-113. Garod Model 3912-
Television Signal-18. Horizontal Blanking and TVFMP-114. Belmont Model 21A21-115. Gen-
Synchronization -18. Vertical Blanking and Syn- eral Electric Model 802-115. Consolidated Tele-
chronization -19. Range of Frequencies in Video vision Model 2315-116. Television Assembly
Signal -20. The Modulated Wave -21. Positive Model F1-101-116. Hallicrafters Model T-54-
and Negative Modulation-22. RECEIVER CIR- 117.
CUITS: GENERAL -22.
CHAPTER 7.
CHAPTER 2.
FREQUENCY CHARACTERISTICS OF THE VIDEO AMPLIFIERS AND D -C RESTORERS 120
TELEVISION SIGNAL 25 Requirements of a Video Amplifier-120. Basic
Television Channels -25. Video Signal Character- Circuit for Video Amplifier-121. Low -Frequency
istics -25. Vestigial-Sideband Transmission-26. Compensation -121. High -Frequency Compensa-
Operating Bandwidth Characteristics -27. The tion-124. AVERAGE BRIGHTNESS AND D -C
Carrier and Intermediate Frequencies-29. RESTORER CIRCUITS -134. D -C Restorer Cir-
cuits -135. Basic D -C Restorer Circuits -136.
CHAPTER 3. Brightness Control-137. Grid Leak -Capacitor
Restorer -137. The Contrast Control -140.
TELEVISION RECEIVING ANTENNAS . . . 31
The Transmitted Television Signal-31. Horizon CHAPTER 8.
Range and Line of Sight -32. The Television Sig- SYNCHRONIZING CIRCUITS 141
nal at the Receiver -34. Voltage and Current Dis-
tribution -34. The (Half -Wave) Dipole Antenna General Requirements -141. Sync Methods -In-
-35. Antenna Resistances -36. Resonance and stantaneous Locking Versus AFC-142. Sync Sep-
Impedance -36. Transmission Lines-36. Imped- aration -143. AFC Circuits -147. Reactance Tube
ance Matching -38. Q of Antenna-40. The Folded
Dipole -41. Length of the Half -Wave Antenna -
42. Indoor Antennas -44. Dipole with a Reflector
Operation -148. Sweep AFC Circuit in RCA
Model 630TS-149. Additional Circuit Features
151.
-
-44. Direct, Reflected, and Blocked Waves-46.
Ghosts or Multiple Images -47. Maximum Volt- CHAPTER 9.
age Input-48. Noise Reduction-49. Installation
and Orientation of Antennas -49. Other Types of SWEEP CIRCUITS 152
Television Antennas -51. Two or More Receivers Electrostatic Versus Magnetic Sweep-152. Sweep
on One Antenna -54. Oscillators -153. The Blocking Oscillator-156.
CHAPTER 4. Required Sweep Waveforms-158. Typical Sweep
Circuits for Magnetic Deflection-158.
R -F AMPLIFIER, OSCILLATOR, AND MIXER
CIRCUITS 55 CHAPTER 10.
The R -F Amplifier-55. Input Circuits -56. Fre- THE PICTURE TUBE 163
quency Converters -57. High -Frequency Oscil-
lators -58. Belmont Model 21A21-59. General Screens-163. Persistency of Vision -163. Focus-
Electric Model 802-61. RCA Model 621TS-62. ing-164. Optical Analogy of Focusing -164. Elec-
Du Mont Model RA -103-64. Westinghouse Model trostatic Focusing-165. Magnetic Focusing-167.
H-181-66. Deflection of the Beam -167. Electrostatic Deflec-
tion-168. Magnetic Deflection-169. Combined
CHAPTER 5. Methods of Focusing and Deflection-169. ION
THE F -M SOUND CHANNEL 68 SPOT -170. Ion Trap Methods -170. PROJEC-
TION TUBES -171.
THE I -F SYSTEM -68. The I.F.'s and Image Fre-
quency Interference -68. I -F Stages-69. Detec- CHAPTER 11.
tion of the F -M Signal -70. THE LIMITER SYS-
TEM-71. Analysis of Limiting Action-72. AVC POWER SUPPLIES 173
from Limiters -75. Input Voltage Considerations Low -Voltage Power Supplies -173. High -Voltage
-75. THE DISCRIMINATOR SYSTEM-75. Cir- Power Supplies-176.
cuit Analysis-76. Resonance Conditions in the
Phase Discriminator -77. Applied Frequency
Higher Than Resonance-79. Applied Frequency CHAPTER 12.
Lower Than Resonance-80. Summary -80. The ALIGNMENT AND SERVICING 182
Discriminator Output Curve -81. Pre -emphasis
and De -emphasis -81. Modifications of the Dis-
criminator Network -82. THE RATIO DE-
ALIGNMENT -182. Equipment Required -184.
Alignment Procedures -184. Sound I -F Section -
TECTOR -85. Simplified Ratio Detector -85.
Practical Ratio Detector -86. AVC From Ratio
Detectors -87. Ratio Detector Modifications -88.
185. Trap Circuits-187. Video I -F Section -188.
Local Oscillator -189. R -F Amplifier and Mixer
189. TROUBLESHOOTING -190. Localizing the
-
Slope Detection -90. Trouble -191. Troubleshooting Chart -193.
PREFACE
IT is our feeling that while television embraces hand -in -hand, we have used as typical examples those
many new techniques, nothing is so complicated circuits or components that are in actual receivers, not
in television that makes it beyond the capabilities forgetting those which might be called unusual. Thus
and comprehension of the members of the service if the reader will digest the contents of these pages,
industry. he will find himself with a widened viewpoint that
In line with this thought, we give in this "How It will be valuable in his advancement toward becoming
Works" explanations of the operation of television a better all-around television technician.
receivers such as are on today's market. A man must The opening chapter presents an over-all picture of
know more than the contents of this book, which is television, for it was felt the reader should at least
merely a stepping-stone, for no textbook can be hoped have a nodding acquaintance with the manner in
-
to convey everything at once. This book will develop which the subject is televised, how the image is built
--
practice will develop up in the camera tube, how the video and sound sig-
more -
a familiarity with television
other texts will add still more knowledge
attendance at a school will be of great benefit past
nals are transmitted, and then picked up, separated
and reproduced in the receiver. The next chapter deals
stone on which to build -

experience on regular broadcast receivers is a corner-
in short, it is a combination
of many things that will make a competent television
with the television channels and the characteristics of
the video signal ; operating bandwidth characteristics,
and the carrier and intermediate frequencies. The
technician. third chapter discusses antennas designed for the re-
One thing is vital while an individual may ac-
: ception of television signals, how they should be ori-
knowledge that he does not know sufficient at the ented and how they should be installed. The next
moment to service a television receiver, he should not eight chapters cover the receiver proper, the titles
develop a fear of what is in such a receiver. It is hoped being the r -f amplifier, oscillator, and converter; the
that the contents of this book will tend to expose what sound channel ; the video i -f system and detector ;
is in a television set so as to show that making the video amplifier and d-c restorer ; sync circuits ; sweep
effort to learn will develop the necessary capability. circuits ; the picture tube, and finally the power sup-
Foremost in a man's mind should be the fact that plies. The final chapter contains practical discussions
unlike a radio receiver which does not have a visual of test instruments, signal generators, etc., used in the
indicator, the picture tube in a television receiver is a alignment and maintenance of receivers, alignment
guide post toward the very many defects which may procedures, and trouble -shooting suggestions.
be found in a receiver and will tend to lead the man We wish to express our gratitude to the many
having the necessary power of interpretation, toward manufacturers who cooperated with us in supplying
the cause of the faults. teclinical data and descriptions of their products. Also
We have incorporated in this book the theoretical we wish to thank the members of the staff of John F.
principles upon which the various circuits of a tele- Rider Publisher who assisted in the preparation and
vision receiver are based, for only if a man has a grasp writing of this book and whose names appear at the
of this underlying theory can he hope to become a heads of the chapters.
competent technician. As theory and practice must go JOHN F. RIDER
CHAPTER 1
GENERAL ASPECTS OF THE TELEVISION SYSTEM

BY WILLIAM BOUIE
No COMMENTS can be made on recent advances in terial is unnecessary, it should be pointed out that
the radio art without recognizing that after many television is an extremely complicated subject, much
years of development, television has finally reached more so than is radio, and that television receivers
the stage where servicemen can no longer of ford cannot be serviced efficiently by any one who does
to ignore its existence. In most parts of the country not understand the principles underlying the opera-
television is already an accomplished fact, thousands tion of the system. The different situations which
of receivers have been sold and installed, and all can arise are so numerous that it is impossible to
signs point toward the gradual expansion of tele- list them, to explain the reasons for each, and in the
vision facilities into sections which do not have cov- case of faulty operation to locate the source of the
erage at the present time. trouble. As in radio servicing, so it is in television
Recognizing, then, that the radio serviceman's servicing: a thorough understanding of the funda-
job is being extended to include not only radio serv- mental principles of operation is invaluable.
icing as we know it today, but also the servicing of Fortunately for the radio serviceman, the advent
television receivers, this section explaining the fun- of television does not mean that an entirely new
damentals of television is being offered in the belief field must be learned. The more you study tele-
that it will be of help to many of you in dealing vision, the more you will come to realize that tele-
with the problems introduced by television. Through- vision embodies every principle that has ever been
out you will find that it is written from the view- used in radio-and more besides. You will marvel
point of the man who is called upon to install and at the ingenuity of television engineers in using the
service television receivers ; and yet, although there time-proved principles of radio, in adapting these
is this emphasis on the practical aspects of tele- to the needs of television, and in discovering new
vision, you will find that an essential amount of principles and new techniques wherever the avail-
theory is included. Lest any of you feel that this able ones were inadequate. Thus, for example, you
inclusion of what might be called "theoretical" ma - will find the simple diode rectifier being used in
TRANSMITTER RECE/VER
SOLINO
RF AF
MICR
RHONE
TRANS-
M/TTER AMGL A MPL .
REPRO-
DUCER )))
SOUND
SOURCE
OF CO/WERT3 SOUND CARRIER CARRIER CONVERTS
60 WAVES INTO (UNMODULATEO) (MOOUL ATEO)
50 UNO ELECTR/CAL ELECTRICAL
WAVES 5IGNAL SIGNAL INTO
.SOUND WAVES
(A)-SOUNO TRANSMISSION SYSTEM
LIGHT
INN/TE BACK
NEON
GLOIVRIBE TRANSMITTER RECE/ VER
P, -/OTO AMA./ TRANS- RF LIGHT
EL(CTEE AMPL . REPRO-
F.ER MITTER
íI1\\ TUB_ DUCER
TO LIGHT
AC
GROUNO GLASS CONVERTS LIGHT CARRIER CARRIER CONVERTS
SCREEN ILLUMINATED FL 1./C TVA T/DNS LINMOOLILATED) (MOOuLA TEO ELECTRICAL
INTO SIGNAL INTO
BY LIGHT VARYING
FROM PARK TO
LIGHT AT /2O^,
ELECTRICAL
SIGNAL fLLIGHT
I/C71/ATION3
(B)-TELEVISION TRANSMISSION SYSTEM
Fig. 1-1.-Part (A) shows an a -m sound transmission system and (B) shows a television transmission system. In the tele-
vision system the photoelectric tube replaces the microphone and the light reproducer replaces the speaker.
1
2 RIDER'S-"HOW IT WORKS"
clipping circuits, in d-c restoring circuits, in video Once the varying light values have been changed
detector circuits, and in many other applications. If into corresponding electrical values by the television
you understand how the diode rectifier operates, camera, the process of transmitting the information
then you will recognize in each one of these "new" follows exactly the same procedure as in the case
circuits an old friend. True, it will take some time of sound broadcasting. Note that the carrier is
before you become familiar with the need for these modulated in the same way, and that it remains
circuits and the specific modifications to achieve stationary in amplitude during the period before the
certain results, still you will be amply repaid for screen is illuminated. Once the neon tube is turned
the effort in the new opportunities which television on and illuminates the screen, the amplitude of the
offers to the trained serviceman and technician. carrier varies in proportion to the amount of illum-
ination. Note that the maximum amplitude of the
Comparison with Sound Broadcasting carrier corresponds to a black image, and that the
image gets progressively lighter as the amplitude of
Just as it is the problem of radio broadcasting to the carrier is decreased. This is called negative
recreate sound at places distant from the actual modulation about which we shall have a great deal
sound, so it is the problem of television to recreate more to say later on.
a scene at places distant from the original scene. For the present, the important thing to note is the "
In the case of amplitude modulated (a -m) sound similarity between the two systems, the one for
broadcasting, as Fig. 1-1 (A) shows, sound vibra- transmitting information on light values, and the
tions produced by, let us say, a tuning fork, are pick- other for transmitting information on sound values.
ed ùp by a microphone which converts these vibra- At the output of the television receiver, we of
tions into corresponding electrical vibrations. This course have an important change. Whereas the out-
electrical signal, which now carries an electrical put of the sound receiver is a speaker which con-
image of the sound vibrations, is amplified, used to verts the electrical impulses into corresponding
modulate the carrier ; the radio wave is radiated, sound impulses, the output of the television re-
picked up by the receiver, amplified, detected ; and ceiver is a "picture tube" or other device which
fî'nally, as you know, the electrical impulses (similar converts the electrical impulses into corresponding
in shape to those which were originally produced light values.
by the microphone) are used to actuate the speaker We see then, that the a -m sound system and the
which in turn recreates the original sound. television system are identical with the exception
Fig. 1-1(B) shows that a system very similar to that the television camera is substituted for the
this is used in television and that in general a very microphone and the picture tube for the loudspeak-
close resemblance exists between sound broadcast- er. We might also mention here that in the RCA
ing and television. To simplify the explanation at television system, the trade -mark name "Iconoscope"
the present time, we assume that only a very small is used for the television camera tube and the trade-
part or element of a scene is being televised. For mark name "Kinescope" is used for the picture tube.
example, we might allow the light from a neon tube As will be explained in detail later on, the Icono-
operated from the 60 -cycle power line to fall on a scope consists of a very large number of minute
small piece of ground glass. The illumination on the photoelectric cells which create an electrical picture
ground glass would change from dark through va- of the scene being televised, while the Kinescope
rious shades of brightness and back again to dark, consists of a cathode-ray tube, on the screen of
and repeat this cycle 120 times per second. (Note which is built up a visible image.
that the rate is 120 cycles because both positive and
negative cycles cause the neon tube to illuminate Scanning
the screen.) In the same way that sound broad-
casting uses a microphone to convert the sound pres- No doubt you have noticed by this time that in
sure variations into electrical variations, so the heart comparing television with a i sound broadcasting
of any television broadcasting system is the "tele- we limited ourselves to televising the simplest type
vision camera" which converts the time variations of object, one which was uniformly illuminated over
in illumination of the scene into corresponding elec- its entire area. We then showed that the two systems
trical variations. In the simple illustration we have are identical provided that the television camera
chosen, because only a single small area is being replaces the microphone and the picture tube re-
televised, an ordinary photoelectric cell can serve places the loudspeaker. Unfortunately, however,
as the television camera. television is not as simple as this or television would
GENERAL ASPECTS OF THE TELEVISION SYSTEM 3
have "arrived" many years ago. In television, we

are confronted with the problem of conveying in-
formation on the light value not at one point but
at every point over the complete area of the scene
being televised. Thus, the scene must be broken up
into a great many elements or elemental areas and
information on the light values over each one of
these elements must be conveyed to the receiver
and finally to the picture tube. And not only must
this information be conveyed, but it must be reas-
sembled in the correct order at the receiver and the
corresponding light value reproduced for every one
of the many elements into which the image has been
broken down.
As a matter of fact, this process of breaking
down a picture into a great number of elements is Fig. 1-2.-Halftone reproduction of a photograph using a
nothing new but is as old and as fundamental as the screen of 50 dots to the inch is shown in (A), at left. The
process of seeing itself. For, in viewing a scene, the greater detail in (B), at right, was obtained by the use of
a screen of 100 dots to the inch.
image is carried to the brain by the eye over a huge Courtesy Westinghouse
network of transmission lines which tells the brain
the intensity and the color of the light at every
point in the field of vision. Because the number of (B) . These are reproductions of the same photo-
elements into which the retina of the eye breaks graph, the only difference being the number of
down the scene is so great, we are not conscious elements into which the picture is divided. If you
that the picture is made up in this way but receive examine these figures closely, you will see that they
the impression that the picture is perfectly blended are composed of a large number of black dots of
or continuous. different sizes, and that Fig. 1-2(B) contains a
Some of you will be surprised to know that even larger number of dots than Fig. 1-2(A). In print-
photographs are made up of elementary particles ing as well as in the processes of seeing and photo-
even though they too appear to be continuous upon graphy, it is necessary for the picture to be broken
casual inspection. Actually the light and dark parts down into a series of small areas before it can be
of a photograph are the result of the presence of printed. The engraver in making his halftone, places
black particles of silver which vary in number over a fine mesh screen in front of his camera so that
the area of the picture. Where the picture is dark, the image is broken down into a series of dots, the
these particles of silver are more numerous than actual size of any one dot depending upon the
where the picture is light. Because these particles amount of light on the area which the dot repre-
are so small they are not ordinarily visible. We sents. Where the picture is dark, the size of the
might note in passing that where photographs are corresponding dot is large, and where the image is
to be enlarged appreciably so-called fine-grain film light the size of the black dot is correspondingly
and special developers are used so that the individ- small. The number of dots into which Fig. 1-2(A)
ual grains or particles will not become visible. is broken down is 50 per inch, while Fig. 1-2(B)
is broken down into 100 per inch.
The great improvement effected by breaking down
Number of Elements Required
the picture into a larger number of dots or elements
It is important for an understanding of television is clearly apparent in the superiority of Fig. 1-2(B)
to appreciate how the number of particles or areas over Fig. 1-2(A). Because of the larger number of
into which a picture is broken up affects the type elements, the former presents more detail, appears
and quality of the reproduction which is obtained. finer, better blended and more continuous than the
As we should expect from the preceding discussion, picture with the fewer number of elements.
the quality of the picture will be improved as the The actual number of elements into which a pic-
number of elements is increased. In order to com- ture must be divided depends upon several factors :
pare the effect of breaking up a picture into a vary- the fineness of detail which it is desired to repro-
ing number of elements, let us consider the two duce, the distance from which the picture is viewed,
reproductions shown in Fig. 1-2(A) and Fig. 1-2 and the size of the picture. Fineness of detail, as
RIDER'S-"HOW IT WORKS"
we have seen, requires a large number of elements from a greater distance than small ones to get the
for a given portion of a scene. In addition, the more same effect.
closely the picture is viewed, the smaller must be An idea of the number of elements needed to re-
the individual elements. This is necessary so that produce fine details can be obtained from Fig. 1-2
the individual elements will appear to the eye to (B) 'This has 100 dots or "elements" per inch in
merge into smooth lines and shades. If a picture of
a given scene is enlarged, either the total number
.
an area 1.5 by 2 inches, giving 150 X 200 - 30,000

total elements, or 10,000 elements per square inch.
of elements must be increased, or the picture must Television pictures may be considered to contain
be viewed from a greater distance. Figs. 1-2(A) and about 367,500 elements, regardless of picture size.
1-2(B) illustrate the first two factors, fineness of Although this does not work out to be so great a
detail and viewing distance. Fig. 1-3 which shows number per square inch on large picture -tube
Fig. 1-2(B) enlarged to twice its original size, screens, the fact that the scene is usually in motion
illustrates the third factor. Fig. 1-2(B) contains 100 compensates for some loss of detail. A point to re-
dots per inch; Fig. 1-3 contains only 50 dots per member about television is that increasing the num-
inch. Each contains the same total number of ele- ber of elements increases the frequency bandwidth
which must be transmitted. Thus there are technical
and economic limitations to the degree of detail
that may be provided.
We can now begin to appreciate the complexity
of the problem with which television is confronted.
For not only must information on the light value
of each one of many thousands of elements be trans-
mitted to the receiver, but also information as to the
order or sequence in which these light values must
be assembled to form the picture. To make the prob-
lem even more complex, all this information must
be transmitted in approximately 1/30th of a second
in order to prevent blurring due to movement in the
scene and in order to make way for the next picture
impression or "frame." In this respect the problem
of television is more difficult than that of facsimile,
since in the latter a still picture is transmitted and
the time consumed may be ten minutes or more in-
stead of 1/30th second.
Need for Scanning
By this time we have seen that to make television
possible, the picture must be broken down into a
large number of elements and information trans-
mitted on the light value at each one of the ele-
ments. At the receiver this information is reassem-
bled in the proper space relationship to form the
original picture. How to transmit this information
Fig. 1-3.-The same detail as in Fig. 1-2(B) can be ob- is the next problem.
tained by viewing this enlargement from twice the distance.
Courtesy Westinghouse Previously we saw that using a conventional a -m
system of radio we could transmit information on
the light value at any one element of a scene by using
ments, therefore the fineness of detail is the same in a photoelectric cell pickup to convert the light value
each. Because Fig. 1-3 has larger dots, it must be to an electrical value, and that this process was es-
viewed from a greater distance. This consideration sentially the same as that of sound broadcasting and
is important in television, since the total number of involved essentially the same transmitting and re-
lines into which the scene is divided is the same at ceiving equipment. This was shown and explained
all times. Therefore, large pictures should be viewed in Fig. 1-1. The first thought that arises is this:
TRAN5M/TTER RECE/ VER

OBJECT /MAGE
1 2 3 a 5 ss rigga©ç©..
r
'"
8 9: /O Ejr /2 iY,mttl®
/3/d l5 /7 /5 W m
/ / /
47;0 2/ 22 29 24
/,l ¡25 27 ,ßp429 30 /.( /

/,U/, /
7.4.
if
/t.\G/
/ /
31 32 33 34 95
/ /
/ / V
/ //l1 /
AREA NOT
/
YET SCANNED
ELECTRICAL SIGNAL Fig. 1-4.-The photoelectric tube scans

FROM P -E TUBE the checkerboard pattern in the order indi-
CONTROL.3 /NTEN1/7Y OF SR7TL/G/T) cated by the numbers. The pattern is re-
produced at the receiver by means of a
MECHAN/CAL L/N/! spotlight, the motion of which is syn-
FOR VERT/CAL CONTROL chronized with the photoelectric tube and
MECHANICAL L/N/C the intensity of which varies in accord-
FOR HORIZONTAL CONTROL
ance with the amount of light reflected
from the square being scanned.
Why not use individual channels of the type shown at which time we can consider that the whole page
in Fig. 1-1(B) to transmit information on the light has been scanned.
values at each one of the points into which the pic- This same procedure is used when a television
ture is divided ? But then, this would require some camera scans a scene which is to be televised. The
100,004 individual pickups and transmission systems only difference is that the television camera breaks
each of which would be similar to the system of down the scene into finer elements than the letters
Fig. 1-1(B) Obviously such a system would be far of a page and that the camera produces electrical
too complex and expensive to be practicable, even impulses which vary in proportion to the amount
if other problems of great difficulty at the receiving of light on each element of the scene being scanned.
end did not exist. The following simple example will help to clarify
A more promising solution, which is used in all the fundamental principles and requirements in-
television systems today, works on the basic idea volved in scanning. Suppose we wish to reproduce
of transmitting information on the light value of by television the pattern shown at the left of Fig.
one element at a time. In this way the picture is 1-4. Let us assume in this example that a television
covered or "scanned" in a systematic way until camera (consisting essentially of a photoelectric cell
finally the image is said to be completely scanned with the proper lens equipment) is available which
when information on the light values at every point can be focused so as to pick off the light values on
in the picture has been obtained. The need for more any one of the squares into which the picture has
than one channel is thus avoided. However, the in- been divided. Let us also assume that we have a
formation must still be reassembled in the proper mechanical arrangement for moving the camera
order at the receiving end before the picture can be both horizontally and vertically so that it will scan
obtained. the object. That is, it is possible to Mart with the
Scanning a scene is sometimes compared with the camera focused on element 1 and to move. it at a
manner in which we read a printed page. For in- uniform rate across the screen until it reaches ele-
stance as you read this page your eyes start at the ment 6. At this point the camera snaps back rapidly
upper left-hand corner of the page and successively to element 7 at the beginning of the second line and
sweep to the right-hand side while examining every moves at a uniform rate along the second line until
letter on the line ; when the first line has been com- it reaches element 12. The camera then returns very
pleted the eye snaps over rapidly to the left-hand rapidly to the left and starting at element 13 on line
side, jumps down one line, and in the same way, 3 it scans the third line. In this manner the pro-
examines every letter on the second line as it prog- cedure continues until the entire pattern is scanned.
resses at a uniform rate toward the right. This pro- Note that at any one instant the camera receives
cedure continues until finally the eye reaches the light only from the particular element at which it
last letter on the lower right-hand side of the page, is aimed and focused and produces an electrical im-
L
pulse which is proportional to the amount of light 2. The image must be scanned at the receiving
reflected by this element. end according to the same systematic plan used by
So much for the scanning at the transmitting end the television camera and the intensity of the light
where the picture is being televised. At the receiv- emitted by the light source in tracing the image
ing end let us assume that we have a projector must vary at every instant in accordance with the
which projects a narrow pencil of light on the screen amount of light which the camera is receiving at
equal in area to one of the square elements. This that instant.
projector like the camera can be moved horizontally 3. At every instant the camera and the light trac-
and vertically so that the light can be focused on ing the image must be synchronized so that the
any part of the screen. Suppose further that the identical portion of the image is being traced out
electrical impulses from the television camera are which corresponds to the element of area being
fed to the projector and arranged to control the scanned by the camera.
intensity of the light emitted by the projector in 4. This scanning procedure or process must be
accordance with the amount of light registered by completed over and over again at a rate of at least
the camera at any particular instant. 30 times per second so that as far as can be deter-
Under these conditions before a picture can be mined by the eye a continuous image of the scene
obtained at the receiver, the motion of the camera is formed.
at the scene being televised and the motion of the
projector at the receiver must be properly coordina- THE CAMERA AND PICTURE TUBES
ted or "synchronized." This means that the televi-
sion camera and the projector must go through the We have already seen that a complete television
same movements together, that the projector must system, like a complete broadcasting system, re-
at all times be focused on exactly the same element quires a pickup at the transmitter and a reproducer
in the picture as that on which the camera is fo- at the receiver, and that the pickup is a photoelectric
cused. In the figure we have assumed a sort of tube and the reproducer is a cathode-ray tube simi-
mechanical linkage between the camera and the pro- lar to those used in oscilloscopes. To avoid con-
jector to accomplish this; actually no such mechani- fusion of trade names let us call the pickup a
cal linkage is possible in television and we shall see "camera tube" (it "takes" the television picture)
later that electrical synchronizing pulses are used and the reproducer a "picture tube" (it reproduces
to control the camera at the transmitting end and the picture). Since we want to get a good general
the projector (or picture tube) at the receiver, so idea of how a television system works, we shall
that both the scene being televised and the image consider both these tubes here, although the service-
which is being reproduced at the receiver are man in the field will naturally come in contact with
scanned in unison-so that the scanning is syn- only the picture tube.
chronized.
In the picture shown, the image has been scanned The Picture Tube
only as far as element 13 ; element 14 is about to Essentially the television picture tube is similar
be scanned. As a result the image at the receiver is to the familiar cathode-ray oscilloscope tube, so that
totally dark beyond this point since the lower ele- those who have read Rider's "The Cathode -Ray
ments have not yet been scanned and hence have Tube at Work" will have a good basis for under-
not yet been illuminated. We shall explain later on
that the observer sees the complete image at one
time even though only one element of it is receiving
light at any particular instant. This is because the
entire scanning process is repeated some thirty or
more times a second, and the eye tends to see the
image after it is no longer illuminated.
We can now summarize the requirements which
must be met before a scene can be transmitted by
television :
1. The scene must be systematically scanned by
the television camera which interprets the light
values at every element of the scene in terms of Fig. 1-5.-A Kinescope (picture tube) with a 12 -inch
corresponding electrical values. screen.
Courtesy RCA
SECONO ANOOE CONOUCTOR

COATED ON /NS/OE
CATHODE - OF GLASS BULB
EM/TS FL L/ORESCENT
ELECTRONS ELECTRON GUN 5CREEN ON
/N5/0E FACE
OF GLASS
NEATER
MOOUL AT/ON
GR/0 -
CONTROLS Q!/ANT/TY
OF ELECTRONS
/N 6EAM 9O0ÿ F/RST ANODE OR
ELECTRON
BEAM (A)
11 I
MAX. FOCUS/NG ANOOE
i0
SCREEN Z50 V. 7000v
GA/0 MAX. MAX.
SECONO ANOOE -
.-41llglll I ACCELERATES ELECTRONS
TOWARD FLUORESCENT SCREEN
Fig. 1-6.-The different elements of a picture tube are

shown in (A) with typical operating voltages. (B) shows
a common schematic representation of this tube; K is
the cathode, G the modulation grid, S the screen grid,
F the focusing anode, and A the second anode.
standing television picture tubes. For those who A cross-sectional view of a picture tube is shown
have not we w11 review the subject here. in Fig. 1-6. As in the usual radio tube, the heater
Let us assume that television (video) signals are causes the cathode to emit electrons and the second
coming into a receiver; as we have said before, the anode (like the plate of the ordinary .tube) strongly
amplitude of these signals is proportional to the attracts them, giving them a high velocity. The mod-
light reflected by the object being televised. We ulation or control grid regulates the number of
want to use these signals to produce a picture. In electrons which pass through it in a given time. In
sound work we know that the signals can be made the picture tube additional elements such as a fo-
to move the diaphragm of a loudspeaker, thus pro- cusing anode form the electrons into a narrow beam
duoing sound waves similar to the original. The pic- so that they will ,strike the fluorescent screen in a
ture tube, then, must be capable of converting the small round spot ; in some tubes a screen grid is in-
electrical video signals into light to produce a pic- serted between the modulation grid and the fo-
ture. It is a property of certain substances called cusing anode to prevent the focusing action from
fluorescent materials that they will glow when they affecting the modulating action. Fig. 1-6(A) shows
are struck by a beam of electrons, and the more the general arrangement of parts inside the tube,
electrons striking such a substance at a given in- and Fig. 1-6(B) shows a common way of repre-
stant, the brighter will be the glow. A picture tube, senting these in schematic form.
then, can be made if we have a source of electrons, With the parts so far mentioned, the tube can
means for controlling their motion and their quan- produce a beam of electrons which will hit the
tity, and a fluorescent screen. center of the fluorescent coating on the inside of the
The external appearance of a typical picture tube picture tube and produce a small spot of light which
is shown by the photograph, Fig. 1-5. This is a can be seen through the glass end. We can focus
glass vacuum Elbe specially shaped to withstand the this spot by varying the potential on the first (fo-
high pressure exerted by the surrounding air, due cusing) anode, and we can vary its brightness by
to the high vacuum within the glass envelope. Serv- applying a suitable potential to the modulation grid.
icemen should remember this, and handle picture The more negative the modulation grid becomes rel-
tubes with great care. Even scratching the glass or ative to the cathode, the dimmer the spot becomes ;
careless handling may cause them to collapse as vio- the less negative the grid, the brighter the spot. It
lently as if they exploded. The white appearance of now remains to provide some means for moving
the large- end of these tubes is caused by the film this spot rapidly enough over the fluorescent screen
of fluorescent material deposited on the inside sur- to give us a complete picture ... in other words to
face ; this, of course, is where the television pic- provide scanning.
tures are formed. Two methods of deflecting the electron beam are
now commonly used electrostatic deflection and
: he located so that the distance AD is the hypotenuse
electromagnetic deflection. The first of these meth- of a right triangle whose sides are equal to DC and
ods, which is probably the simpler to understand, AC. Naturally, if V2 is made positive with respect
takes advantage of the familiar fact that particles to V1, the beam will be deflected vertically down-
of matter having like charges of electricity repel ward, and if H2 is made positive with respect to
each other, while particles having unlike charges at- H,, the beam will be deflected horizontally to the
tract each other. Since the electron beam consists left. The whole process of electrostatic scanning is
of negative charges, we see immediately that the simply a matter of varying the potentials on the de-
beam can be deflected by means of suitably shaped flecting plates in the picture tube so that the spot on
electrodes which are charged either positively or the screen traces out the desired picture according
negatively as required. In picture tubes, as in oscillo- to a regular plan. In a later section, we shall study
scope tubes, this is done by building into the tube the details of the plan actually used.
two pairs of metallic plates, arranged approximately The second method of deflecting the electron
as shown in Fig. 1-7. beam is called electromagnetic deflection, because
This figure shows that plates H1 and H2 are coils of wire carrying' current act on the beam the
parallel to each other but in a plane at right angles way a magnet would. See Fig. 1-8(A) : The mag-
to plates V1 and V2. If no potential is applied to netic lines of flux from a permanent magnet pass
any of these plates, the electron beam will pass from the north pole (N) to the south pole (S) ; the
straight along the axis of the tube and cause a spot electron beam passes between the poles. If the mag-
to appear on the screen at A. Now if we leave plates net were not present, the beam would produce a
H, and H2 alone, but make plate V, positive with spot at point A on the screen. When the magnetic
respect to plate V2, plate V1 will tend to attract field acts on the beam as in Fig. 1-8(A), however,
the negative electrons which make up the beam, the beam is deflected vertically upward to produce
thus causing the beam to bend, so that it strikes the a spot at point B. Compare this effect with that
screen at a new point, say at point B. We have thus produced by the electrostatic field in Fig. 1-7, and
deflected the beam upward in a vertical direction a you will notice that in Fig. 1-7 the electron beam
distance AB. Similarly, if we leave V, and V2 was deflected .in the direction of the lines of elec-
alone, but make H1 positive with respect to H2, H1 trostatic flux existing between plates V, and V2,
will deflect the beam horizontally to the right until whereas in the case of magnetic deflection, the beam
we get a spot at a point such as C. is deflected at right angles to the lines of magnetic
If we combine these effects, by making both V1 flux. Thus in Fig. 1-8(B) the beam is deflected
and H1 positive at the same time, the beam will horizontally to the right, when another magnet is
bend sidewise and upwards, causing the spot to ap- introduced whose influence is at right angles to the
pear at D. If the potentials of V, and H1 have one shown in Fig. 1-8(A).
been the same as used in the first two tests, D will Of course, a permanent magnet gives a steady
Hog/ z. £EFL. PLATES VERT. OEPL . PLATES
7H15 PA/R OF PLATES TN/3 PA/R OF PLA7ES
OEFLECT5 BEAM OCPLECTS BEAM
NOR/ZONTALLY VERT/CALLY
pEfE TEp E.T/_A_L~

:-
gEAM
-
9E,9.41 NOT OEFLECTEO-N.,J
VZ BGAM OEFLECTEO NO.PzZONTALLY
EL EC TRO.v/
GUN
--I II.11I.i
IO O
Fig. 1-7.-The electron beam in this picture tube is electrostatically deflected by the two pairs of plates. H, and H2 deflect
the beam horizontally, and V1 and VE deflect it vertically.
magnets
Fig. 1-8.-The magnetic deflection coils in (C) and (D) provide a magnetic field similar to that of the permanent
in (A) and (B). Note that the beam is deflected at right angles to the lines of force.
field and. would therefore produce a constant de- The Camera Tube
flection, whereas in television we must be able to Whereas the picture tube contains a fluorescent
vary the extent of the deflection from zero to maxi- screen to convert an electric current into light, the
mum in various directions. Since we can also pro- camera tube contains a photosensitive screen to con-
duce a magnetic field by passing a current through vert light into an electric current. Large and small
a coil, and in addition, can vary the strength of the RCA camera tubes, called Iconoscopes, the type in
field by varying the current, we use coils for elec- general use in this country at the present time, are
tromagnetic deflection, as shown in Figs. 1-8(C) illustrated in Fig. 1-10. The essential parts of the
and 1-8(D). Two .pairs of coils are used, and these large tube are shown in the cross-sectional view in
are often combined in a single compact cylindrical Fig. 1-11: these are the mosaic, the signal plate,
unit called a "deflecting yoke." Fig. 1-9 shows an the collector, and the electron gun. The important
RCA deflecting yoke. accessories external to the tube are the lens, the
A point to remember about deflection systems is deflecting yoke, and the load resistor.
that when one tube has electrostactic deflection, a The mosaic consists of millions of individual
change of voltage on the deflecting plates is required
to move the beam, whereas in electromagnetic deflec-
tion, a change of current through the deflecting coils
Cur -AWAY VIEW
is required. Of ASSEMBLED YOKE
Recently it was discovered that a metallic film

placed over the luminescent material on the inner IRON
vERTICAL-
SHELL
surface of the Kinescope tube improved the distri- CCrLECTING COIL
bution of light over the screen. This metallic film

is in the form of an aluminum backing which acts
HORIZONTAL-
as a reflector. In operation, the electrons pass DEFLECTING COIL
through the aluminum and strike the luminescent PAPER TUBE
material causing it to fluoresce in the usual manner. Courtesy RCA

However, very little light is lost within the tube
because of the reflecting properties of the aluminum Fig. 1-9.-The magnetic deflecting yoke contains the hori-
surface. The result is a gain in the brightness of zontal and vertical deflection coils which are designated here.
the received image. It is probable that this new
type of Kinescope may be used quite extensively
in new models of television receivers.
lo RIDER'S-"HOW IT WORKS"
of light in the image of the scene. The resultant cur-
rent which flows in the load resistor causes a volt-
age drop. This constitutes the video signal. This
signal is then amplified and used to modulate the
television transmitter.
The following more detailed description of the
action of the Iconoscope is presented for those in-
terested. When a scene is focused on the mosaic,
each photosensitive globule emits electrons in pro-
portion to the amount of light falling upon it ; the
more light that falls on a given area the more elec-
Fig. 1-10.-Both a large and a small Iconoscope are shown. trons are emitted from that area. This process is
Courtesy RCA called "photoemission." Since a loss of electrons
means a more positive charge, photoemission sets
up a variety of different charges over the surface
photosensitive globules like metallic droplets depos- of the mosaic in accordance with the distribution
ited on one side of a thin sheet of mica. Each globule of the light it is receiving. So far no video signal
is like a minute island on the mica, so that the has been produced because the globules are insu-
globules are insulated from each other. On the other lated from the signal plate.
side of the mica there is a layer of conducting ma- Since the path from the mosaic to the signal plate
terial ...the signal plate. Because the globules are is through a capacitor, the only way to get a signal
separated from the signal plate by the mica, the across is to produce a sudden change of potential.
mosaic consists of a myriad of mica -dielectric capac-
itors, all having one plate in common. Fig. 1-12
shows in a general way what the mosaic looks like
Fig. 1-12.-An enlarged AI/CA SHEET
when viewed from the edge and enormously mag- view of the way the photo-
nified. The collector ring is a metallic coating ap- sensitive mosaic is arranged METALL/C-
with respect .to the signal F/LM
plied in the form of a ring around the inside of the plate of a camera tube.
SIGNAL
PLATE
tube and also extended down the neck of the tube
near the electron gun.
Briefly the action of the Iconoscope is as fol-
lows : An optical image of the scene to be televised is In the Iconoscope type of camera tube, this change
focused on the mosaic by the lens. The electron gun of potential is produced by the action of the electron
projects a stream of electrons over the mosaic, beam from the gun. (The necessary orderly scan-
scanning it under control of the deflecting yoke, as ning action from left to right and top to bottom is
already described in connection with the picture provided by suitable currents through the deflecting
tube. The electron beam in traversing the mosaic yoke, as in the case of a magnetically deflected pic-
causes a succession of voltages to appear on the sig- ture tube.) In order to understand the effects pro-
nal plate which are proportional to the distribution duced by this scanning beam, let us first consider
three possible conditions on the surface of the mo-
saic before scanning. It has been found that the
COLLECTOR
LENS
EDE RANG
.e^mA.W06)
PHOTVJEN5/77
.OSA/C SCREEN ordinary Iconoscope mosaic in total darkness (black
FOCUS/NG
1,7,46E scene) asumes a potential of -1.5 volts relative to
ON Al/CA 3//G47"
MOSA/C the collector ring. When strong light (white scene)
SCENE" SIGNAL PLATE falls on a portion of the mosaic, a considerable num-
I (METAL,
ber of electrons are lost by photoemission, and the
EL EC TRON potential is changed from, say, -1.5 volts to 0
GUN Y/OEO OUTPN
volt. Illumination of medium intensity (gray scene)
L OAO
RES/5TOB may cause the potential to change from -1.5 volts
DEFLECT/NG
YOKE
to -0.8 volt. These potentials of the mosaic, then,
may be approximately as follows for these portions
Fig. 1-11.-The elements of the large tube shown in Fig. of the mosaic black area, -1.5 volts ; gray area,
:
1-10 are easily identified. Light images formed on the signal

plate are transformed into electrical energy by the scanning
-0.8 volt ; white area, 0 volt.
beam. The electrons in the scanning beam travel so rap-
idly that when they strike a surface such as the Icono- The important things to notice here are (1) that
:
scope mosaic, they knock off a great many electrons, we get a video signal across the load resistor (Fig.
more, in fact, than even bright light does. This kind 1-11) only when a change of potential is produced
of action is called "secondary emission." So many on the mosaic, and that this is accomplished by the
electrons are thus lost by secondary emission that scanning beam; and (2) that the resultant video
the portion of the mosaic directly under the action signal is proportional to the light coming from the
of the scanning beam is driven to a positive poten- scene being televised, the signal being of maximum
tial of +3 volts. Each portion of the mosaic is amplitude for black portions of the scene, minimum
driven to +3 volts during the instant the scanning amplitude for white portions.
beam acts upon it and this +3 -volt potential is The collector ring (2nd anode) serves to accele-
reached regardless of the light conditions prevailing rate the electrons in the scanning beam, and also to
... a black area goes to +3 volts as well as a white collect some of the electrons emitted from the
or gray area. Although the peak potential reached mosaic. In Fig. 1-11 you will notice that this ele-
by every area as the scanning beam passes over it ment is grounded ; it is, however, 500 to 1000 volts
is the same, the change in potential which this area positive with respect to the cathode, of the electron
undergoes at this time will depend on the illumina- gun, because the gun is at a corresponding potential
tion ithas received. Thus the black area will change negative with respect to ground. This will not be
from -1.5 to +3 volts, a change of 4.5 volts ; the the last time that you will find television tubes with
gray area will change from -0.8 to +3 volts, a elements at very high negative potentials, and the
Change of 3.8 volts; the white area will change from possibility of such circuits existing must always be
0 to 3 'volts, a change of 3 volts. It is this sudden borne in mind as a safety measure.
change of potential which is induced on the signal Another type of RCA camera tube more recently
plate that causes the video signal current to flow developed is the Image Orthicon. This tube retains
through the load resistor. The difference between the essential features of the Iconoscope but it is
the changes for black, gray, and white areas is what much more sensitive and efficient. Its high degree
indicates the difference in the illumination over the of sensitivity permits the pickup of scenes in can-
mosaic surface as it scanned, and this difference dlelight and varying degrees of darkness, and makes
is proportional to the distribution of light over the possible "round-the-clock" television coverage of
mosaic. news and special events. This tube, shown in Fig.
1-13(A), is incorporated in the portable super -sen-
sitive television camera manufactured by RCA Vic-
tor Division.
Fig, 1-13.-An Image Orthicon is shown in Fig. 1-13(A).

Note the size of the tube. The elements of the tube (includ-
ing the deflection yoke and focusing coil) are shown in
Fig. 1-13(B). RCA
Courtesy
CATHODE (ZERO) DECELERATING RING

(ZERO)
SECONDARY
ELECTRONS
SECONDARY
ELECTRON IMAGE
ELECTRONS
SIGNAL OUTPUT PHOTO -CATHODE

ELECTRODE (-600 V )
(+1500V) TARGET SCREEN
(ZERO)
ALIGNMENT COW TWO-SIDED TARGET
(B)
A simplified cross-section of the Image Orthicon
is shown in Fig. 1-13(B). In operation, the scan-
ning beam is projected from the cathode (extreme
left) toward the target at the right on which elec-
trical charges have been built up for focusing an
image of the scene on the photo -cathode. The scan-
ning beam is given just enough velocity to reach the
target. If the electrons in the beam strike a positive
charge they are absorbed. If the target charge is
negative, electrons in the beam are returned to the
area near the cathode. Incorporated within the tube
near the cathode are reflecting electrodes known as
dynodes. These dynodes are coated with a material
capable of high secondary emision. Referring to Fig. 1-14.-A simple progressive scanning pattern in which
the picture is scanned in a series of lines which start at the
Fig. 1-13(B), when the electrons in the return beam upper left and slope down to the right.
strike the first of the dynodes, secondary electrons
are emitted. These secondary electrons in turn strike below point B. In fact, all the lines in Fig. 1-14
other dynodes, greatly increasing the total number slant downward. This slant is necessary to provide
of electrons and hence the current carrying the pic- the vertical (top -to-bottom) part of the scanning
ture information. This procedure is repeated for process. In fact, the left -to-right component .of
each element in the televised scene. scanning is called "horizontal scanning" or "line
scanning"; the top -to -bottom component is called
SCANNING AND SYNCHRONIZATION "vertical scanning" (also called "frame scanning"
or "field scanning," for reasons which will be clear
In previous sections we discussed (1) the neces-
later on). This slant-line system of scanning is used
sity for scanning both the camera tube and the pic-
because the characteristics of the electrical circuits
ture tube, (2) the essential requirement that these
which provide scanning make it impossible to scan
operations must be synchronized, and (3) the de-
in a series of truly horizontal lines, with an abrupt
vices which make it possible to control the motion
vertical drop at the end of each line.
of the electron beam in these tubes. In this section
After the beam reaches point b it is deflected back
we shall describe the details of the path of the
along the dotted line bc to start a new line, cd. On
scanning beam on the tube screen, the waveshapes
the picture tube the line bc would be much fainter
necessary to produce this path, and some of the
than line ab because the time allowed for the beam
problems which arise in connection with scanning.
to move from b to c is only about one -tenth the time
The Scanning Pattern allowed for it to travel from a to b. Lines like ab,
cd, etc., are called "line traces"; lines like bc, de,
As we have already mentioned, the usual scan is etc. are called "line retraces" or "line flybacks."
from left to right and from top to bottom. In cer- Here some questions may arise : Why does the
tain cases this order must be reversed, but the motion of the spot across the screen appear as a
principle is the same. In Fig. 1-14 is shown a sim- line ? Why is the flyback dimmer than the trace ?
.
plified diagram, of a scanning pattern ; this is often The eye is responsible for these effects. Any one
called a "raster." The rectangular diagram formed who has ever swung a flashlight or a "sparkler" in
by the light lines connecting points A, B, C, and D a circle must have noticed that as the speed of the
represents what we may call the "picture space." swinging is increased, the individual spot of light
The width of this rectangle (AB) is drawn so as merges into an apparently continuous circle of light.
to be 4/3 times the height (AC) ; this is the propor- The velocity of the scanning spot across the picture -
tion used in motion pictures. The ratio of width to tube screen is so rapid (several thousand miles per
height is called the "picture aspect ratio" and its hour) that the eye cannot distinguish the spot from
value of 4/3 is one of the FCC television standards. a continuous streak. This property of the eye is
On a picture tube, the heavy lines ab, cd, etc., are called "persistence of vision."
caused by the fluorescent glow which occurs as the The difference in brilliance between the trace and
electron beam moves from point a to point b (in the the flyback is due to the fact that the brilliance of
direction of the arrow on line ab). You will notice the fluorescence produced on the picture-tube screen
that points a and A coincide, while point b is slightly depends on the time during which the electron beam
frame between two rasters (called "fields") , each

containing half the total number of lines required
to make up a frame. These fields, transmitted at the
rate of 60 per second, get rid of the flicker effect.
The required number of lines per frame is supplied
by sending the, odd -numbered lines along with one
field and the even-numbered lines along with the
following field. The fields are transmitted so rapidly
that as far as the eye can see, all the necessary lines
appear in their proper positions on the raster simul-
taneously.
The method of scanning just described is called
"interlaced scanning," and is the system now in use.
Fig. 1-15.-An interlaced scanning pattern in which the A simplified pattern of a frame produced by inter-
odd and even lines are scanned on successive fields.
laced scanning is shown in Fig. 1-15. The general
acts upon it. Although in both cases the actual time principles are the same as for progressive scanning;
is very short, the difference in time (about 10 to 1) in fact, each field is itself progressively scanned.
does result in a very marked difference in brilliance. The lines transmitted during the first field scan, re-
A complete set of lines such as shown in Fig. 1-14 ferring to Fig. 1-15, are lines 1, 3, 5, and the first
is called a "frame." This particular frame is com- half of line 7. The beam is then deflected to the
pleted when the spot reaches point 1 ; it then returns top of the picture space to begin the second field.
to point a to start a new frame. A frame like that During the second field, line 7 is completed and
shown in Fig. 1-14 is said to be built up by "prog- lines 2, 4. and 6 are traced, thus completing the en-
ressive scanning" (the lines follow each other in a tire seven -line frame in two steps. (As in Fig. 1-14,
continuous chain) . Six complete lines (retraces are the arrows show the direction of travel of .the beam
not counted) are shown in this figure; a complete spot.) The use of an odd number of lines has been
technical description of this figure would be "A :
found advantageous in interlaced scanning; this
six -line frame (or raster) produced by progressive method is called "odd line interlacing." American
linear scanning." Of course, so few lines as we have practice calls for 262.5 lines in each field, making
shown would be insufficient to give a good picture, 525 total lines per frame. Sixty fields are transmit-
and a very great many more are actually used. In ted each second, giving, a field frequency of 60 per
order to give the eye the illusion of motion, many second.
complete frames must be traced out every second. The frame frequency chosen depends upon the
The standard American procedure is to trace 30 a -c power-line frequency; it must be an even multi-
complete frames every second; each frame consists ple or submultiple thereof. The reason for this is
of 525 lines. that if there is some hum present in the deflecting
circuits, a certain amount of distortion will appear
Flicker and Hum on the Raster in the picture. If this distortion appears to be
stationary, and is not excessive, it can probably be
It has been found that if the raster is bright a tolerated. On the other hand if the distortion moves
flicker will be observed when the number of frames so that the picture appears to have moving ripples
per second (frame frequency) is less than a certain in it, even slight distortion is most objectionable.
critical value. Although a satisfactory illusion of mo- By making the frame frequency an even submultiple
tion might be produced by about 12 frames per of the power -line frequency, this type of hum pat-
second, as many as 48 frames per second may be tern can be rendered stationary. Since most Ameri-
required to remove objectionable flicker. The more can power lines use 60 -cycle a.c., the frame fre-
frames per second a television transmitter sends quency chosen was 30 per second. (In England,
out, each having a large number of lines, the higher with 50 -cycle a.c., the frame frequency is 25.)
the modulation bandwidth required. One way to
keep the bandwidth down is to send twice as many Scanning Waveform
frames per second, each frame having half the num-
ber of lines. But a large number of lines is neces- In describing the picture tube, we showed that
sary for detail in the picture. It was therefore de- the beam can be deflected by applying suitable po-
cided to divide the total number of lines in each tentials to the deflecting plates or suitable currents
RIDER'S-"HOW IT WORKS"
1/6 the time allowed to the line trace; the field re-
trace is allowed 1/13 the time of the field trace.
The total time intervals allowed for these waves
indicate that the line -scan frequency is 15,750 cycles
and the field -scan frequency is 60 cycles. These
waveforms are called saw -tooth waves ; circuits for
generating them are described in a later section.
M
TIME
N
/S,%SO
-. .5ECONO V =
T/ME-
6O SECOND
Over-all View of a Television System
We are now in a position to obtain a bird's-eye
Fig. 1-16.-At the left is shown the saw -tooth waveform view of a complete television system. Such a step is
required for horizontal scanning. A similar waveform, but advisable at this time because of the complexity of
of much lower frequency is required for vertical deflection
and is shown on the right. the system and the desirability of not losing sight
of the function of the principal parts in the maze
of detail associated with all the individual elements
through the deflecting coils. The nature of these of the system.
voltages and currents depends on the kind of de- Fig. 1-17 shows the general role played by each
flection to be produced. In Fig. 1-15 is shown the of the major units of a television system in causing
kind of deflection required in television scanning. the picture of the televised scene to appear on the
The trace deflection required is a steady motion screen of the picture tube. Referring to this illus-
from left to right along a line slanting slightly tration, you will note that the camera tube is focused
downward to the right; the retrace travels from on the scene to be televised with the result that an
right to left along a line slanting downward to the image of the scene is formed on the photoelectric
left. We found out that such a motion will occur mosaic of the camera tube. In this way each point
when both sets of deflecting plates or coils are in of the mosaic takes on a voltage which is propor-
operation simultaneously. Thus in scanning, two tional to the light value associated with this point
deflection circuits are acting at once, one moving the of the image. In order to transmit this information,
beam horizontally (horizontal deflection circuit), the electron beam of the camera tube completely
the other moving it vertically (vertical deflection scans the image on the photoelectric surface 30 times
circuit). in each second. As a result of this the video portion
The shape of the voltage (or current) wave which of the television signal is produced in which the
has to be applied to the deflecting plates (or coils) electrical variations correspond with the light varia-
to produce the desired scan is shown in Fig. 1-16. tions on the screen of the camera tube.
At the left is shown the waveform for horizontal In order to fulfill the requirements of synchroni-
or line scanning; at the right the waveform for zation, the need for which has already been ex-
vertical or field scanning. Notice that the time of plained, a synchronizing signal (abbreviated "sync"
the line wave is equal to the time of the field wave signal) is applied to the camera tube deflecting cir-
divided by the number of lines per field (1/60 cuits so that the scanning of the electron beam is at
second divided by 262.5 = 1/15,750 second). Also all times under the timing control of this sync sig-
notice that the line retrace is allowed only about nal.
CAMCRA
TUOC
TRANSM/TTER RECE/VER RERPOLV/CLO
V/OCO /MAGL
t V/OCO & SYNC
S/GNAL S
SIGNAL
PICTURE
LRINT FROM RAO/D V/OCO
SCENE WOCO Y/OCO
AMPL/F/ER
RAP/o
TRANS -
eCE/ E4 AMPL/P/CR
!UGLY/MN
SEAM
M/TTER l
ceFLCCT/NG
CIRCUITS SYNC Sr1C
S/QNAL JEPaRAT.^.F LL EC TROLA
VERT. BEAM
SL1CF[ELJ.
YAJCNRGVFz
/NG ACF[fCT/NG
SYNC 3/6NAL CIRCUIT
SYNC
-SIGNAL
it C/RCU/TS
7 NOR/Z
PEAL EC T.
Fig. 1-17.-The principal elements of a complete television system. Note the provision made for synchronizing the scanning
at the picture tube with that at the camera tube.
At the same time this sync signal must also form scanning beams-the one in the camera tube at the'
a part of the television signal which is broadcast in transmitting end and the other in the picture tube
order that the scanning at the picture tube in the re- at the receiving end-in perfect synchronism. In
ceiver can be kept in synchronism (in step) with the this way the image of the scene is traced out by the
scanning at the camera tube. For this reason you moving spot of light on the screen of the picture
will note that the sync circuit also feeds the same tube.
sync signal to the video amplifier, and as a result
the complete signal contains information not only A more detailed discussion of synchronizing cir-
on the light values but also the necessary control cuits is given in chapter 8. This discussion is sup-
signals to synchronize the scanning at the picture plemented by examples of specific circuits. These
tube with that at the camera tube. examples are drawn from television receivers on the
market today.
The complete television signal is amplified by the
In the above description we have omitted a con-
video amplifier circuits in the transmitter and fed
sideration of the sound broadcasting which almost
to the modulating circuits of the transmitter where
invariably is a part of the television broadcast. For
it modulates the high -frequency radio wave which
the present, it will be sufficient to understand that
serves as the carrier. According to the present fre-
the sound is transmitted and received in the same
quency allocations assigned by the Federal Com-
way as a conventional sound broadcast even to the
munications Commission, the carrier frequencies used
extent that an entirely separate carrier is used to
lie within the range between 44 and 88, and between
carry the modulation of the sound accompanying
174 and 216 megacycles. The reasons for the use
the television broadcast.
of carrier frequencies in the ultrahigh frequency
range and the actual make-up of the signal will be
THE TELEVISION SIGNAL
discussed later in detail.
We are now in a position to consider more fully
The signal which is radiated by the transmitting the nature of the signal which is used to transmit
antenna is picked up by the receiving antenna and the television image. Up to this point we have ex-
fed to the television receiver. The television signal plained in a general way that this signal contains
is amplified in this receiver which is almost invaria- the electrical image of the scene being televised and
bly of the superheterodyne type. After sufficient in addition contains the information required to
amplification the signal is demodulated in the second synchronize the scanning at the camera tube with
detector of the receiver and the video and sync that at the picture tube. We will now go into great-
signals are recovered. The video amplifier following er detail as to the structure of this signal because
the detector further amplifies the signal which is of the bearing which it has on the operation and
finally impressed on the control grid of the picture servicing of television receivers.
tube. The fluctuations of voltage on the control grid
of the picture tube cause the intensity or brightness It will be helpful in understanding the nature of
of the scanning spot to vary in accordance with the the television signal to review briefly the audio sig-
amount of light on that element in the scene which nal used in sound broadcasting and later to com-
at that particular instant is being scanned by the pare the two. In Fig. 1-18(A) is shown a typical
electron beam of the camera tube. sound or audio signal. As you know, the amplitude
of this audio signal represents the intensity (loud-
The receiver contains separate circuits for deflect- ness) of the sound, and the number of cycles per
ing the beam of electrons horizontally and vertically second represents the frequency (pitch) of the
so as to accomplish the scanning of the image at the sound. During periods when the sound intensity
picture tube. In general these circuits are similar to is zero, the amplitude of the sound wave is, of
those used to deflect the electron beam at the camera course, zero; on the other hand, during periods
tube. To insure absolute synchronization between when the sound intensity is high, there will then be
the scanning at the picture tube and that at the cam- a proportionate increase in the amplitude of the
era tube, the receiver contains circuits (called sync signal. An important characteristic which you
separator circuits) for separating the synchronizing should note is that in a sound wave, the amplitude
pulses from the complete television signal. As is of the wave has both positive and negative peaks
noted in the figure, these impulses are applied to the and extends equally in both directions from the
deflection circuits in the receiver and keep the two zero axis.
1 C, RIDER'S-"HOW IT WORKS"
The Video Signal (B), as we have seen, is of this type. If we call
the black -level voltage zero volt in Fig. 1-18 (B),
A typical video or picture signal, that is, the elec- all other shades of brightness will produce negative
trical signal which represents the variations in voltages, the whitest part of the scene having the
brightness over the elements of a scene, is shown greatest negative voltage. A signal having negative
in Fig. 1-18(B). The horizontal line in the left- picture polarity is used to modulate the television
hand portion of the signal shown represents the carrier in present-day American television. Of,
signal voltage produced by an unilluminated or course it is perfectly possible to have a video signal
black portion of the scene, which is called the "black in which the signal voltage increases as the picture
level." In discussing video signals this black level brightness increases ; in this case the signal is said
is used as a reference from which the light values to have a "positive picture polarity." In fact we
corresponding to all other signal voltages are meas- shall see later on, in considering receiver circuits,
ured. The reason for this is apparent from inspec- that the signal undergoes a reversal of polarity in
tion of the rest of Fig. 1-18(B), which represents passing through each stage of the video amplifier.
various signal voltages corresponding to parts of The important difference between the audio
the scene reflecting varying degrees of brightness. signal and the video signal is that the video signal
Note that as the brightness of the scene increases is always located on only one side of the black
from black, the voltage of the video signal de- reference level, whereas the audio signal contains
creases, and that in any case the picture brightness variations on both sides of the zero -signal level.
never results in a signal voltage greater than the The video signal is therefore a pulsating voltage
black level. The black level is therefore a suitable with a d -c component, not an a -c wave like the
reference, because its voltage is fixed and easily audio signal.
reproduced. "White" would not be a suitable ref- Let us now consider the video signal in greater
erence because the signal voltage corresponding to detail. Fig. 1-19(A) shows the output of the cam-
"whitest white" depends upon the maximum in- era tube for two successive lines of the image. At
tensity of illumination available. the same time Fig. 1-19(B) shows these two lines
A video signal in which the signal voltage de- as they appear on the scanning pattern or raster.
creases (from the black level) as the picture bright- Starting at a, the beginning of the field, the beam
ness increases toward white is said to have a traces the first line a -c; referring to the signal [Fig.
"negative picture polarity." The signal of Fig. 1-18 1-19 (A) ] it can be seen that the image is black at
L OUD
SOUND
NO SOFT
/..50(//V0 SOUND
A)
J
BLACK
LEVEL
GR AY (b
{YH/TE
LEVEL
Fig. 1-18.-Comparison between a sound and video signal. In the video signal, black is represented by a fixed level, and vari-
ous shades of brightness are represented by voltages displaced proportionately from the black reference level.
GENERAL ASPECTS OF THE TELEVISION SYSTEM 1 ï
HORIZONTAL BLANK/NG
PULSE OR PEDESTAL
L/NE 1 LINE
BLACK O BLACK
LE vEz_
GRAY
(A) LEVEL
WH/TE
LEVEL
a Fig. 1-19.-The video signal and horizontal blanking pulse for
d _e - -- f two successive lines of a scene. In Fig. 1-19(B) at left are shown
the lines of the scanning pattern corresponding to the signal. Note
9 - that the picture is blanked out during the horizontal retrace
Cg) periods, as shown at cd and fg.
a, changes gradually to a brilliant white at b, and Signal and Sync Pulses

finally changes gradually to black at the end of the
line C. In the previous section we showed the video
During the retrace time c -d, the signal level is signal without discussing the modifications which
maintained at a uniform black level as shown by must be made in the signal in order to provide the
c -d on Fig. 1-19(A) This "black" interval during necessary synchronization. We shall now describe
which the retrace is carried out is called the "hori- how this synchronizing information is added to the
zontal blanking period," and the pulse c -d is called signal.
the "horizontal blanking pulse," or the "horizontal Consider the video wave shown in Fig. 1-20.
blanking pedestal." As we shall see later in more This shows the wave for the last two lines of the
detail, the pedestal performs two functions : (1) It field, just preceding the vertical retrace period dur-
blanks out the return trace so that it will not appear ing which the beam returns to the upper portion of
on the screen of the picture tube and (2) it pro- the field. Starting at the left of the figure, the first
vides a platform on which the horizontal synchro- thing you will note is the addition of a pulse which
nizing pulse is erected. Note that the second scan- is erected on top of the horizontal blanking pulse ;
ning line d -e-f is essentially the same as the first this is called the "horizontal sync pulse." As we
line and again the line is terminated on the signal shall see later in the discussion of receiver sync
wave by a blanking pulse f-g. In this way the entire circuits, this small rectangular pulse provides the
field is scanned and the picture signal corresponding means which keeps the horizontal line or scanning
to the light and dark variations of the field is pro- oscillator in the receiver in synchronism with the
duced. scanning at the camera tube. An important thing
VERTICAL. BLANK/NG TOP OF -

BOTTOM OF PICTURE PICTURE
PER/OO
BLACK
LE EL
--- - 1
1/
LINE
7/7
PICTURE
SIGNAL CONTA/N.5 I
ES
VERT/CAL SYNC PULSES
GRAY SYNC PULSES
.EQUALIZING PULSES
WHITE
LEVEL --
-HOR/ZONTAL
SYNC
PUL 5E5
J H= 7/ME FOR 7L/NE ---/
/5,750
- = 63.5 MICROSECONDS
Fig. 1-20.-The video signal showing the last two lines of a field followed by a vertical blanking period. Horizontal sync
pulses are shown on the horizontal pedestals and the vertical sync pulse is transmitted during the vertical blanking period.
to note is that this sync pulse is located in the izing pulses both preceding and following the verti-
"blacker -than -black" region so that the screen of cal sync -pulse interval. After the last equalizing
the picture tube is kept dark during the period of pulse, the horizontal sync pulses are resumed ; by
the horizontal retrace. Thus the portion of the sig- this time the beam has been returned to the upper
nal more positive than black is used for synchro- portion of the screen so that shortly thereafter the
nizing information, whereas the voltage more neg- normal video signal is resumed. To summarize, the
ative than the black level is used for the picture in- first line (A) shows the complete signal as it exists
formation. It is also important to observe that this for any one field both before and after the trans-
line -synchronizing pulse appears at the end of each mission of the vertical blanking period.
line so that constant synchronization of the line Part (B) of this figure describes the signal as it
oscillator is maintained. exists 1/60 second later for the following field.
Since the scanning is interlaced, note that the line -
At the end of the last line at the bottom of the sync pulses in (B) appear between the line -sync
field, represented by point 1 in Fig. 1-20, the beam pulses in (A), thus providing the timing which is
is ready to return to the upper edge of the field. essential for interlacing. Again the last line in this
As in the case of horizontal scanning, a sync pulse field is followed by a vertical blanking period at the
is required to return the beam to the top at the end of which the video signal is resumed. Note that
proper instant. All the information associated with for both parts of the figure, the reference point
the end of the field, required to return the beam from which time is reckoned is the beginning of
to the top of the field in preparation for the one the vertical sync pulse, which is designated as tak-
to follow, is contained in the interval designated as ing place at any time represented by t = ti. Using
the "vertical blanking period." Essentially the fol- this time reference, it of course follows that the
lowing three functions are performed during the vertical sync pulse for the next field must begin
vertical blanking period : (1) A field synchronizing 1/60 second later (since there are 60 fields per
pulse is provided (the exact nature of this pulse second) ; this is shown by part (B) of the figure so
is described later) so that the beam will be return- that the two vertical sync pulses are directly below
ed to the start of the frame at the proper instant. each other but 1/60 second apart in time.
(2) The entire signal is blanked out so that the The description which follows shows in greater
field retrace and lines scanned during this interval detail those parts of the complete video signal
will not be apparent to the observer. (Actually, of which have already been described.
course, the line -scanning circuits continue to func-
tion, but the beam does not exist because of the
negative voltage on the control grid of the picture Horizontal Blanking and Synchronization
tube during this interval.) (3) The line synchro- As shown in Fig. 1-21, the horizontal sync signal
nizing pulses are maintained during the vertical is transmitted at the end of each line and consists
blanking period, which lasts for about 15 lines, so of an essentially rectangular pulse erected on the
that the horizontal deflecting circuit in the receiver horizontal blanking pedestal. The amount of time
will not slip out of synchronism during this period. allowed for the blanking pedestal is specified as
In addition to the vertical and horizontal sync 15% of the total time from the beginning of one
pulses, two groups of so-called "equalizing pulses" line to the beginning of the next line. Since the
are transmitted during the vertical blanking period ; time for each line (including the retrace) is 63.5
these are required for reasons which will be ex- microseconds (1/15,750 second), the time devoted
plained later. to horizontal blanking is about 9.5 microseconds.
This interval of 9.5 microseconds has been found to
Standard Television Signal be just large enough to allow for the retrace time,
to allow the spot to assume normal scanning speed
It is desirable at this point to show the complete at the left edge of the picture, and to maintain re-
signal which is used as the standard in this country. liable synchronization.
To illustrate the make-up of this signal, Fig. 1-21
shows the signal for two successive fields in the The whole of the horizontal blanking interval is
neighborhood of the vertical blanking pulse. Ac- not utilized for the synchronizing pulse as can be
cordingly the left-hand portion of (A) shows the seen by examining the enlarged view of the wave
last four lines of any one field. This is followed by between C -C [Part (B)] shown in the detail view,
the vertical blanking period which contains equal- Part (C) of the figure. Actually only about half the
total blanking time is used, and the front or leading ceding the vertical sync pulse are identical for alter-
edge of the sync signal is placed as close as possible nate fields. The need for these equalizing pulses
to the beginning of the blanking pulse. The small arises because of the interlacing of alternate fields.
allowance which is made takes care of some varia- As you can see from Fig. 1-21, the lines in the
tion in timing and insures that the sync pulse will second field (B) are interlaced with those of the
not run into the video portion of the signal and preceding field (A). If the equalizing pulses were
thus upset the line timing. eliminated and the vertical sync pulses were insert-
ed in (A) at the end of the last line, then the verti-
Vertical Blanking and Synchronization cal sync pulse would have to appear in the next
field (B) at the middle of the line; the reason for
The vertical blanking interval follows the last
this is that 1/60 second later the beam is in the
line of each field and consists of the following four
middle of the line because of the interlacing. Thus
parts which will be considered separately.
without the equalizing pulses, the conditions pre-
(1) Six equalizing pulses one-half line apart ceding the vertical sync pulse would be different for
precede the sync pulse and accomplish (a) the each of the two fields. This would tend to produce
maintenance of horizontal or line synchronization a different type of vertical sync pulse for alternate
and (b) the "equalization" of the intervals preced- fields, upset the synchronization and give rise to
ing the vertical sync pulse so that conditions pre - the distortion known as "pairing of the interlace."
525 LINES. JO FRAMES PER SEC., 60 FIELDS PER SEC.. INTERLACED
NORrZONTAL SYNC
[O:ALIPNG
YT[RVAL
x ULxS
SOT TOM Of NKTURE
.--TOP OF PKTORE
.5N W%MlY CARNIER qXTK(
UIJLJLfl 1 0
SYICK vu or uA%. EANRER
.KTURE LAKIMG
(A )
L- OTV MIN (SEE MOTE H
PKYLRE-I RTKAL BLANKMG (c%
O v wcN
11
NOMzO.ITAL
10. ANKING l INE
SCC
I.III1^(I
n nn
I
nn
1I 1
B)
A & R.SHOW BLANKING & SYNCHRONIZING SIGNALS
IN REGIONS OF SUCCESSIVE VERTICAL BLANKING
PULSES.INORIZONrL DIMCNSONS NOT TO SCALE. ALL OUCN-
!ORS ARC IR. BLACK LEVEL UNLESS or.CRNIx SPCCAIKALLV
INOICATE01
VCJITCAL NIaN
MOR.
MOON. &MN ,rI .0058 WAX
AK r- A«
11&11q( ' , PICTURE MOR STNC
L
ICON BLANONO
REPETITION RATE
15,750 NCR 5CC
(c)
DETAIL BETWEEN C -C
OR4N3AINS W SCALE
(-DIAGRAM C SHOWS ENLARGED DETAIL VIEW OF SIGNAL IN VIEW a BETWEEN LINES C -C

2- DIAGRAM D SHOWS ENLARGED DETAIL VIEW OF SYNC SIGNAL IN VIEW A DETWEEN LINES O-0
3 HTIME FROM START OF ONE LINE TO START OF NEXT LINE.I/IS,750 SEC.

4. V. TIME FROM START OF ONE FIELD TO START OF NEXT FIELDI/Ou SEC..162.SH
5 LEADING AND TRAILING EDGES OF BOTH HOR. AND VERT. BLANKING PULSES HAVE
SLOPES (NOT INDICATED IN A&) WHICH SHOULD BE KEPT AS STEEP AS POSSIBLE
6 -RECEIVER VERTICAL RETRACE SHALL BE COMPLETE AT END OF .07 V
ED) DETAR. BETWEEN D-O

Minima TO SCl1LE
Fig. 1-21-The standard television signal. Part (A) shows the signal at the end of any one field; part (B) shows the sig-
nal at the end of the next field, 1/60 second later. The difference between the two fields is caused by interlacing.
In a paired interlace, the even-scanned lines do not the vertical sync -pulse interval are the same for
, lie midway between the odd -scanned lines, because both fields. Note that the lagging equalizing pulses
of the difference in timing on alternate fields. are also one-half line apart so that line synchro-
In connection with the maintenance of line syn- nization is maintained for both the "odd" and
chronization during the vertical blanking interval, "even" fields.
note that the leading edges of the equalizing pulses (4) The lagging equalizing pulse interval is
function to maintain synchronization. Not all the terminated before the end of the vertical blanking
pulses are used for each field, however. Thus note period so as to prepare the line oscillator for the
that because of the interlacing, the first, third, and normal horizontal sync pulses which are to follow.
fifth equalizing pulses are used on the first field In practice, the video signal is blanked out for a
(A), and the second, fourth, and sixth pulses are period of from 7 to 12 lines following the last
used on the succeeding field (B). This explains equalizing pulse so that the line oscillator (which
why six pulses are used, each spaced one-half line may have been operating at double line frequency
apart, rather than three pulses spaced one line during the preceding period) has a chance to settle
apart. It would, of course, be possible to use three down to being under control of the normal type
different pulses in each field, but if this were done of sync signal. At the end of the vertical blanking
the signal preceding the vertical sync pulse would interval, the blanking is of course removed and the
be different for succeeding fields and there would be video portion of the signal again controls the in-
a resulting absence of equalization. tensity of the beam in the picture tube.
(2) The vertical sync pulse (or field synchro-
nizing pulse) follows directly after the equalizing Range of Frequencies in Video Signal
pulse interval and consists of six broad pulses in
which the edges are serrated or cut at one-half Unlike audio signals which contain frequency
line intervals. The function of the vertical sync components ranging from a low value of about 20
pulse is to provide the control signal which tells cycles per second to a high value of about 15,000
the vertical oscillator that it is time to begin cycles per second, video signals include a range
the retrace and thus to return the beam to the top from practically zero frequency (produced over
for the beginning of the next field. The pulses in areas where there is little variation in light inten-
the vertical sync -pulse interval are considerably sity) to as high as 4 or more megacycles (pro-
broader than the line pulses so that the sync separa- duced over areas where there is a very rapid varia-
tor circuit will be able to distinguish between the tion in light intensity).
two types of pulses and thus be able to separate the It is interesting to examine the manner in which
vertical sync pulses from the horizontal sync pulses. the maximum frequency required in the video sig-
At the same time the edges of the serrations at half- nal is related to the amount of detail which is re-
line intervals provide the necessary control for produced. This can be arrived at from the follow-
maintenance of horizontal synchronization. The ser- ing considerations : Let us assume that we wish to
rations are required at half-line intervals because transmit a picture in which the same resolution or
of the interlacing; the reasoning used in connec- detail is desired in the horizontal direction as in the
tion with the equalizing pulses also applies here. vertical direction. In the vertical direction we have
(3) It was explained previously that in order to a total of 525 lines or 525 elements. Since a hori-
provide identical conditions for the two successive zontal line is 4/3 times as long as a vertical line
fields preceding the vertical sync pulses, six equal- (the picture is 4/3 times as wide as it is long), it
izing pulses were inserted in front of the vertical follows that there are 4/3 as many elements in a
sync pulse in each field. It is just as necessary to horizontal line as there are in a vertical line. This
keep the conditions following the vertical sync - makes a total of 525 x 525 x 4/3 or 367,500 ele-
pulse interval the same for the two successive ments in the complete picture.
fields (A) and (B) for this reason six lagging
; Since we wish to calculate the maximum fre-
equalizing pulses appear in both (A) and (B) after quency which is required to reproduce the light
the vertical sync pulse. If you examine the vertical and dark variations over each one of these 367,500
sync -pulse interval in both (A) and (B) you will elements, let us assume that alternate elements in
see that, although the lines in the two fields are dis- the picture are black and white so that the image
placed by one-half line because of the interlacing, resembles a checkerboard pattern. This type of
nevertheless the conditions in the neighborhood of scene requires the highest possible frequency for
faithful reproduction because of the rapid iriation

* a 1000-kc carrier with an audio signal which con-
in light intensity as the beam goes from one ele- tains frequency components ranging up to 5 kc,
ment to the next. The exact opposite of this type then the resulting modulated wave contains, in ad-
of scene would be one which was uniform over its dition to the carrier frequency, new frequencies
entire area, for in this case it would be necessary which extend to the limits of 5 kc below the carrier
for the system to transmit only very low frequen- frequency and 5 kc above it. In other words, the
cies. For the checkerboard pattern under discussion, process of amplitude modulating the carrier results
the variation from black to white in scanning two in the introduction of two sets of sidebands which
adjacent elements requires a certain amount of time. extend outward from the carrier to a value equal
This amount of time represents the duration of one to the highest frequency in the modulating wave.
complete cycle and one divided by this number repre- The sideband which contains the frequencies lower
sents the highest frequency which must be trans- than the carrier frequency is called the "lower side -
mitted. band," and that which contains the frequencies
Let us calculate this time interval required for higher than the carrier is called the "upper side -
the beam to scan two adjacent black and white ele- band."
ments. Since there are 367,500 elements in a com- It would be impossible to use a broadcast -band
plete picture and it requires 1/30 second to trans- carrier for television work. This can be seen from
mit this picture, the time allotted to the transmission the fact that unlike sound, the video modulating
of information on two elements (1 cycle) is equal frequencies would themselves be higher than the
to carrier frequency. In order to make the modulation
process work, it is necessary that the carrier fre-
2\72- 1 second 1 second
quency be at least several times the highest fre-
367,500 30 cycles 3,900,000 cycles
quency of modulation. For this reason the carrier
The .constant, -\/2, is determined by several factors, frequency in television must be several times as
the most important of these is the spot diameter. high as the maximum video frequency, or several
The frequency generated when the scanning beam times 4 mc. Actually the carrier frequencies which
passes over these two elements is thus equal to have been chosen for television are at least ten
3,900,000 cycles per second, or 3.9 megacycles. times the highest video frequency since the lowest
As we have previously seen, the maximum fre- carrier frequency which is being used for 525 -line
quency which is present in the video signal is related television is 44 mc.
directly to the amount of detail required in the im- Some of you may recall that a number of years
age. It has been, found that for the smaller picture ago, carrier frequencies only a little higher than the
tubes satisfactory detail is obtained when frequency
variations up to 2.5 megacycles are transmitted and SOUND
ENVELOPE
that little is gained by transmitting the high fre-
quency components produced in the scanning at the
camera tube. However, where a comparatively large
picture tube is used, additional detail and a finer
image can be produced by transmitting frequency
995 /000 /005
components ranging up to about 4 megacycles. RC
CARRIER
The Modulated Wave V/pEO

ENi ELC.AE
L C/vER uPPEa
In previous sections we have described the video 5/pEdANp 5/PESANO
wave produced when a scene is scanned and the

modifications which are made in this wave in order
to provide for both line and field synchronization.
We now will consider the make-up of the modu- 40 44 a
MC
CARRIER
lated wave which is produced when this video wave ¡CARR/ER LINE SYNC
modulates a high -frequency carrier. FREQUENCY 4.1/LM (B)
First yet us review the result of amplitude mod-
ulating a carrier with a conventional audio signal. Fig. 1-22.-The sidebands in an a-m sound signal occupy
a bandwidth of only about 10 kc, whereas the sidebands in
As Fig. 1-22(A) shows, if we amplitude modulate a television signal require a bandwidth of about 8 mc.
broadcast -band were assigned to experimental tele- lation we run into somewhat similar terms-posi
vision. This merely illustrates our point, however, tive and negative modulation-which are related
since those early low -definition pictures contained not to the polarity of the video signal but to the
comparatively few elements and consequently had modulated wave itself. A television carrier is said
maximum video frequencies far below those found to have positive modulation when an increase in
in present-day high -definition work. carrier amplitude corresponds to a brighter area in
Returning to the comparison between a broad- the scene being scanned. Thus for a wave with
cast-band carrier amplitude modulated by a sound positive modulation, the lowest carrier amplitude
wave and a modulated television carrier, we show corresponds to black while the maximum carrier
in Fig. 1-22(B) the wave which results when amplitude corresponds to the brightest part of the
a 44 -mc carrier is modulated with a video signal. image. On the other hand, for negative modulation,
As in the case of the 1000-kc carrier, the process a decrease in carrier amplitude corresponds to an
of modulation introduces two sets of sidebands, and increase in the brightness of the image. Thus for a
for the example shown the two sidebands extend wave with negative modulation, the lowest carrier
to 4 me below and 4 me above the 44-mc carrier. amplitude corresponds to maximum white in the
It is interesting to compare the bandwidth re- image and the highest carrier amplitude corresponds
quired for the transmission of a scene by television to black.
with the bandwidth required in amplitude modu- Because negative modulation offers certain ad-
lated (a -m) sound broadcasting. As Fig. 1-22 vantages in improved performance and simplified
shows, an a -ni sound broadcast requires only a 10- receiver design, it is used as standard in this coun-
kc channel whereas the television channel (arranged try. As shown in Fig. 1-23, the maximum amplitude
for double-sideband modulation) requires 8000 kc of the carrier is used for the synchronizing pulses
or 800 times as much space as the sound channel. and lies in the blacker -than -black region. The term
Although it is possible to locate approximately 100 "blacker -than -black" merely means that the sync
sound channels in the broadcast band, it would re- signals have higher amplitude than black picture
quire more than 8 times the space provided by the signals. It has been found that reliable synchroni-
entire broadcast band for the transmission of a sin- zation can be secured so long as the amplitude of
gle television channel with double-sideband mod- the synchronizing pulses is from 20 to 25 percent
ulation. The large amount of the radio spectrum of the maximum carrier amplitude. Actually, then,
required for television is one of the reasons for not more than 80 percent of the total carrier ampli-
the choice of the ultrahigh frequency range for tude is available for transmitting information on the
television. light values in the scene, the rest of the wave being
used for synchronization.
Positive and Negative Modulation
In discussing video signals we pointed out that RECEIVER CIRCUITS: GENERAL
the video signal is said to have a positive or nega- Having examined the fundamental principles of
tive picture polarity depending upon whether the television, let us now investigate the operation of
changes from the black level take place in a posi- receiver circuits. These circuits are especially im-
tive or a negative direction, respectively. In modu- portant because the work of servicemen in the field
SYNC deals primarily with the installation and main-
PULSO
tenance of receivers. In order to show the interre-
1 20-26 lationship between the many components that make
T- --BLACK LEVEL up a receiver, we shall first break down the receiver
,00 into its major sections and later consider the func-
ifibitaid odi NH/TE LEVEL.
tioning of these in more detail.
Fig. 1-24 shows a block diagram of a typical
10" 91
1-«-f LINE --.
television receiver arranged to show the general
character of the signal and the function performed
by each section. For convenience we shall assume
that the receiver is tuned to the 44-50 me channel.
Fig. 1-23.-A video wave with negative modulation. The In accordance with the preceding description, this
sync pulses are located in the blacker -than -black region and means that the frequency of the video carrier is
occupy from 20 to 25 percent of the maximum carrier
amplitude. 45.25 me (1.25 me above the low-frequency end of
the channel) whereas the frequency of the audio which is broader than usual to minimize the effects
carrier is 49.75 mc (0.25 mc below the high -fre- of oscillator drift), and is demodulated at the sound
quency end of the channel) . detector. The avc voltage is supplied in the usual
Both these signals, together with their sidebands, manner to control the gain of the stages in the
are picked up by the antenna and fed through a sound i -f amplifier. In accordance with FCC stand-
transmission line to the input of the r -f amplifier. ards, frequency modulation (f.m.) is used exclu-
Essentially the function of the r -f amplifier is the sively for all television sound.
same as that of the r -f amplifier in any superhet- Returning to the video signal, we have seen that
erodyne receiver-to amplify the signal and to re- a 25.75 -mc i -f signal is produced by the mixer and
ject unwanted signals in adjacent and other chan- that this signal carries the video modulation. As the
nels. In this case, the r -f amplifier is broadly tuned diagram shows, this signal is amplified in the video
so that both the video and sound carriers, which are i -f amplifier, which usually consists of several
separated by 4.5 mc, are amplified equally. stages, and finally reaches the video detector where
After being amplified in the r -f amplifier, both the signal is demodulated. The video signal recov-
signals are fed to the mixer circuit where the con- ered at this point is essentially the same as the out-
version of the signals to the intermediate f requen- put of the camera tube, so that it contains all the
cies (i.f.) takes place. Since there are two radio fre- information required to reproduce the picture, and
quencies, it of course follows that two separate in- in addition, includes the blanking and sync pulses.
termediate frequencies are produced. The video detector is followed by the video ampli-
In accordance with present practice, the oscilla- fier which, in terms of a sound receiver, corresponds
tor operates at a frequency approximately 26 mc to the audio amplifier. The function of the video
above the video carrier frequency. For the channel amplifier is to amplify the video signal so that its
being received, the frequency of the oscillator in the amplitude will be great enough to "swing" the mod-
receiver is 71 mc. Since the oscillator frequency is ulation grid of the picture tube. For the average
25.75 mc above the video carrier frequency, in this picture tube this requires approximately 25 volts,
case, 25.75 mc will be the value of the video i.f. In peak-to -peak.
the same way, the i.f. of the sound signal is equal Note in the diagram that the polarity of the video
to the difference between the oscillator frequency signal is reversed 180 degrees for a single stage of
and the sound carrier frequency, 71-49.75 mc, or video amplification and that the receiver is arranged
21.25 mc. so that the signal which reaches the control grid of
Following the mixer, the sound channel in the the picture tube has a positive polarity. As a result
block diagram is entirely independent of the rest of the synchronizing impulses appear in the blacker -
the receiver and in practically every detail is simi- than -black (highly negative grid -bias) part of the
lar to a conventional receiver used for broadcast picture -tube characteristic so that the beam is block-
reception. Thus the 21.25 -mc sound i -f signal passes ed during the retrace part of the line and field
through the sound i -f amplifier (the selectivity of sweeps.
SOUND CHANNEL
RECC/eAEA
uNE o TO
04 -so. SOUND SOUN0 AUD.O
Cw'Aww1cL 2 No LOUOSPEAKER
' ANTCNN.aE
1/
AMP(. OC T. V AMPL.
RF
59YA10
2Y.?fMC
097lMC
AVC
CET.
V/DEO VIDEO CHANNEL

45. 25MC
V/OED y 0E0 V/ Oe 0
AMPL . °.,
E7. -11\7- Ah/PL .
TO 4 MC TO 4 MC
25.75MC
/5.7LLE
OSC. CCYC50SS
N.C./Z
7/ rrC
AVC SYNC A n
/5, 750
A PEFL. vv
SEPAR- CCT.
ATOR CYCLES
V Y
TO 4 MC CO CYCLES
60 CYCLES
SEPT
OEFL.
CCT
Fig. 1-24.-A block diagram of a typical television receiver. Note the changes in the signal as it passes through the receiver.
al&
In addition to supplying the video signal and the throughout the receiver with the exception of the
signal which actuates the avc system, the detector high -voltage requirements for the picture tube. The
supplies the video signal to the synchronizing sep- latter, which may include voltages as high as 30,000
arator. The purpose of this separator is to remove volts, is supplied by a separate high -voltage power
the picture component from the complete video sig- supply which has its own transformer, rectifier, and
nal and then to separate the horizontal sync pulses filter.
from the vertical sync pulses. As is shown, the The two most popular high -voltage power sup-
horizontal sync pulses are arranged to control the plies in use at the present time are the "kick -back"
timing of the horizontal deflection circuit, while the power supply and the r -f power supply. Television
vertical sync pulses are arranged to control the tim- power supplies are covered in chapter 11.
ing of the vertical deflection circuit. And now, having discussed the fundamental prin-
The power supply is not shown in the block dia- ciples of television here, detailed coverage of the
gram. A single low -voltage power supply can be various television receiver circuits will be given in
used to take care of all voltage requirements succeeding chapters.
CHAPTER 2
FREQUENCY CHARACTERISTICS OF THE TELEVISION SYSTEM

BY SEYMOUR D. USLAN
In order to understand many of the features of CHANNEL FREQUENCY CHANNEL FREQUENCY

television it is very important to know something
about the frequency standards concerning televi-
NO.
1 -
(MC)
44
54-60
50
NO.
7 174 --
(MC)
180
sion broadcasting as set up by the Federal Com-
muications Commission (FCC). Such information
2
3 60-- 66
8
9
180
186 --
186
192
will be very helpful because in the discussions to
follow reference is quite often made to these stand-
4
5
66
76 -- 72
82
10
11
192
198 --
198
204
ards. The frequency range of television, the differ-
ent channel frequencies, the bandwidth involved,
and the video and sound carrier frequencies are
6 82 88 12
13
204
210 - 210
216
other group, usually called the high band. Such a
some of the topics which will be discussed in this grouping is readily understandable when one con-
chapter. siders the 86 -mc difference between the end of the
sixth channel and the beginning of the seventh. Such
Television Channels a grouping is helpful in analyzing image frequency
interference and other problems.
Television today occupies quite a large frequency
range. ,The FCC has set aside 13 television channels Video Signal Characteristics
in the frequency spectrum between 44 and 216
megacycles. These 13 television channels do not When a (high frequency) carrier is modulated
exist over this complete frequency range but only with a video signal, the carrier undergoes changes
over part of it. The term "television channel" ac- in amplitude. Thus we see that television is an-
cording to the FCC means-a band of frequencies other form of amplitude modulation and possesses
6 megacycles wide in the television broadcast band many of the characteristics. of regular a -m radio
and designated either by number or by the extreme broadcasting. In this latter type of broadcasting,
lower and upper frequencies. This means then for when a carrier signal is modulated by audio the re-
the 13 television channels, the television broadcast sultant modulated signal consists of a center fre-
band occupies 13 x 6 or 78 mc. However, this 78 -mc quency component, equal in frequency to the carrier,
band is not continuous as are the f -m and a -m com- and two sideband frequency components. One side -
mercial broadcast bands. band component is called the upper sideband and
From the 44 -to -216 -mc frequency range the fol- the other is called the lower sideband. The frequen-
lowing frequencies are not part of the television cies of these sidebands are equal to the center fre-
band 50-54 mc, 72-76 mc, and 88 to 17.4 mc. The
: quency plus or minus the frequency of the applied
other frequencies in this 44 -to -216-mc frequency audio modulating signal. The relative amplitudes of
spectrum are assigned to the television band. These the components of the a -m signal are determined by
remaining frequencies are divided into 13 tel- the percentage of modulation. The frequency sep-
evision channels and each channel is designated by aration between the components is determined by
a number. In other words the channels are number- the frequency of the audio signal. In a -m radio
ed 1 through 13. Consequently with the width of broadcasting (550 to 1500 kc) the adjacent channel
each television channel standardized at 6 mc, the separation is 10 kc as set down by the FCC. This
frequency ranges of the television channels are shown means that there is a maximum of only 5 kc on
in the accompanying table. either side of an assigned station frequency.
These television channels are often divided into In regular a -m broadcasting it is possible to use
two separate groups. Channels 1 through 6 are con- audio modulating frequencies as high as 15 kc. The
sidered as one group, often termed the low band, frequency of the audio modulating signal deter-
and channels 7 through 13 are designated as the mines the bandwidth of the transmitted signal.
25
However, since the maximum bandwidth as set rier plus or minus the frequency of the video sig-
down by the FCC is equal to 10 kc, then the maxi- nal. All the video intelligence is vested in the side -
mum audio frequency that can be passed is 5 kc. bands and none in the center frequency component.
With this much understood we can illustrate the Conditions for the percentage of modulation in tele-
frequency spectrum of a typical a -m wave (audio vision are the same as in regular a -m broadcasting.
modulated) with regard to both amplitude and fre- The video frequencies are as low as 30 cycles
quency. In Fig. 2-1 we have illustrated the fre- and as high as 4 mc. This represents quite a prob-
lem for the design of proper circuits to pass such
Fig. 2-1. - The
frequency spectrum
a band of frequencies. The frequency spectrum pic-
ture of the video -modulated television signal for
of a typical 100 -
RC
percent-modulated 100 -percent modulation is illustrated in Fig. 2-2. In
a -m wave. The am- this drawing Fc is the carrier frequency, FL the
plitude of the side -
FL FC FU bands is half that lower sideband, and Fu the upper sideband. The
FREQUENCY of the carrier. maximum bandwidth is shown to be equal to 8
mc. This video bandwidth plus that of the sound
quency spectrum of a 100 -percent -modulated a -m f -m signals would take up too much space in the
wave. In a 100-percent a -m wave, the amplitudes assigned frequency spectrum for television. As
of the upper and lower sidebands each is equal to
half that of the carrier component. In Fig 2-1 the Fig. 2-2.- The
frequency spectrum
frequency of the lower is designated as fL and that of a television sig-
nal at 100 -percent
of the upper sideband fu, and for maximum audio modulation. Note °^ e--
transmission they are shown as being 5 kc on either that each sideband
is spaced 4 mc FL
side of the center frequency component which is from the carrier. FREQUENCY
-
designated as f0. If the frequency of the carrier is

mentioned, according to the FCC, each television
1000 kc, then for a 5-kc audio modulating signal
channel occupies an over-all bandwith of 6 mc.
the upper and lower sideband frequencies are re-
This includes the video -modulated signals and the
spectively 1005 kc and 995 kc. Thus an over-all
sound f -m signals.
bandwidth of 10 kc exists.
The intelligence (i.e. audio signal) is completely
Vestigal-Sideband Transmission
vested in the sideband components, with each side -
band containing half of the power of the transmit- Since the maximum 8 -mc bandwidth of the video -
ted intelligence. (The center frequency component modulated signal is in itself too broad for the
contains no intelligence at all.) Upon demodula- assigned television channels, some form of adjust-
tion of the a -m signal, each sideband contributes to ment must be made. We have previously stated that
the amount of audio power output. If only one side - if only one sideband of an audio a -m signal were
band were detected, all the audio frequencies in- transmitted, the primary difference in the receiver
volved at the transmitter would be reproduced but output would be in the amount of power. In other
only half the audio power output would be obtained words all the audio modulating frequencies would
as compared with that resulting when both side - be reproduced but with half the power they would
bands are detected. Consequently we can see that have under normal transmission (double-sideband
whether one or both of the sidebands are detected transmission).
all the audio frequencies will be reproduced and Such a system of single-sideband transmission is
only the power output at these frequencies will be also possible in television. The maximum bandwidth
different. necessary for the video -modulated signal would then
In video signal transmission the situation is very be only 4 mc-half that of the 8-mc bandwidth
much the same regarding the intelligence and the under double - sideband transmission. Single - side -
sidebands. In television transmission, the picture band transmission is not used in television because
signal (i.e. video signal) is similar to the audio sig- it may cause distortion in the output of the detector
nal in a -m transmission. The video signal is, there- system of the receiver and also because it is diffi-
fore, used to amplitude modulate a high -frequency cult to obtain the necessary sharp cutoff filter for
carrier signal. The resultant modulated signal con- complete suppression of one sideband. Since the
sists of a center frequency component and two side - television channels are 6 mc wide and since one side -
bands, an upper and a lower. The frequency of the band occupies a width of only 4 mc, approximately
sidebands will be equal to the frequency of the car - 2 mc (discounting the small bandwidth required
FREQUENCY CHARACTERISTICS OF THE TELEVISION SIGNAL 27
for the sound f -m signal) will not be used if single- one that is partially suppressed. The upper fre-
sideband transmission is employed. By utilizing the quency sideband will be at a frequency limit of
the extra bandwidth we can transmit a television 4 mc plus the video carrier signal and for channel
signal that has one complete sideband plus part of number 1 this will be 4 plus 45.25 or 49.25 mc.
the other. Such transmission is often called vesti- Thus, from Fig. 2-3, this upper sideband is seen to
gal-sideband transmission, and it is the type em- fall well within the frequency response of the
ployed in television today. channel without infringing upon that part reserved
Vestigal-sideband transmission (sometimes called for the sound f -m signal which will be at a maxi-
sesqui-sideband or quasi-single-sideband) refers to mum of 50 kc wide. Consequently we see that in
an a -m wave where one sideband is partially sup- the vestigal-sideband transmission of a video -mod-
pressed and the other sideband not suppressed at ulated signal, it is the lower sideband which is
all. The suppression of the one sideband is such
partially suppressed without suppression of the up-
that the transmitted video -modulated signal in con- per sideband.
junction with the sound f -m signal occupies a band- Examining the video part of the frequency re-
width that just conforms to the 6 -mc television sponse characteristic of Fig. 2-3 brings a few perti-
channel bandwidth. nent facts to light. It is the high -frequency end of
To understand the character of the transmitted the lower sideband that is suppressed and the low -
f requency end that remains. (Those parts of the
video -modulated signal. as far as its bandwidth is
concerned, let us refer to an ideal television chan- sidebands closest to the video carrier signal con-
nel response as seen in Fig. 2-3. This response tain the low video frequencies. This means that in
this type of vestigal-sideband transmission the high-
Fig. 2-3.-Ideal re- er frequency video signal components are weakened,
sponse curve for tele-
VIDEO CARRIER SOUND CARRIER
vision channel 1. The which effectively means that the lower frequency
curve is 6 mc wide, components are strengthened. Much of the video
with the video carrier
1.25 mc away from energy is located primarily in the low-frequency
44 4525 Ì SO
the lower end of the signals, These low -frequency components of the
4575 band and the sound
FREQUENCY (MC) carrier 4.5 mc above video signal determine the background of the re-
the video carrier. ceiver picture detail.
curve is assumed to be that for channel number 1, Operating Bandwidth Characteristics

although it will hold for all the other channels
with the proper upper and lower frequency limits. The bandwidth characteristics of Fig. 2-3 are
This curve represents the ideal characteristic of the ideal and in practice the straight line sides of the
transmitted television signal as well as that of the response curve are not obtained. In the television
r -f section of the television receiver. According to transmitter, special filter networks are employed
the FCC, the sound carrier shall have a frequency to suppress the lower frequency sideband and the
0.25 mc lower than the upper frequency of the side of the response curve that results is not a
channel in question. This means that for channel vertical line as was shown in Fig. 2-3. It is quite
number 1 (44-50 mc) the sound carrier frequency difficult and expensive to design a filter that has
would be 49.75 mc. The FCC also stipulates that a straight vertical line cutoff characteristic. For sin-
the video carrier frequency shall be 4.5 mc lower gle-sideband transmission without any frequency
in frequency than the sound carrier frequency. This discrimination of the sideband passed and likewise
means that for channel number 1 the video carrier without passing any part of the suppressed side -
would be at a frequency of 45.25 mc. Since each band, a very sharp cutoff filter would have to be
channel is 6 mc wide and we have already used employed. With vestigal-sideband transmission in
4.75 mc of it (from 45.25 to 50 mc) all that re- television, such a sharp cutoff filter is no longer
mains is 1.25 mc, which is at the low frequency necessary.
end of the channel. For channel number 1 this 1.25 - In the television system we have to deal with a
mc limit is between 44 and 45.25 mc. number of frequency response curves to understand
Since a double-sideband video -modulated signal fully the different bandwidth characteristics of
will have a total bandwidth of 8 mc with each side - the system. There is the over-all video -modulated re-
band having video frequencies up to 4 mc, then sponse of the video or so-called television trans-
from the ideal response characteristic of Fig. 2-3 mitter, the response of the sound f -m transmitter,
we see that the. lower frequency sideband is the the response of the input or r -f section of the re-
ceiver, and the over-all video response of the re- 5.25 -mc point and 'the 5.725 -mc lower frequency
ceiver. This last frequency response characteristic limit of the f -m signal is 0.475 mc. This separation
includes that of the video i -f stages as well as the is sufficient to prevent any interaction between the
r -f stage. For most practical purposes the response limits of the video and sound signals. The distance
of the r -f section of the receiver should be a com- between the 5.775 -mc upper frequency limit of the
bination of the video -modulated and sound f -m f -m signal and the 6 -mc limit of the television chan-
transmitted signals. nel is 0.225 mc. This separation is often referred
The sound f -m signal itself occupies a very small to as the guard band of the television channel. This
part of the 6 -mc channel. Its maximum bandwidth guard band is intended to prevent any interaction
is 50 kc (25 kc on either side of its center fre- between adjacent television channels.
quency) for 100 -percent modulation. Let us refer
to Fig. 2-4 which is an over-all picture of the band-
width of a 6 -mc channel which takes into account
the video -modulated and sound f -m signals. This
bandwidth characteristic is illustrated for 0 to 6
mc so it will be applicable to any channel. To use
it for any channel all that must be done is to apply
the proper frequency limits. The left-hand side of
the curve illustrates the video response and the
right-hand side shows that of the sound f -m re-
sponse. The video response is seen to have sides
0.475 mc 4
that are sloping rather than straight lines. The up- 25
g+o.225MC-0..
KC
per video frequency sideband which is a maximum FREQUENCY (mc) EACH
of 4 mc wide lies in the channel between 1.25 and Fig. 2-5.-An enlargement of the upper frequency end of
5.25 mc. The sound f -m signal has a maximum the television channel.
bandwidth of 50 kc (25 kc on either Side of its cen-
ter frequency) and thus occupies only a small por-
tion of the channel. The video modulation frequencies should be so
The 0.5 -mc separation between the sound carrier reproduced in the television receiver that they have
and the upper limit of the video frequency side- the same relative amplitudes as when modulating the
band is enough separation to prevent any overlap- carrier at the transmitter. This is important so that
ping of the video and sound f -m signals. This is the reproduced picture at the receiver takes on the
better illustrated in Fig. 2-5 which is an enlarge- same appearance as the scene at the studio. In
ment of the upper frequency end of the television vestigal-sideband transmission, however, the low
channel. The frequency separation between the video frequencies are transmitted double sideband,
and upon reproduction these low video frequencies
VIDEO SOUND
will consequently be stronger than the high video
CARRIER CARRIER- 50KC frequencies. In order that the video frequencies of
the received signal be a faithful reproduction of
those at the studio, the receiver incorporates with-
in its video channel correction circuits in which the
low video modulating frequencies are corrected for
proper reproduction.
0.5 1.25 5.25 5.75 6.
The low video frequencies must undergo a certain
I
amount of attenuation so that the signal being ap-
4 MC «.5 i.25 + plied to the picture tube will be a reproduction of
mc mc
that at the output of the camera tube. Fig. 2-6 illus-
4.5 me
trates the over-all bandwidth characteristics of the
video channel of the television receiver. That part
6 MC CHANNEL WIDTH of the curve to the right of point A is the same as
FREQUENCY (MC) that portion of the video section of the curve of
Fig. 2-4. Point A falls about 2.5 mc away from the
Fig. 2-4.-Actual over-all response curve of a television low -frequency end of the channel. Between zero
chanhel. Note that the response drops off and then rises again
to a sharp peak at the sound carrier. and this 2.5 -mc point, the characteristic of the curve
FREQUENCY CHARACTERISTICS OF THE TELEVISION SIGNAL 29
of Fig. 2-6 differs appreciably from that of Fig. sloping curve of Fig. 2-6 to the left of the video
2-4. This new curve is such that at the video car- carrier where the video frequencies of the lower
rier frequency it will be down 50 percent or 6 db. sideband are attenuated in an increasing manner.
In order that there be faithful reproduction of all This over-all method of attenuation of the low
the frequencies, the shape of the curve is such that video frequencies should be such that, if a video
for all frequencies there will effectively be 100 - frequency, of say, 0.75 mc is attenuated 15 percent
percent transmission. The reference to 100-percent in the upper sideband, then that same video fre-
transmission means the average level of the requency of the lower sideband should be attenuated
sponse characteristic, such as that at the high video 85 percent in the (suppressed) lower sideband.
frequency end of the band. This means the over-all video frequency of 0.75 mc
For instance, the low video frequencies of the passed to the video amplifiers will be equivalent to
upper sideband are attenuated in a gradually de- 100 -percent transmission. In other words the upper
creasing manner to about 1.25 mc from the carrier sideband contributes 85 percent of the video fre-
frequency. This means that as the video frequencies quency and the (suppressed) lower sideband con-
from 0 to 1.25 mc of the upper sideband are in- tributes 15 percent, making it 100 percent over-all.
creased, the amount of attenuation is decreased. The same analysis holds for the other video fre-
To have the reproduction of these low video frequencies from zero to 1.25 mc. The video f requen-
quencies at so-called 100 -percent transmission cies past 1.25 mc (which are all at the same level)
throughout the entire channel, those low video fre- are contained only in the upper sideband and, there-
quencies on the suppressed or lower sideband have fore, do not have to be attenuated. In effect we are
to compensate for the attenuation of those of the putting the low and high video frequencies on the
upper sideband. This compensation must, however, same level so that upon reproduction there will be
be in proportion to the amount of attenuation of no discrimination between the video frequencies.
the upper sideband so that not less than or more
than approximately 100 -percent transmission exists The Carrier and Intermediate Frequencies
at these frequencies. We have already stated that due to FCC stand-
In order that the low frequencies of the sup- ards the sound carrier for each television channel
pressed or lower sideband contribute the proper is located 0.25 mc below the upper frequency limit
of the channel. The video carrier should be located
VIDEO 4.5 mc below the sound carrier. This means that the
CARRIER video carrier frequency is 4.75 mc lower in fre-
quency than the upper limit of the television chan-
nel or 1.25 mc higher in frequency than the low-
f requency limit of the channel. Thus for each one
4 MG
of the 13 television channels the sound and video
UPPER SIDEBAND carrier frequencies have a fixed position in the
6 channel and are 4.5 mc apart. In other words, no
matter what service area is covered by the television
Fig. 2-6.-The over-all bandwidth characteristics of the transmitter the sound and video carrier frequencies
video channel.
for every number 1 channel are the same, those for
every number 2 channel are the same, and so on.
amount of signal strength to these same low fre- Below is a tabulation of each of the 13 television
quencies of the upper sideband for 100 -percent channels with the video and sound carrier fre-
transmission, the suppressed lower sideband also quency used in each channel.
has to be attenuated but in a reverse manner. From
Fig. 2-4 it is seen that the lower sideband (that Most manufacturers include the tabulations below
portion of the channel to the left of the video car- in their television service data but they also pro-
rier) consists only of the video frequencies to 1.25 vide a separate tabulation of the oscillator fre-
mc, and most of these low frequencies are at the quencies for their particular receiver. This oscilla-
same (maximum) amplitude. Consequently, the at- tor frequency is a fixed but different value for each
tenuation of the (suppressed) lower sideband will channel, but for every channel it is used in con-
be such that as the low frequencies are increased junction with both the video- and sound -modulated
the attenuation also increases. This is seen from the signals.
30 RIDER'S -"HOW IT WORKS"
VIDEO SOUND the accompanying sound i.f.'s falling 4.5 ,mc below
CHANNEL CARRIER CARRIER the video.
CHANNEL FREQUENCY FREQUENCY FREQUENCY It appears that there is no definite standardiza-
NO. (MC) (MC) (MC) tion of the i.f.'s as used in today's television re-
1 44-50 45.25 49.75 ceivers since there are so many different ones in
2 51 60 55.25 59.75 use. The television field is relatively new and per-
3 60-66 61.25 65.75 haps in time to come some standardization may be
4 66-72 67.25 71.75 made on the i.f.'s used. For the sake of the service-
5 76-82 77.25 81.75 man it is hoped that the near future will see some
6 82-88 83.25 87.75 such standardization.
7 174-180 175.25 179.75 Some examples of typical sound and video i.f.'s
8 180-186 181.25 185.75 are included in the listing below.
9 186-192 187.25 191.75 SOUND I.F. VIDEO I.F.
10 192-198 193.25 197.75 21.25 25.75
11 198-204 199.25 203.75 21.6 26.1
12 204-210 205.25 209.75 21.75 26.25
13 210-216 211.25 215.75 21.9 26.4
22.25 26.75
The front-end section of the television receivers In all of the above instances, the oscillator fre-
today have a response characteristic the width of a quency tracks above the incoming television sig-
television channel-namely 6 mc. This front end nal. When the oscillator frequency beats with the
usually consists of an r -f stage, an oscillator stage, television signal in the mixer tube, a number of sig-
and a mixer stage for frequency conversion action. nals of different frequencies result. The two fre-
This complete front end functions for both the quencies that we are interested in and for which
video and sound signals. Since both the video and special tuned circuits are employed are the so-called
sound carrier signals have a fixed amount of fre- video i.f. and sound i.f. The video i.f. produced is
quency separation, one oscillator frequency setting quite wide, and when the output video i.f. of the
is used to produce both the video i.f. and sound mixer tube is mentioned, we mean that frequency
i.f. (The front end is discussed in greater detail in which is the .difference between the video carrier
chapter 4.) Once the oscillator frequency is known, and the receiver oscillator frequency. The same
it is a simple matter to find out what the center f re- situation holds true for the sound i -f signal.
quency of the video and sound i -f signals will be. The reason for this clarification of the video i.f.
The video and sound i.f.'s are not fixed in fre- is that the video i -f transformers used in almost all
quency for each channel, which means that the the television receivers are not tuned to the video
oscillator frequency of each receiver is determined. i.f. produced by the mixer tube. The video i -f trans-
by the i.f.'s used. However, for every television re- formers are in most cases stagger tuned with each
ceiver there is a fixed frequency difference of 4.5 adjacent i -f transformer, and in many instances all
mc between the center i.f. of the video and sound i -f transformers, tuned to a different frequency.
signals. This difference is predetermined by the When these frequencies are mentioned they repre-
transmitter which has the video and sound carrier sent only the tuned frequency of the video i -f trans-
frequencies separated by 4.5 mc as specified by the formers and not the video i -f signal produced by the
FCC. mixer tube. In chapter 6 the video i -f stages will be
The center frequency of the i.f.'s has been found discussed in detail along with the theory of stagger
to vary anywhere from about 25 mc to 27 mc with tuning.
CHAPTER 3
TELEVISION RECEIVING ANTENNAS

BY SEYMOUR D. USLAN
There is no doubt that the television receiving an- television antenna is primarily confined to the hori-
tenna is today considered a vital part of the receiver. zontal plane. The necessary service area should be
There has been many an argument launched as to the covered with the greatest amount of field intensity in
necessity of a receiving antenna, but due to the nature a 360° horizontal plane. A certain amount of vertical
of the transmitted television signal, the surrounding radiation is unavoidable, -but this radiation is kept to
terrain, and the location or site of the receiver a re- a minimum.
ceiving antenna is generally considered a must in
television. Aside from the direction of radiation of the trans-
Because of the relative importance of this compo- mitted signal, there is the problem of polarization to
nent for the proper reception of television signals, a consider. When used with reference to radio waves,
separate chapter is devoted to it. In this chapter we polarization means the direction of the electric field
will discuss the more common types of receiving of the signal. The radiated signal is an electromag-
antennas and such associated topics as impedance netic signal, which means there are two alternating
matching, resonance, folded dipoles, reflectors, ghosts fields involved, the electric field and the magnetic
and so forth. field. These fields are 90° out of phase with each
other. When polarization is discussed, reference is
The Transmitted Television Signal made only to the electric field. This means that the
signal can be either horizontally or vertically polar-
The nature of the transmitted television signal is
ized, depending upon the position of the antenna.
quite different from that of radiobroadcast signals
When the radiating element is horizontal, the electric
because of the frequencies involved. The television field alternates in a lateral motion such that the signal
frequencies are short-wave signals, and as such their
is horizontally polarized; and when the radiating ele-
propagation characteristics differ greatly from the ment is in the vertical plane, the electric field alter-
longer wavelength signals in the a -m broadcast bands. nates in an up-and-down motion such that the radi-
These longer wavelength signals follow the curvature ated signal is vertically polarised. Although it has not
of the earth (ground waves) and also use the atmos been standardized by the FCC, horizontal polarization
phere (sky layers) as a means of propagation. When is used by practically all the television stations in the
these signals use the sky layers as a means of propa- United States.
gation, they are reflected or refracted from them and
thus find their way back to the earth. There are two good reasons why horizontal polari-
The antennas employed in regular a -m broadcast- zation is preferred to vertical polarization. In the first
ing, therefore, radiate the signal in all directions and place, with vertical polarization much greater losses
do not confine it to any one particular plane. At the in the radiated wave are caused by ground and at-
higher frequencies (shorter wavelengths) such as mospheric attenuation than with horizontal poláriza-
those used in television and f.m., the signals are not tion. Secondly, many types of interfering signals are
able successfully to utilize the sky layers or curvature vertically polarized. If the radiated television signal
of the earth as a means of travel. Due to the high - were vertically polarized, the interfering vertically
frequency nature of these signals, they will pass polarized signals would be easily picked up by the
through the sky layers without being reflected. If the receiving antenna.
signals try to follow the curvature of the earth, the Television broadcasting- stations are generally lo-
ground will attenuate them very rapidly ; the higher cated approximately in the center of their service
the frequency the greater will be the attenuation. areas, so that the horizontal radiation pattern bas to
In order that the television signal reach the receiver be somewhat circular in shape in order to cover the
with sufficient strength, the signal radiated from the necessary area.
31
32 RIDER'S - "HOW IT WORKS"
Horizon Range and Line of Sight of antenna B and the new line of sight is represented
by line CI).
Television receiving antennas cannot be as simple
as those employed for the longer wavelength signals If a receiving antenna is located in a position where
and still as efficient because of the directive nature of it is beyond the horizon range of the transmitter and.
the television signal. Since the propagated television does not fall within the line of sight of the transmit-
signals have their energy concentrated in one plane, ting antenna, reception of the signal will, in most in-
the receiving antenna should be located at a site where stances. not take place. In other words, since the sig-
it will intercept the signal. nal is directional and cannot successfully use the
Besides the actual design of the receiving antenna, atmosphere or ground as a means of propagation, it
will be blocked by the earth and not be able to reach
two other factors play an important part in the proper
the receiving antenna.
reception of television signals the height of the an-
:
tenna above ground and its distance from the trans- The alxwe analysis indicates that ill order to in-
mitting antenna. In determining these, the horizon crease the possibility of signal pickup, the horizon
range and the line of sight must be considered. By range and, hence line of sight should be increased.
horizon range is meant that distance covered by the This increase can be achieved if either the trans-
direct wave from the antenna to the curvature of the mitting or receiving antenna is raised. In practice the
earth. The line of sight is the maximum distance that transmitting antenna is made as high as is possible
can be seen from the top of the transmitting antenna in the service area it is supposed to cover. The receiv-
to the top of a receiving antenna before the earth ing antennas should likewise be as high as possible in
blocks out the view. order to prevent the signal from being blocked by
Besides the distance between the transmitting and hilly terrain or otlier intervening objects that effec-
receiving antennas, the height of these antennas also tively cut down the horizon range.
determines the horizon range and line of sight. Fig. The actual distance of the horizon range and the
3-1 makes this somewhat clearer. Points A and B line of sight can be very easily calculated. The line of
represent two antennas, A a transmitting and B a sight can be considered as consisting of two horizon
receiving antenna. According to the above definitions, ranges-one from the transmitting antenna and the
line AX represents the horizon range from the trans- other from the receiving antenna. Thus if we have a
mitting antenna and line AB represents the line of method of calculating each horizon range, we can add
sight between the two antennas. If the height of the them together and obtain the line of sight distance.
transmitting antenna were raised from A to C, and This can be easily seen from Fig. 3-2. The horizon
if an observation from the top of point C were then range of the transmitting antenna in this drawing is
made, the horizon range would be seen to have in- equal to distance AB, and at point B it is tangent to
creased from distance AX to distance CY. The line the earth. This means that at point B it makes a right
of sight of the transmitting antenna, now at point C, angle with the radius of the earth drawn from this
is also said to increase. By this is meant that, if an- point. This radius is indicated by line BC. Any other
tenna B is moved so that from its top to the top of straight line drawn from the circumference of the
earth to its center is also a radius of the earth. An-
other such radius is line CD. The height of the an-
tenna is represented by distance AD and when this
line is extended- to the center of the earth at point C
it completes line AC. Lines AB, BC, and AC form
a right triangle. With the height of the antenna known
and the radius of the earth likewise known, it is very
easy to calculate the horizon range.
In ally right triangle the two sides that form the
right angle, as sides AB and BC in Fig. 3-2, are
FIG. 3-1.-The line of sight is shown between two antennas called the legs of the triangle. The side connecting the
at A and B. If the antenna A were raised to position C, the
line of sig:it would be extended to D. two legs, side AC. is called the hypotenuse of the tri-
angle. The hypotenuse of any right triangle is always
antenna C a new line of sight is formed, this new line larger than either one of its legs but smaller than the
of sight will be greater than AB. This is shown in arithmetical sum of the two legs. An old geometrical
Fig. 3-1, where point I) represents the new position theorem about right triangles, called the Pythagorean
TELEVISION RECEIVING ANTENNAS 33
theorem, states that the square of the hypotenuse is height of either t, the transmitting, or r, the receiving,
equal to the sum of the séuares of its two legs. Refer- antenna will increase S, the line of sight.
ring to Fig. 3-2 the theorem is expressed symbolically For example, if the height t of the transmitting
as follows : antenna is 1000 feet and if the receiving antenna,
(AC)2 = (AB)2 + (BC)2 Eq. 3-1 which falls beyond the horizon range of the trans-
Calling the radius of the earth R and the height of mitting antenna, has a height r of 50 feet, the line of
the transmitting antenna t and the horizon range dis- sight S would be calculated as follows :
tance AB by the letter X, the above relation reduces S= 1.23 (Vt+Vr)
to the following : S= 1.23 (x1000+ V50)
(t+R)2= X2+ R2 Eq. 3-2 S= 1.23 (10/10+5V2)
S = 1.23 (31.62 + 7.07)
which reduces to the following : S= 1.23 (38.69)
t2 + 2tR + R2= X2 + R2 S = 47.59 miles
Eq. 3-3 Thus the line of sight between the two antennas is
and subtracting the R2 from either side of the equa- equal to 47.59 miles. If the receiving antenna is raised
tion we have : to 200 feet the line of sight would also increase. Thus :
t2 + 2tR = X2 Eq. 3-4 S = 1.23 (V 1000 + V200)
The radius of the earth, R, is approximately 4000 S= 1.23 (10/10+ 10/2)
miles. Compared with this the height of the antenna, S = 1.23 (31.62 + 14.14)
t, even when squared will be very much smaller than S = 1.23 (45.76)
S = 56.28 miles
the expression 2tR and therefore the term t2 can be Increasing the receiving antenna height from 50 to
neglected in the last equation. Consequently. the final 200 feet has thus increased the line of sight distance
resultant equation is :
by approximately 8.7 miles.
X2 = 2tR Eq. 3-5
or
TRAN9hiITTINCa A
X = V2tR Eq. 3-6 ANTENNA
Since R equals approximately 4000 miles, we can t
further simplify equation 3-6 as follows :
X=V2xtx4000=V8000t=88.6Vt
Eq. 3-7
Equation 3-7 will give the horizon range X in units
of miles when the antenna height t is also given in
units of miles. Since the antenna height is usually
CENTER OF
given in terms of feet, equation 3-7 can be changed to : THE EARTH
X = 1.23 Vi Eq. 3-8
where X is equal to the horizon range in miles and t
the antenna height in feet. Ftc. 3-2.-The horizon range is the distance between points
A and 13. The horizon range from the transmitting antenna
The same type of analysis will hold with the re- and the horizon range from the receiving antenna can be
added together to obtain the line of sight.
ceiving antenna. That is, the horizon distance in miles
from the receiving antenna is equal to the product of
Propagation of television signals can cover dis-
1.23 and the square root of the height of the receiving
tances beyond the line -of-sight range when the effec-
antenna in feet.
tive radiated signal from the antenna is strong (that
As mentioned before, the maximum distance of the
is, either a high-powered transmitter or a transmit-
line of sight is equal to the sum of the horizon ranges
ting antenna with a high power gain, or both). This
of the transmitting and receiving antennas. Thus,
is primarily caused by the directed radiated television
with height of the transmitting antenna designated as
signal undergoing diffraction and refraction as it
t and that of the receiving antenna designated as r
travels toward the receiver. This is a result of the
the line of sight distance, call it S, is equal to the
nature of the atmosphere. principally the troposphere,
following :
that part closest to the earth. Because the signal un-
S = 1.23 \/ + 1.23 -c/ r
t Eq. 3-9
or we can write equation 3-9 in simpler form as dergoes diffraction and refraction, it has a tendency
follows :
to follow the curvature of the earth through the at-
S= 1.23 ( + Vr) Eq. 3-10 mosphere and thus cover distances beyond the line of
From equation 3-10 it can be seen that raising the sight range.
The Television Signal at the Receiver ferent types of receiving antennas perform these func-
tions in one manner or another, as will be seen later.
It has often been said that a television receiver is
only as good .as its antenna. But some authorities have When we refer to a signal-to-noise ratio at the in-
also held that a piece of wire will serve as an antenna, put to the television antenna, the noise factor does not
or even that no antenná is needed. These contradic- include any tube or thermal noise relative to the tele-
tory statements arise from the fact that good television vision receiver circuits. Actually, therefore, two noise
reception has been attained using just straight pieces components have to be taken into account when the
of wire, and thus, the necessity of a television antenna television signal reaches the detector circuits of the
is disputed. However, the only cases of such reception receiver. One of the. components results from the
are of' the particular kind where the television receiver noises and interfering signals picked up directly by
happens to be located in a very favorable region with the antenna, and the other is attributed to the inherent
respect to some of the broadcasting antennas. The noise characteristics of the components within the re-
signal strength is then relatively high, and good recep- ceiver itself. Since it is difficult to reduce the inherent
tion may be obtained without the use of a television noise within the receiver, the best thing to do is either
receiving antenna. to reduce the noise pickup by the antenna system or to
increase its signal pickup, so that the signal-to-noise
If a rigid test is made in such an area with a per-
ratio of the television signal just before the detector
fectly matched antenna system, the reception of the
stages of the television receiver will be high.
television signal will be better than without any an-
To understand why a half -wave dipole is used as
tenna at all. In practically all installations it can be
the basis of the television antenna, it is necessary to
shown that television reception is definitely improved
understand something about the voltage and current
with the proper type of antenna. At points which are
distribution, antenna resistances, resonance condi-
a considerable distance from, but within the service
tions, and the like with respect to such types of
range of the transmitter, the signal pickup of a con-
antennas.
ventional single wire is usually so small that good re-
ception cannot be obtained.
Voltage and Current Distribution
If you are in doubt whether to use a television an-
tenna in yotir location and if you do not know too In choosing the type of antenna for television re-
much about the surrounding terrain and where the ception, as well as other types of reception, the voltage
television broadcasting antennas are located, then it is and current distribution along the antenna must be
definitely advisable to use a television receiving known for an understanding of the reason a certain
antenna. antenna is used. In this respect, let us examine the
full -wavelength straight -wire antenna as seen in Fig.
The nature of the transmitted televisions signal is a
3-3. The distribution is such that of the ends of the
criterion in orienting the television receiving antenna
wire the current is a minimum and the voltage a
in order to obtain the maximum amount of signal
maximum. Due to certain characteristic phenomena
pickup for proper reception. As mentioned previously,
of open wires used as antennas, such as that appearing
the transmitting antennas of most television stations
in Fig. 3-3, standing waves of both voltage and cur-
employ horizontal polarization and thus the signal is
rent exist along these antennas. By standing waves is
horizontally directed.
meant the following: A radio -frequency wave travel-
This necessitates the orientation of the television ing to the ends of the antenna will be reflected back
receiving antenna in a horizontal position for maxi- to its starting point. The reflected wave will meet the
mum signal pickup from the passing television waves. original (so-called incident wave) in such a manner
More will be said later about the orientation of the that the individual voltage and current curves of these
television receiving antenna. waves will add algebraically such that voltage and cur-
The antenna generally employed for television re- rent waves always exist on the antenna. These waves
ception is a single half -wave dipole or some combina- are referred to as standing waves. Antennas in gen-
tion of half -wave dipoles, with or without reflector eral are referred to as radiating elements, or a series
elements. All types of television receiving antennas of such elements, and for the antenna to be a good
are designed to perform two main functions, namely, radiating element (or receiving element), standing
obtaining the necessary signal pickup and supplying waves should exist on it.
an output impedance that can be properly matched to Fig. 3-3 shows the standing wave distribution for a
the receiver for maximum energy transfer. The dif- full -wavelength antenna. Distribution of the voltage
and current is sinusoidal. Since the voltage and cur- receiver. The standing wave ratios quoted by antenna
rent are represented by sine waves, it is evident that manufacturers include the loss introduced by the
they change polarity at certain points along the full - transmission line the lower the standing wave ratio,
;
wavelength wire. The current is a maximum and the therefore the better the antenna and transmission
voltage zero at one -quarter of a wavelength from line setup.
either end of the wire. Since the full wavelength of
wire represents one complete cycle (360 degrees of The (Half -Wave) Dipole Antenna
electrical length of either voltage or current), the Let us now consider a half wavelength of wire or
standing waves of voltage and current are said to be
just half that in Fig. 3-3. This half -wave wire is illus-
trated in Fig. 3-4 and is generally representative of
the current and voltage distribution of the half -wave
dipole antennas used with many television receivers.
A half -wave antenna is often referred to as just a
dipole antenna. The terminology (dipole) for these
FULL WAVE
ANTENNA antennas has originated from the fact that voltage dis-
tribution along the half -wave antenna is such that at
the ends of the antenna the voltages are of opposite
H ri WAVE
polarity (that is, positive and negative charges). This
FULL WAVE
is readily evidenced by the half cycle of voltage in
Fig. 3-4. Since the half -wave dipole antenna is the
FIG. 3-3.-A full -wavelength straight -wire antenna with
standing current and voltage waves. Note that at the mid- basic type used for television receivers, let us examine
point of the antenna the current is a minimum and the voltage some of its characteristics as shown in Fig. 3-4. At
a maximum.
any one instant, the current standing wave is of the
90° out of phase. This 90° phase difference is the same polarity at all points, and there is, ideally, zero
same as saying that the reversals of polarity of the current at the antenna ends with maximum current,
voltage and current waves occur one -quarter of a which is referred to as a current loop, at the center of
wavelength apart. the dipole at one -quarter of a wavelength from either
The numerous qualities of an antenna apply to both end. The voltage changes polarity, and the change is
transmitting and receiving antennas, so that those stich that there is, ideally, zero voltage at the center
properties discussed under receiving antennas apply of the dipole, one -quarter wavelength from either end.
equally well to transmitting antennas and vice versa.
Consequently, when we talk about the voltage and
current distribution of the standing waves on one FIG. 3-4.-The current
type of receiving antenna, it applies equally well to and voltage distribution
along a half -wave antenna.
the same antenna when used for transmitting pur- Note that at the ends the
current is a minimum and
poses. the voltage a maximum.
However, the transmission line used with the an- Compare with Fig. 3-3.
tenna should have no standing waves on it, so that
the line will not radiate any of the energy but rather
transfer it to the receiver. If standing waves do exist At the ends of the dipole, the voltage is a maximum,
on the transmission line, some of the energy picked but the voltage at one end is of opposite polarity to
up by the antenna will be lost from the line as this the voltage at the other end.
energy is transferred to the television receiver input A drawing of a half -wave dipole antenna is illus-
and also "ghosts" may appear on the picture tube trated in Fig. 3-5 (A) along with the waves of volt-
screen. This will be discussed in detail later on in this age and current distribution. In Fig. 3-5 (B) is a
chapter. In order to make the transmission line a non - picture of a typical half-wave dipole antenna as used
radiating element having no standing waves (that is, today. The half -wave dipole antennas illustrated are
no reflection of the signal), the line must be termi- those that have center lead-ins and are termed cur-
nated in its characteristic impedance. This means that rent-fed antennas, because the feed -in lead is at a
the impedance of the transmission line must be prop- current loop or maximum. Half -wave antennas can
erly matched to the impedance of the television re- also be end -fed, but television receiving antennas are
ceiving antenna and also to the input of the television generally center -fed dipoles. For proper reception, the
half-wave dipole antenna should have a length ap- 73 ohms. The actuál value of radiation resistance of
proximately equal to one-half of a wavelength at the the simple dipoles, as used, varies somewhat away
center frequency of the frequency spectrum it is to re- from 73 ohms, depending on the exact length of the
ceive. Consequently, the length of each section should dipole and presence of other physical factors.
CURRENT VOLTAGE DIPOLE

WAVE WAVE ELEMENTS Resonance and Impedance
The half -wave antenna behaves very much like a
tuned circuit. The antenna has inductive, capacitive,
R SUPPORT and resistive components, but they are all distributed
throughout the antenna in contrast to the lumped L,
TRANSMISSION l' C, and R components in the regular tank circuit. The
LIESSION
simple half -wave dipole antenna that is center tapped
(B) Courtesy Technical Appliance Corp. behaves like a series resonant circuit. This is shown
FIG. 3-5.-Current and voltage distribution in a half -wave in Fig. 3-6 (B) where L, C and R represent the in-
antenna (A). In (B) is shown a typical dipole antenna, its ductive, capacitive, and resistive components respec-
transmission line, and a method of support.
tively. The resistive component is primarily the
be about one-fourth of the wavelength of the middle radiation resistance of the antenna since the real or
frequency which the antenna is designed to receive. ohmic resistance is relatively very small, even at the
This will be elaborated upon later in this chapter. increased value it has at high frequencies.
If the dipole is approximately a half wavelength
Antenna Resistances long, or about one -quarter wavelength on either side
All the power input to a transmitting antenna is of the center tap, the antenna will act as a series reso-
dissipated in one form or another. The so-called re- nant circuit to the frequency for which it is a half -
sistances of an antenna determine how this power is wave antenna. In a series resonant circuit the imped-
dissipated. Any ántenna, whether it is used for re- ance is a minimum and purely resistive, the current a
ceiving or transmitting, effectively contains two types maximum, and the voltage a minimum. This is also
of antenna resistance. One type is the usual ohmic
resistance of the metal parts of the antenna, often CURRENT
WAVE
called the real resistance; the other type is often called

the imaginary or radiation resistance. In other words, DIPOLE
ANTENNA
we are concerned with two types of power, the power
dissipated due to the actual ohmic resistance of the VOLTAGE
WAVE FEED
LINES (B)
antenna and the power radiated from the antenna. (A)
The former power is readily understandable and,
from this, the ohmic resistance is conceivable. The
FIG. 3-6.-A half -wave dipole antenna (A) and its equiva-
resistance dissipating the latter portion of the power lent series resonant circuit (B), where the resistive compo-
is not, in reality, a physical resistance in the sense that nent R is primarily the radiation resistance of the antenna.
the other one is and, hence, often is known as the
imaginary or radiation resistance of an antenna. Con- seen from Fig. 3-6 where, at the center point of the
sequently, when there is talk of power dissipation .of half -wave antenna, the current is a maximum and the
an antenna, the 12R total power loss should be under- voltage a minimum. At other points along the antenna
stood to encompass both types of resistances. Resist- the impedance is not purely resistive because it en-
ance R is the series combination of the ohmic re- compasses some reactance. This is clue to the chang-
sistance of the antenna and the radiation resistance ing relationship of the voltage and current standing
of the antenna. waves and, therefore, the impedance is greater away
In most types of antennas, the ohmic resistance is from the center point, being a maximum at either end
much smaller than the radiation resistance, so that of the half -wave antenna due to the minimum value
practically all the power is dissipated through the of current and maximum value of voltage.
radiation resistance. For a half -wave dipole antenna
in free space (that is, no intervening objects including Transmission Lines
ground effects, buildings, mountains, etc.) the radia- The transmission line connecting the antenna to
tion resistance is found to be equal to approximately the receiver is as important a factor in obtaining the
correct input signal to the receiver as the antenna simply measuring the voltage and current at the input
itself. The transmission line is primarily intended to end whether the line was of a finite length or infinitely
provide a ready path for the television signal from the long. In other words, when a finite line is terminated
antenna to the receiver, with a minimum amount of in its characteristic impedance, the input impedance of
signal loss. To adequately provide this ready path, that line is equal to the characteristic impedance.
the transmission line must match the impedance of When this fact is known, maximum energy transfer
the antenna to the input impedance of the television from the antenna to the receiver via the transmission
receiver. For this so-called impedance match, which line is easily attained. This will be seen in the next
is discussed in detail in the following section, the section on impedance matching.
characteristic impedance of the transmission line has If the leakage conductance and series resistance per
to be known. unit length of line is negligible, as is usually the case,
No matter what type of transmission line system the characteristic impedance of a line is purely re-
is used, each one is distinguished by its own char- sistive and can be represented by a relationship be-
acteristic impedance which is expressed in ohms. tween the inductance L and capacitance C per unit
There are two main types of transmission lines used length of line. Denoting the characteristic impedance
with today's television receiver-the open -wire line, by the common expression Zo, it is represented mathe-
sometimes also called twin -lead, twin -pair, or ribbon matically as follows :
type, and the coaxial line. The characteristic imped-

ance, sometimes called surge impedance, iterative Zoz=VL
Y C Eq. 3-11
impedance, or just plain impedance is determined by This expression holds for both twin -lead trans-
the physical make-up of the line and can be calculated mission lines and the coaxial type. The characteristic
by knowing the inductance and capacitance per unit impedance can also be given in terms of the physical
length of line (assuming a lossless line in the latter characteristics of each type of line. For the twin -lead
case.) transmission line, as seen in Fig. 3-7, the character-
The impedance value of a transmission line is not istic impedance Z is given by the following :
a direct function of the length of the line, although

this is often thought to be the case. In other words,
if the known characteristic impedance of a transmis-
Z = ;, logro
2Dd
Eq. 3-12
where D is the distance between centers of the wires

sion line is equal to, say, 1.00 ohms, this value of im- d is the diameter of the wire (whether one
pedance does not refer to any specific length line but wire or a number of small wires grouped
to the type of line itself. That is, 500 feet of the line together)
will not have twice as much (characteristic) im- e represents the dielectric constant of the ma-
pedance as 250 feet of the same line but will have the terial surrounding the wires. For air e = 1
and for polyethylene e = 2.25. (This value
same characteristic impedance. may vary slightly.)
If a transmission line had a voltage applied across (Equation 3-12 is obtained from equation 3-11 by
its input terminals, the ratio of this voltage to the substituting the expressions for L and C for the par-
current at the input end would be the input im- allel transmission line of Fig. 3-7.)
pedance. If the line were uniform and of infinite
length, the input impedance would become the char-
acteristic impedance of the line. Thus if a line in- Flo. 3-7.-Cross-sectional view of
a twin -lead transmission line. D is
finitely long had a 50-foot section cut off, the charac- the distance between centers of the
teristic impedance of the line would not change, be- twin leads and d is the diameter of
each lead.
cause subtracting a finite length of line from one that
is infinitely long still leaves the line infinitely long.
The piece cut out is a part of the infinitely long line In using such a formula for computation, the units
and as such the characteristic impedance is a means of D and d should be the same, whether they are
of identifying this part. The input impedance of this measured in inches or centimeters. After computing
50 -foot line may be different from the characteristic 2D/d the logarithm of the answer (logio) can be
impedance of the infinitely long section. However the found in any book containing tables of logarithms.
50 -foot piece is still said to have the same character- Remember, however, that logro means "logarithm to
istic impedance as the infinite line. the base 10" and refers to a table of "common loga-
If the finite line were terminated in its character- rithms" and not natural logarithms, which are com-
istic impedance, we would not be able to tell by puted to a different base.
38 RIDER'S "HOW IT WORKS"
For example, let us compute the characteristic im- Thus

pedance of a parallel -wire transmission line that is 138 D
covered by polyethylene plastic insulator (E = 2.25) Z° = logro
U E
and measures 22/64 of an inch between centers with 138 19/64

the wire having a diameter of 3/64 of an inch. Using Zo
x/2.25 logro
equation 3-12 for the characteristic impedance Zo we 138 logro
have: Zo
3
Z=
276
logro
2D
d
and since = 2.25, D = 22/64", and d = 3/64"
E
= 92 log10 6.33
= 92 (.8014)
= 73.7 ohms
276 2(22/64) Such a coaxial line is often said to have a 75 -ohm
Zo
x/2.25 logro impedance.
= 276 44 The parallel wire and coaxial type of transmission
1.5
logro lines are available in various characteristic imped-
3
= 184 log10 14.66 ances. Most antennas used in television reception
= 184 (1.662) llave either about a 75 -ohm impedance as the simple
= 306 ohms half -wave dipole antenna or about 300 ohms as the
Such a transmission line would be listed as having folded dipole. Consequently, the transmission lines
a 300 -ohm characteristic impedance. used should have either a 75 -ohm or a 300 -ohm char-
Referring to Fig. 3-8, the characteristic impedance acteristic impedance. When the coaxial line is used,
for the coaxial transmission line is as follows :
it is primarily employed for impedance matching be-
138É D tween a low -input impedance receiver and a simple
Zo = logro Eq. 3-13 dipole antenna. For high -input impedance (about 300
where D is the inside diameter of the outer conductor ohms) television receivers, the 300-ohm twin -lead
(or the outside diameter of the dielectric) transmission line is usually employed. The above
d is the diameter of the inner conductor methods are the most commonly used, although varia-
(whether one wire or a number of small tions do exist.
wires grouped together)
E represents the dielectric constant of the di-
electric material. For air E = 1, for polyethy- Impedance Matching

lene E = 2.25. (This latter value may vary So that the maximum amount of energy pickup will
slightly).
be transferred from the half -wave antenna to the re-
(Equation 3-13 is obtained from equation 3-11 by
ceiver input, the antenna has to be properly "imped-
substituting the expressions for L and C for the co-
ance -matched" to the receiver input. A fair idea of
axial transmission line of Fig. 3-8.) In equation 3-13,
impedance matching can be had from Fig. 3-9. Look-
as well as equation 3-12, both I) and d have to be in
ing in the direction of the r -f amplifier, through the
the saint units.
primary of the input transformer, we see an imped-
ance equal to Z1; and looking in the direction of the
transmission line and antenna, we see an impedance
FIG. 3-8.-Cross-sectional view of of Z2; for maximum energy transfer Z1 should equal
a coaxial transmission line. D is the
inner diameter of the outer con- Z2. Under this condition the maximum amount of
ductor and d is the outer diameter energy possible, not all the energy, will be transferred
of the inner conductor.
from the antenna to the grid of the first tube. For the
best match to occur, the transmission line should first
be impedance matched to the antenna. Then these
The method of using equation 3-13 to compute the units through the medium of the receiver input trans-
characteristic impedance of a coaxial transmission line former should be matched to the input impedance of
from its physical dimensions is similar to the example the r -f amplifier. If there is any mismatch, there will
just discussed for the parallel wire. For example let be a loss of energy, and the maximum possible amount
us compute Z° for a coaxial line that uses polyethy- of energy will not be transferred.
lene as a dielectric (E = 2.25), where the diameter D As has been pointed out, the input impedance at
of the dielectric and the diameter d of the inner con- the center of the half-wave dipole ideally is a pure
ductor are 19/64 and 3/64 of an inch respectively. resistance and equal to 73 ohms. In practice, this
value of input impedance of the dipole can vary any- In reality, almost any good type of balanced trans-
where from 50 to 100 ohms as a result of such vary- mission line, besides the simple twisted pair, can be
ing physical factors as the construction of the antenna, used to match the antenna as long as the characteristic
the obstacles near the antenna, and its height. A very impedance of the line is close to the radiation resist-
simple impedance match to such an antenna is made ance of the antenna. Many receivers use television
by using a twisted -pair transmission line (usually, antennas supplied by outside manufacturers ; some
ordinary rubber -insulated wire) for the feeder sec- receivers are not directly supplied with an antenna,
tion, as shown in Fig. 3-10. The so-called character which must be bought separately. In the latter case
istic impedance of such a transmission line is some- knowledge of the input impedance of the receiver's
where around 75 ohms. If different types of wire are input transformer is necessary to secure the proper
impedance match. Many television receivers today
have an input impedance equal to about 300 ohms, so
that the transmission line and antenna have to be
R -F AMPI.
properly matched to this 300 -ohm impedance for
RALF -WAVE
DIPOLE ANTENNA proper energy transfer. Transmission lines with a
INPUT TRANSFORMER
characteristic impedance of 300 ohms would normally
be used to match the line to the transformer, but the
300 -ohm line would be mismatched to, for example,
TRANSMISSION a 75 -ohm simple half -wave dipole.
LINE
Some mismatch can be tolerated under certain con-
ditions. First of all, let it be understood that for
maximum efficiency the transmission line should
Fic. 3-9.-In order to effect a maximum transfer of energy
from the antenna to the control grid of the r -f amplifier tube, match the impedance of the antenna to the input im-
the impedance ZQ of the transmission line and antenna should pedance of the receiver. However, a certain amount
equal the impedance Zl seen looking into the primary of the
of mismatch may be allowable provided that satisfac-
input transformer.
tory performance of the television receiver is ob-
chosen for the twisted line, and if the physical hook- tained. Most television receivers are sensitive enough
ups to the antenna can be varied somewhat, the im- to suffer a certain amount of energy loss in the re-
pedance of this transmission line can be varied on ceived signal and still operate satisfactorily. This
either side of 75 ohms for the desired impedance means that if there is a small amount of impedance
match. Beside giving a good impedance match, this mismatch between the antenna and receiver there will
type of line minimizes pickup by the lead-in due to be a certain amount of energy lost by the setting up
the twisting effect and closeness of spacing between of standing waves in the transmission line, and if this
each individual wire, and thus noise pickup by the loss is small, it can be tolerated. Even though mis-
transmission line is reduced. matching by as high as a 4 to 1 ratio has resulted in
Where the antenna and line are properly matched,
the input transformer of the receiver should have an
impedance approximately equal to that of the line or TWISTED PAIR
antenna. For example, if the antenna resistance and TRANSMISSION LINE
characteristic impedance of the line are each about

100 ohms, then looking into the primary of the input
transformer, toward the r -f amplifier, the impedance
seen should also be 100 ohms. Under these circum-
stances, the maximum amount of energy transfer will
be made from antenna to receiver. In other words,
for maximum energy transfer in any kind of sys-
tem, the impedance of the load should equal the im- FiG. 3-10.-In order to match the impedance of a half -wave
pedance of the source. If the impedances are not dipole antenna with that of the transmission line, the latter
can be formed of twisted pair, which has a characteristic im-
purely resistive (that is, contain reactances) then the pedance of approximately 75 ohms.
reactances in the load and those in the source must
be of opposite sign. In television receiving antenna so-called satisfactory receiver performance, this type
systems, the impedances involved are considered pure of mismatch is definitely undesirable and should be
resistances for matching purposes. avoided wherever possible. Besides energy loss such
40 RIDER'S - - "HOW IT WORKS"
a mismatch can also cause "ghosts" on the picture the impedance increases and, hence, the impedance
tube. This will be discussed in detail later on. Since mismatch decreases. Therefore, for these off-reso-
most antenna and transmission line manufacturers nance frequencies the dipole antenna does not create
list the impedances of their products, it should be a 4-to -1 mismatch with a 300 -ohm transmission line.
quite easy to obtain a match between the 'antenna and However, this 4-to -1 mismatch still exists at the cen-
receiver. The difficulty in the complete matching proc- ter frequency to which the antenna is cut, but if the
ess is in knowing the input impedance of the televi- antenna is cut to a frequency which has a relatively
sion receiver. A great many of these receivers have strong field in the vicinity of the antenna, the input
an input impedance equal either to approximately 300 signal may he strong enough to overcome the loss the
ohms or about 75 ohms, but the impedance of some mismatch entails and, hence, reception at this fre-
differs from these. It is readily seen that it is difficult quency may be satisfactory. It should be remembered
to effect a complete impedance match when it is nec- that the increased impedance on either side of the
essary to guess the input impedance of the receiver. center of the antenna is not purely resistive but en-
If the mismatch is too great, there may not be compasses some reactance.
enough signal input to the television receiver to over- However, variations of the simple half -wave dipole
come the inherent noise of the receiver. The signal-to- antenna and variations in transmission line hookups
noise ratio will be low, and the reproduced picture exist, so that impedance matching can be closely at-
signal, as well as the audio signal, will be noisy and tained. There are folded dipoles, dipoles with re-
incoherent. flectors, specially constructed simple half -wave dipoles
It has been stated that for the best reception the line that change the effective radiation resistance, and
should be matched to the antenna. However, in cer- other unique types of dipole arrangements. Analysis
tain instances simple dipoles of 75 -ohm input imped- of some of these types of television receiving antennas'
ance have been used with 300 -ohm transmission lines, is made later.
yet the signal input to the television receiver is said For any combination of transmission line and an-
to be enough for proper reception. With this mis- tenna, the smaller the length of the line, the less will
match of 4 to 1 the set still operates satisfactorily. be the line losses, no matter how well it is impedance
Many television antenna manufacturing companies matched. If the line must be long, due to the location
stipulate that, if a 300 -ohm transmission line is used of the antenna, low -loss transmission lines 'should be
with their simple half -wave dipole antenna's, which used to prevent excessive reduction in the signal
have an input impedance around 75 ohms, proper reaching the receiver.
reception will still be attained. Knowledge of the con-
In conclusion, it may be said that, although correct
struction, the Q, and the impedance of the dipole is
impedance matching will give the best over-all per-
necessary to understand completely how such a mis-
formance, the primary purpose in most television
match can be tolerated.
antenna installations is to obtain the necessary voltage
The input of a simple dipole is, as mentioned, about
input to operate the receiver satisfactorily.
75 ohms and resistive in nature. This impedance
varies from the center to either end, with the imped-
Q of Antenna
ance at the ends being a maximum and somewhere in
the vicinity of 2500 ohms. Since this is so, if the an- The inductance and resistance of a straight piece
tenna can be tapped at some point away from the of round wire will decrease. as the diameter of the
center, where the impedance is about 300 ohms, we wire increases. In other words, both L and R are
have a good place to match.a 300 -ohm line. This pro- inverse functions of the diameter of the wire. How-
cedure is somewhat impractical, but the same effect is ever, at operating frequencies within the television
obtained in another way. The 75 -ohm impedance that band the inductance decreases at a faster rate than
appears at the center of the dipole antenna is meas- the resistance for changes in diameter.
ured at the frequency for which the antenna is cut. Since the Q of a circuit or a wire may be defined
On either side of this center frequency the impedance as the ratio of inductive reactance to resistance, the Q
of the antenna at the center leads will increase-the of the dipole is seen to decrease as its diameter in-
amount of increase being determined by the frequency creases. This is clue to the fact that the inductance
change. Thus an impedance mismatch of 4 to 1 for a and, hence the inductive reactance, decreases at a
300 -ohm line attached to a simple half -wave dipole much faster rate than the resistance as a result of the
antenna occurs only at the center frequency for which diameter increase. This is the same as saying that the
the antenna is cut ; on either side of this frequency resistance effectively increases with respect to the
inductive reactance. Consequently, it can be said that Those shown in parts (A )and (B) are typical out-
the Q of a dipole antenna is an inverse function of its door folded dipoles, while that shown in part (C) ,
diameter. The. inductive reactance, and hence the Q although suitable for outdoor use (with proper me-
of a wire, increases with increase in wire length, but chanical support), is especially useful as an indoor
for the foregoing analysis of the Q of an antenna the antenna. This latter antenna can be placed under a
length of the dipole, as well as the frequency of opera- rug, in a closet, or some other convenient place in the
tion, is considered constant. The lower the number of home. It is made completely out of 300 -ohm twin -lead
wire used for the dipole, the greater will be the diam- transmission line ; its construction is discussed in a
eter and hence the lower the Q. following section on indoor antennas.
Since the low-frequency (channels one through Let us now study the folded dipole in general and
six) and the high -frequency (channels seven through see how this type of antenna increases the input im-
thirteen) groups of the television band are quite wide, pedance compared with a simple half -wave antenna.
the antennas used for either or both groups should The folded dipole is similar to an autotransformer
have as broad a response as possible. This broad re- where the primary of the transformer is analogous to
sponse is necessary in order to receive all the signals that part of the folded dipole which has the trans-
within each television channel with little or no dis- mission line attached to it and the secondary of the
crimination. By lowering the Q of the antenna it is transformer is analogous to the other half -wave sec-
possible to secure a fairly broad -band characteristic. tion of the folded dipole. Accordingly it is readily
However, since a reduction in Q "flattens" out the re- seen that a mutual impedance exists between both
sponse characteristic, there is also a reduction in sig- half-wave sections of the folded dipole in the same
nal pickup. Consequently, a compromise has to be way that mutual inductance exists between the wind-
effected between the tolerable loss in signal pickup ings of a transformer.
and the reduction in Q. Each individual half -wave section of the folded
dipole is considered as a resonant half -wave antenna
The Folded Dipole and fundamentally resistive at the resonant frequency.
The folded dipole antenna has a great advantage This is the same as having both the primary and sec-
over the simple dipole antenna in that it exhibits a ondary of the transformer tuned to the same resonant
much higher impedance, thus allowing for a better frequency in which the impedance of each will be re-
sistive in nature. This means that looking into the
APPROX. tapped portion of the folded dipole the complete im-
V2 WAVELENGTH pedance seen is the sum of the tapped half -wave sec-
tion plus the reflected impedance from the other half -
}FOLDED
DIPOLE wave section. Reflected impedance means that be-
cause of mutual impedance between both half -wave
sections of the folded dipole there is an impedance
reflected into the primary which changes the over-all
TRANSMISSION LINE input impedance of the antenna. This reflected im-
pedance is mathematically equal to the square of the
FIG. 3-11.-A folded half -wave dipole antenna, where d, the

distance between the two half -wave sections, is small in com-
parison with the wavelength.
impedance match to receivers having 300 -ohm inputs.
A folded dipole antenna is the simple half -wave an-
tenna illustrated in Fig. 3-5 (A) with another half -
wave antenna section joined to it at the ends. A draw-
ing of a typical folded dipole is illustrated in Fig. 3-11.
From this drawing the folded dipole is seen to be (A) Courtesy Technical Appli-
ance Corp.
about a full wavelength of wire (or appropriate tub- (B) Courtesy Camburn Inc.
(C) Courtesy Insuline Corp. of
ing or rod) bent so that it takes on approximately the America
shape shown in Fig. 3-11. In Fig. 3-12 are shown FIG. 3-12.-Three types of folded dipole antennas. Those
illustrated in (A) and (B) are for outdoor use and that in
three typical folded -dipole antennas as used today. (C) is for indoor or outdoor use.
mutual impedance (which is purely reactive in nature) of the folded dipole is equal to four times the value of
divided by the impedance of the untapped half -wave either self -impedance. For the half -wave folded dipole
section. When multiplying the mutual reactance by these impedances are resistive, as previously men-
itself, the result is a purely resistive component, and tioned, and, thus, the total input resistance of the
since the impedance of the untapped half -wave sec- folded dipole is equal to four times the input resistance
tion of the folded dipole is resistive (at resonance), of a single half -wave section of the folded dipole.
the reflected impedance is, also, effectively resistive. Since a simple half -wave dipole antenna has approxi-
Consequently the total input impedance of the com- mately the same input resistance as a single half -wave
plete antenna system, as seen from the feed-in half - section of the folded dipole, the input resistance of a
wave section, under the above conditions of resonance, folded dipole is about four times as great as that for a
is resistive. simple half -wave dipole.
Since the distance d separating the two half -wave Under ideal conditions the input resistance of a
sections of the folded dipole as seen in Fig. 3-11 is half -wave dipole is equal to about 75 ohms. So that
much smaller than the half -wave length of the an- under similar conditions the input resistance of the
tenna, the mutual impedance is considered to ap- folded dipole is equal to 4 x 75 or 300 ohms. Thus
proach the maximum possible value. In other words, for television receivers having a 300 -ohm input im-
using the analogy of the transformer again, the co- pedance, a 300 -ohm transmission line can be used to
efficient of coupling of the half-wave sections of the match the folded dipole to the receiver for maximum
folded dipole is said to be approximately unity. Since energy transfer.
each individual half -wave section has its own so-called
self -impedance, similar to the self-inductance of the Length of the Half -Wave Antenna
windings of a transformer, the total self-impedance of Throughout this chapter we have constantly used
both sections (since they are connected) is equal to such nomenclature as the length of the dipole being a
the sum of their individual self -impedances. The total half -wavelength long. This was all preliminary to
self-impedance of the folded dipole is not the complete solving the proble-ns of the actual physical length of
impedance, because the folded dipole antenna im- the antenna and cl oosing a particular length. If only
pedance as a whole takes into account the mutual one frequency is g )ing to be picked up (that is, for a
impedance as well as the total self -impedance. The fixed frequency re :eiver), the antenna length is easy
wire, rod, or tubing used for both half-wave sections to calculate and is based on that signal frequency.
is usually macle of the same material, so that the self - However, we are concerned with television broadcast
impedance of both half -wave sections are about the receivers, and the television band today is essentially
same, neglecting the small difference caused by the from 44 to 88 mc and from 174 to 216 mc, so that the
slight spacing of the feed-in section of the half -wave antenna length for either group must be chosen so
element that has the transmission line attached. that it will be responsive to all frequencies within the
For an' autotransformer wired in series aiding, the group.
total inductance of the unit is equal to the sum of the Most antennas that are in use today are designed
individual self-inductances plus twice the value of the primarily for use on the low -frequency television
mutual inductance. If the individual self-inductances band. This means that the antenna should be broad-
are equal and if the coefficient of coupling is unity, the band to all frequencies in this band, namely 44 to 88
value of the mutual inductance will be the same as mc. However these antennas are not usually satisfac-
either self-inductance. Under these circumstances, the tory when used on the high -frequency (174 to 216
total inductance of the autotransformer is equal to mc) part of the television band. For the high-fye-
four times the self-inductance of one part of the quency channels, a different antenna generally has to
winding. be used.
The same is true, for the most part, of the folded Since often some signals are stronger than others,
dipole. The individual self -impedances of the half - the antenna length should favor the weaker stations.
wave sections are about equal, and the coefficient of For most practical purposes, however, the length of
coupling between these sections is considered to be the antenna is designed for the center frequency of
unity, so that the mutual impedance is the same as the band of frequencies it is to receive. Accordingly,
either individual self -impedance. The total input im- the dipole antenna for the low -frequency channels is
pedance for the folded dipole is, therefore, the sum usually cut to 66 mc, the midfrequency of 44 to 88 mc,
of the individual self -impedance plus twice the mutual and for the high -frequency channels it is usually cut
impedance. Consequently, the total input impedance to 195 mc. This is based on the assumption that the
different television signals in each band are essen- mc, the effective electrical length will increase fur-
tially of the same signal strength. However this is ther, which means the physical length should be re-
not actually so. But in areas where the signal duced by more than five percent. For most practical
strengths are relatively strong, this analysis can he purposes at the frequencies within the low -frequency
applied. In those areas where one signal (in either television band, the physical length of the half-wave
the high- or low-frequency television band) is the antenna should be reduced by about 6.5 percent. This
weakest of the group, the antenna is often cut to this means that the preceding formulas, equations 3-15
frequency for maximum possible reception on the and 3-16, have to be multiplied by 93.5 percent (the
channel in question. We will now analyze the method difference between 100 and 6.5 percent) to give the
of computing antenna length. correct effective half wavelength.
It is known that.a 10 -meter wavelength means a fre- Thus,
quency of 30 mc, but the simple formula telling how 492 x.935 460
in feet Eq. 3-17
this is brought about is often forgotten. The wave- f (mc) f (mc)
length of a specific frequency is found by dividing this or
5904 x .935 5520
frequency into the velocity of radio waves.1 The veloc- in inches Eq..3-18
ity of radio waves is equal to 300,000,000 meters per
f (mc) -
f (mc)
second, and thus with the frequency, f, in cycles per where L is equal to the effective length of a half -wave
second, the wavelength, in meters, is given by the antenna.
following :
For most center lead-in half-wave antennas, each
half of the dipole is approximately equal to half the
Wavelength -
300,
in meters Eq. 3-14 values of L found in the foregoing formulas.
This is for one full wavelength. If we change the units With respect to the high -frequency television band
of this formula and divide the right hand side by 2, (174 to 216 mc) , the end effect is even more pro-
we will find that nounced. For most practical purposes at these fre-
492 quencies, the physical length of the half -wave antenna
in feet Eq. 3-15 should be reduced by about 8.5 percent. Consequently,
f (mc)
or equations 3-15 and 3-16 have to be multiplied by 91.5
f5(904) percent to give the correct effective half wavelength
L = in inches Eq. 3-16
me for the high -frequency television band. Thus :
where L is equal to the length of a half wavelength in 492 x .915 450

in feet Eq. 3-19
free space and f is the frequency in megacycles per f (mc) f (mc)
second. 5904 x .915 5400
In half -wave antennas, a so-called "end effect" is in inches Eq. 3-20
f (mc) f (mc)
attributable to the material supporting the antenna Let us now consider a practical example for each
and other physical factors, all of which make the elec- television band. Thus for the center frequency of 66
trical length of a half -wavelength antenna effectively mc of the low -frequency band, the length of a half -
longer than the physical length. To make sure that wave antenna would be (using equations 3-17 and
the electrical length of the antenna used is effectively 3-18) as follows :
one-half wavelength long, the actual physical length

is made less than that for a half wave in free space L= 46066
=6.97feet
which has no end effects. The end effect is. more pro- or approximately 7 feet, and
nounced at the higher frequencies, which means that
the effective electrical length increases with increase L - 555220
= 83.6 inches
in frequency. From about 5 to 30 mc, the physical or approximately 84 inches.
length of the half -wave antenna should be reduced by For the center frequency of 195 ntc of the high -
about five percent to make it operate effectively as a frequency band, the length of a half -wave antenna
half -wave antenna. Since the 44- to 88 -mc low -fre- (using equations 3-19 and 3-20) would be as follows :
quency television band has a center frequency of 66
1Many sources in establishing the wavelength of radio waves
L = 450
195
- 2.31 feet
state that the frequency of the wave is divided into the velocity or approximately 2.3 feet,
of light. However, since both light waves and radio waves are
electromagnetic waves, they basically travel at the same 54°°
L= = 27.7 inches
velocity, namely 300,000,000 meters per second. Hence, in the 195
above problem we have omitted the reference to the velocity or approximately 28 inches.
of light.
If it is desired to compute the antenna length for ends. Next, twist together and solder the two bare
frequencies other than those at the center of each leads on each end, as shown in Fig. 3-13. Then center
channel group. equations 3-17 through 3-20 can still cut one lead of the twin line and strip away the insu-
be used. All that has to be done is to insert the proper lation front each cut end to half the distance between
frequency into the denominator of the equations. the parallel wires of the transmission line. Finally
Channels 1 through 6 use equations 3-17 and 3-18 bend these two bare pieces of wire at right angles at
and channels 7 through 13 use equations 3-19 and the beginning of the insulator. This part of the an -
3-20. The frequency that is used for the channel de-
sired is the center frequency of that channel. For
channel 1, frequency range 44 to 50 mc, the frequency
to use would be 47 mc in equations 3-17 or 3-18. For f
channel 7, frequency range of 174 to 180 mc, the fre- SOLDER
quency to use would be 177 mc in equations 3-19 or SOLDER
HERE
HERE SOLDER
HERE
3-20. PLASTIC
TWIN 300
OHM LEAD
INSULATOR
Since each section of a dipole is effectively a quarter
wavelength long, the physical length of each such sec-
tion would be half that computed in the above ex-
amples. However, due to the gap in the lead-in dipole FIG. 3-13.-Construction details of a folded dipole antenna
made from a 300 -ohm twin -lead transmission line.
for the transmission line, the actual physical lengths
of the individual sections of the dipole are somewhat tenna represents a folded dipole. Next, take another
less than calculated.
piece of the 300 -ohm twin -lead line and strip off some
All the relations to the length of the half -wave an- of the plastic insulator at one end, so that the twin
tenna in this section apply equally as well to folded leads are bare. This latter twin lead represents the
dipoles as to simple dipoles. It has been found in many transmission line, and it is usually any length that
instances that for best results the length of the folded will make its wiring to the receiver input most con-
dipole should be made slightly shorter than its calcu- venient, but the shorter the better. Solder these two
lated value. bare leads to those of the center part of the folded
dipole as shown in Fig. 3-13. The impedance of this
Indoor Antennas
folded dipole antenna is about 300 ohms, similar to an
Li places where an outdoor antenna cannot be em- outdoor folded dipole.
ployed with a television receiver, an indoor antenna
can be used. There are a few different types of indoor Dipole with a Reflector
antennas on the market today. One of the most com-
mon types used is made of a 300 -ohm twin -lead trans- In many localities the television signal pickup re-
mission line as indicated in Fig. 3-12 (C). This type quired for proper reception is greater than that ob-
can be very easily constructed for any television band tainable with a half -wave dipole or folded dipole an-
or channel desired. A drawing of the construction is tenna alone, so that something has to be done to
illustrated in Fig. 3-13. The letter L on the drawing increase the signal pickup. This is especially necessary
refers to the length of the twin -lead transmission line when the receiver is located at a great distance away
for the frequency to which it is to be cut. This length from the transmitting antennas. Since the signal sur-
L can be computed through the use of equation 3-15 rounds the antenna, it is easily realized that signal
or 3-16 and a multiplying factor of 0.825 for both energy exists at points other than the immediate
television bands. vicinity of the dipole itself. This leads to the idea that,
For instance, if the indoor antenna is to be cut for if some of this signal energy from the surrounding
the center frequency of the low -frequency television area could be directed toward the antenna, the an-
band, then L would be equal to about 74 in. Thus a tenna would effectively have a greater signal input.
length of 300-ohm twin -lead line equal to a little more To increase this signal pickup effectively, the an-
than 74 in. is cut ; the extra length is used to twist the tenna employed is equipped with a "reflector" ele-
aids together. The general construction of this indoor ment. This is shown in Fig. 3-14 Where a simple half -
antenna is illustrated in Fig. 3-13 and is as follows. wave dipole is shown, and placed behind this dipole,
After computing L, take a piece of 300-ohm trans- in the same plane as the dipole, is the reflector ele-
mission line slightly, larger than the computed length ment. This reflector conductor is usually of the same
and strip away some of the plastic insulator from the material as the dipole itself, and it should be slightly
longer than the dipole. The reflector should be placed effective input impedance is altered. The exact value
on the side of the receiving antenna away from the of the change in the antenna's input impedance de-
transmitting antenna from which the signal is to be pends upon the degree of coupling, the value of
received. This means that the desired signal will be mutual impedance between the elements and, hence,
the amount of reflected impedance. When used with
a simple half -wave dipole or folded, dipole, the re-
REFLECTOR DESIRED SIGNAI flector usually decreases the input impedance, so, be-
BEING REFLECTED
sides increasing the signal pickup, the reflector can
change the input impedance to, perhaps, a better im-
pedance match from the antenna to the receiver.
I7 MAST
In Fig. 3-15 are illustrated some typical receiving
antennas as used today. In part (A) the reflector is
SIMPLE
DIPOLE used with a simple half -wave dipole element and in
part (B) the reflector element is used with a folded
TRANSMISSION dipole element.
LINE DESIRED SIGNAL
Since the reflector increases the signal pickup of
the dipole element, it is customary to state how much
FIG. 3-14.-In order to increase the pickup, a reflector ele- this signal pickup is increased over the dipole element
ment is placed behind the dipole in the same plane at a (lis -
tance 1), that is between one -tenth and one -quarter of the alone by so many decibels gain. Consequently, the re-
wavelength for which the dipole is cut. flector element is said to add so much decibel gain to
the antenna.
approaching the antenna dipole in the direction indi-
cated in Fig. 3-14. That part of the signal that passes
the dipole and hits the reflector conductor will be re- DIPOLE DIPOLE
flected back to be picked up by the dipole. REFLECTOR
The distance D that the reflector is spaced from the REFLECTOR
dipole is a factor in the amount of increased signal GUIDE LOOPS
pickup, and it is usually somewhere from one -tenth to TRANSMISSION LINE
(A)
one -quarter of a wavelength away from the feed-in FOLDED DIPOLE
dipole element. Most manufacturers specify the spac-
ing in their service instructions accompanying the
antenna. This is to make sure that the reflected signal TRANSMISSION LINE (B)
picked up by the receiving dipole is aiding the signal
directly picked up by the same dipole, so that the (A) Courtesy IIIS,fI ne Corp. of America
(B) Courtesy American Phenolic Corp.
maximum possible total energy pickup is available.
FIG. 3-15.-Reflector elements are used not only with simple
The *ay in which the signal is increased can also dipole antennas (A), hút also with folded dipoles, as shown
be compared with transformer action. A certain in (B).
amount of mutual impedance exists between the di-
pole and reflector, determined primarily by the dis- When used with a reflector, half -wave antennas be-
tance separating the elements and the self-impedance come unidirectional (that is, become more directional
of the elements. In brief, when the signal hits the re- to signal pickup in one direction than any other) be-
flector, a voltage is induced which causes a corre- cause there is very little signal pickup from the re-
sponding current to flow in the reflector. This reflector flector side of the arrangement. In other words, the
current, by analogous transformer action, induces a reflector element prevents those signals approaching
voltage into the lead-in dipole element, the magnitude the rear of the antenna from being picked up by the
depending upon the mutual impedance and the phase dipole.
relation of the voltage depending upon the spacings This action of the reflector makes it extremely use-
between the elements, which is often one -quarter ful in preventing pickup of undesired signals by the
wavelength. Thus, it is seen how a receiver antenna dipole element. In many localities reflectors are not
with a reflector can have an effective increase in signal needed to increase the signal pickup because the signal
pickup over a half -wave antenna without a reflector. strength in those areas is relatively strong. However,
Since a mutual impedance exists between the ele- they are often used in such areas to prevent pickup of
ments and since the signal pickup is increased, the undesired signals.
Direct, Reflected, and Blocked Waves these structures. Waves thus diverted are commonly
The use of a reflector element is recommended for known as reflected waves. In the drawing of Fig. 3-16
those localities where there are surrounding buildings, the reflected signals are picked up by the receiving
hilly terrain, or other signal obstruction mediums antennas.
even if the signal pickup in the area is relatively As long as the dipole element (that which has the
strong. This is primarily to prevent the pickup of transmission line attached) will intercept a television
signals that may become reflected when hitting one signal, no matter where it comes from, reception is
of these obstructions. If such a reflected signal were possible. Thus reception by means of either direct or
to be picked up by the dipole element, there is the reflected waves or even a combination of these signals
very likely possibility that "ghosts" on the picture is possible. In Fig. 3-16 antenna A is seen to be re-
tube would occur. Before discussing what these ghosts ceptive to only a reflected signal because reception of
are and how they are formed, let us study the differ- a direct wave is blocked by the building it is on. How-
ent paths by which the transmitted television signal ever receiving antenna B is receptive to both direct
may reach the receiver. and reflected signals.
In Fig. 3-16 a number of direct waves are illus- If the reflector element used on antenna B is so
trated emanating from a television transmitting an- oriented that the reflected wave will be coming di-
tenna. Two television receiving antennas are also rectly toward this reflector element (that is, the re-
shown in the drawing. Receiving antenna A does not flector element is broadside to the reflected signal),
receive any direct wave because the building on which then this reflector element will greatly reduce pickup
it is located blocks all these waves. A direct wave that of this reflected signal by the dipole element. As men-
is obstructed by some medium and is thus unable to tioned, when an antenna receives both direct and re-
reach an antenna is termed a blocked wave. Such a flected waves of the same transmitted frequency, then
wave is illustrated in Fig. 3-16. Of the other three ghosts on the picture tube of the receiver will prob-
direct waves illustrated in the drawing, one makes ably result. This is especially so when the signal
direct contact with receiving antenna B and the others strength of the reflected signal is appreciable com-
strike two buildings at an angle and are reflected from pared to the direct signal.
RECEIVING
ANTENNA
A
B`ocKED WAVE
TRANSMITTING _________ -1110.-------
ANTENNA
----------- RECEIVING
ANTENNA
s.
DIRECT % REFLECTED
WAVES S WAVE
3-16.-Receiving antenna A picks up only a reflected wave because the building it is on blocks the direct waves. Receiving
antenna B is in the open and receives both a direct wave and a reflected wave, thus creating ghosts in the television receiver.
Ghosts or Multiple Images ing antenna B is shorter in length than the path of
the signal leaving the transmitting antenna, reflecting
We have used the word ghosts a few times in this from a building, and finally reaching receiving an-
chapter without stating exactly what they are. Ghosts tenna B. This difference in the length of the signal
need no introduction to many readers as they are paths means that one signal will reach the receiving
quite familiar with this type of interference and know antenna before the other. The time difference in signal
that it is essentially caused by reflected signals. Since pickup produces ghosts or multiple images. The
this is still an important problem in the reception of greater the number of reflected signals picked up, the
good television pictures and also because many other greater the number of images on the screen.
readers do not fully understand the different methods
In Fig. 3-17 (B) the ghosts appear to the right of
by which ghosts are formed, a discussion of ghosts
the main picture and this form of multiple image re-
is in order.
production is known as "lagging ghosts." In other
In brief, ghosts are multiple images formed on the
words, the reflected signals (considered as the unde-
screen of the picture tube. In Fig. 3-17 (A) a normal
sired signals when a direct signal is also picked up)
picture is illustrated and in Fig. 3-17 (B) is a pic-
are lagging in time with respect to the main or direct
ture showing multiple images or ghosts. In general
signal. When ghosts are due to reflections from sur-
ghosts are caused when the dipole element picks up
rounding buildings, the use of a reflector element and
two or more signals of the same transmitted fre-
proper orientation of the receiving antenna is the best
quency arriving at the receiving antenna at different
remedy for getting rid of such ghosts.
instants of time. For instance in Fig. 3-16, the direct
wave from the transmitting antenna takes a certain Ghosts may be caused in a television receiver by
amount of time to reach receiving antenna B and the two other methods. In each case the ghosts are formed
reflected wave (if picked up) takes a different amount because the signals arrive at the input to the first tube
of time to reach receiving antenna B. Due to the dif- of the receiver at different instants of time. The mul-
ference in time between the two signals, two separate tiple images caused by one of these methods are
images will appear on the screen of the picture tube, known as "leading ghosts." In other words, the "un-
one for each signal received. desired" signal arrives at the input to the first tube
The direct signal is the strongest of the received of the receiver before the "desired" signal. This situ-
signals and the reflected signals are the ones that pro- ation is illustrated in Fig. 3-18.
duce the ghosts. Remember that each ghost is an Both signals involved are direct waves, but one
exact image of the direct signal because all the signals signal is picked up by the antenna and the other sig-
causing images on the picture tube leave the trans- nal is picked up directly at the input to the receiver.
mitting antenna at the same time. Due to certain ob- Since the path for signal A from the transmitting an-
structions some signals are reflected and therefore tenna to the receiver input is longer than that for
take a longer path to the receiving antenna. This is signal B, there will be a time difference between the
readily evident in Fig. 3-16, where the path of the reception of each of these signals. Ghosts will, there-
direct wave from the transmitting antenna to receiv- fore, result, especially if signal B has appreciable sig-
(A) (B)
Courtesy R.C.A.
FIG. 3-17.-A normal picture is shown at (A) and a picture with ghosts at (B).
nal strength when it reaches the input of the television cepted by the receiver, thus causing a ghost. This
receiver. At the input to the receiver, signal A will type of ghost will become more evident if the mis-
usually be the stronger because the antenna, which is matched transmission line is quite long. In many
a broadly tuned circuit, has a better response to signal instances, however, well-defined images are not read-
A than the receiver itself to signal B. This means ily evident, but a blurring of the pattern is the result.
that the ghost or image caused by signal B will be Consequently we can see that besides causing loss
displaced to the left of the main picture, and the ghosts in signal pickup, improper impedance matching of the
in this instance are known as leading ghosts. In other transmission line to the receiver can also cause double
words, the undesired signal (B) will be reproduced images or blurring of the scene on the picture tube.
in the picture tube first and the desired and stronger A good impedance matching system using as short a
signal will be lagging the image, or ghost, in this case ; transmission line as possible, is the best remedy for
this also means that the ghost is leading the desired this type of trouble.
signal. This is the reverse of Fig. 3-17, where lagging
ghosts are illustrated. The formation of leading ghosts Maximum Voltage Input
is not as common as that of lagging ghosts. The best
remedy in preventing these leading ghosts is to have It is well known that a maximum voltage input to
the television receiver well shielded so that reception the receiver is desired, but the feed-in to the simple
of signals directly through the receiver will be half -wave dipole antenna or folded dipole is center
negligible. driven and at this point the current is effectively at a
Ghosts may also be caused by improper impedance maximum. How, then, can we conceive of a maximum
matching of the transmission line to the receiver. voltage input to the first r -f tube ? There are numer-
Standing waves are consequently set up in the trans- ous way of explaining this, one of which follows.
mission line because reflections from the receiver will In Fig. 3-19 a dipole antenna and input circuit to a
travel back to the antenna. When these signals hit the television receiver is illustrated along with the current
antenna proper, they act effectively as a reflected and voltage curves effective at the dipole. Since the
signal and enter the receiver retarded in time with center point of the antenna is at the loop of the current
respect to the original signal. In effect then, two sig- curve, a maximum amount of current will flow
nals of the same type but differing in time are ac - through the primary of the input transformer. This
RECEIVING
ANTENNA
SGNAL P _ -
TRANSMITTING
ANTENNA
TRANSMISSION
LINE
r7,4 rinr,
TELEVISION
RECEIVER
3-18.-Two signals are received at the television receiver, one is picked up by the antenna and the other signal is picked up
directly by the receiver.
maximum amount of current sets up a maximum This reduction in noise pickup is only in reference
magnetic field which causes a maximum voltage across to that picked up by the transmission line and not that
the secondary due to induction. Since the parallel noise picked up by the dipole itself. This latter noise,
tuned secondary circuit is resonant at the frequency as well as the desired signal input, finds its way into
the receiver.
VOLTAGE
INPUT
CURRENT
HALF -WAVE
TUBE
Installation and Orientation of Antennas
DIPOLE
INPUT
It is beyond the scope of this book to discuss the
TRANSFORMER
installation and orientation procedures for all the dif-
ferent television antennas manufactured. The follow-
ing notes, however, cover outdoor receiving antennas
TRANSMISSION and apply to all types, unless otherwise specified.
LINE
One of the primary requisites in all installations is
Ftc. 3-19.-Half-wave dipole and input circuit to the r -f that the antenna system be mechanically secure. The
amplifier. The arrows indicate the current flow, which being
a maximum in the transformer primary, induces a maximum first necessity is to make sure that high winds, rain,
voltage in the secondary at the frequency of operation. ice, or snow will not affect the mechanical secureness
of the system. The reason for this is obvious, since it
of operation, maximum voltage is developed at this
is desirable that the antenna require little attention
desired frequency, which is applied to the grid of the once it is installed. Most types of antennas are sup-
input tube. plied with the necessary mechanical supports to at-
tach the system to some object such as a roof, chim-
Noise Reduction
ney, or side of a house.
In many television antenna systems, the primary of The antenna should be placed as high as possible
the input transformer has its center point grounded and away from any interfering objects which might
to reduce noise interference through the medium of reflect the television signals. Consequently, such high
the transmission line. This was illustrated in Fig. 3-19, installation points as roof tops are used. The antenna
but the left-hand side of this figure is redrawn in also should be mounted some distance away from any
Fig. 3-20, to make this system of noise reduction of the metallic objects commonly found on a roof top,
somewhat clearer. Since the. transmission line used such as water drains, pipes, and wires. It is quite as
may cover a greater area than the antenna itself, it important to make sure that the transmission line
has a tendency to pick up noise voltages, especially if used (especially if it is the twin -lead type) is also
the transmission line is quite long. To reduce this kept away from metallic objects. By placing the trans-
noise pickup and, hence, increase the signal-to-noise mission line near such objects, noise and other types
ratio at the input to the television receiver, the center of undesired reflected signals can be picked up by the
tap of the primary is grounded. It reduces noise pick- line.
up in the following manner.
The noise signal, when it hits the transmission line, - TRANSMISSION
induces equal voltages in each lead of the transmission LI NE
line which in turn produces noise currents that flow in
the same direction in the transmission line as indi-
cated in Fig. 3-20. By center tapping the primary of
the input transfonner to ground, this circuit becomes PRI. OF
symmetrical, and the noisé currents both flow toward INPUT
NOISE T RAN SF.
this ground connection. This effectively makes the PICK-UP
BY
individual currents out of phase, and, since they are TRANSMISSION LINE
equal in magnitude, they produce magnetic fields
which cancel each other and, hence, the total noise FIG. 3-20.-In order to reduce noise pickup through the
voltage induced in the secondary of the input trans- medium of the transmission line, the center of the primary
winding of the r -f transformer is grounded. This increases the
former is virtually zero. signal-to-noise ratio at the input of the receiver.
Once the spot with the most freedom from metallic the receiving or transmitting sense, is a conventional
objects has been chosen, two remaining important figure -eight pattern, as shown in Fig. 3-21, with the
factors must be taken into account the orientation of
: points of maximum and minimum signal pickup 90°
the antenna proper and the correct installation of the out of phase with each other. Therefore, when going
transmission line from the antenna to the receiver. from a minimum signal pickup to a maximum, the
Because of the horizontal polarization of the radiated antenna should be rotated by 90°. This field pattern
television signals, the radiating elements of the re- shows that the dipole is bidirectional.
ceiving antenna must be in a position to receive the The installation of the transmission line is as.im-
maximum possible signal pickup. Most television re- partant an item as the antenna itself. The transmission
ceiving antennas, especially in installations containing lines most commonly used, as previously mentioned,
dipole elements, are so oriented that the elements are are of two types : the twin -lead line (with plastic in-
horizontally situated. sulation) and the coaxial line. Both types can be
In using antennas that are unidirectional or bi- supplied with different characteristic impedances. The
directional, it is important to know something about importance of correct impedance matching for maxi-
the television transmitting stations within the cover- mum energy transfer should be remembered, espe-
age area of the receiver in question. If the field inten- cially in localities where the received signals are weak.
sities of most of the transmitting stations within the One advantage of the coaxial line is that it has low
vicinity of the receiver are about the saine, not much inherent signal loss per section and low noise pickup,
difficulty will be encountered in orienting the antenna. as compared with the twin -lead type. However, the
If most of these stations are quite near each other as coaxial type is primarily used for low impedance
far as the receiver is concerned, the receiving antenna matchings because high impedance coaxial lines are
can be oriented simply by placing the pickup element not available for these high -frequency impedance
broadside to the oncoming television signals. The re- matching systems. The twin -lead type of transmission
ceiving antenna thus serves its most useful purpose line at 300 ohms is the kind generally used for high
in trying to obtain equalized signal pickup as nearly impedance matching. Some of the factors that help
as possible for all of the television stations in its area. determine the type of transmission line to use are the
If the antenna is placed in a different position there type of antenna (high or low impedance), input im-
may be a noticeable loss in signal pickup when the pedance of receiver, and length of transmission line
receiver is in operation. This is particularly true when needed.
the receiver is quite a distance from the television sta- In attaching either type line from the antenna to the
tions. However, as pointed out, in localities where the receiver, the length of line used should be as short and
receiver is in the vicinity of strong local stations, as rigid as possible. The requirement for a short line
variation in the position of the antenna may make arises from the fact that the longer the line, the more
little difference. signal will be lost. The rigidity requirement is neces-
If the strengths of the various television signals in sary to prevent the line from swaying with the wind.
the locality of the receiver in question are quite dif- A number of things may happen if the line does sway
ferent from each other, and if they approach the in the wind. The line may be moved near some metal-
receiver from different directions, it is advisable to lic object, it may be scraped constantly against some-
thing which will make for quick deteriorating of the
insulator and then perhaps shorting of the line ele-
ments, or the line may even be snapped by a strong
wind.
The mast and supports of the antenna should have
FIG. 3-21.-The ideal field pattern of
a half -wave dipole. Note that each side some provision for stand-off insulators or guide loops,
of the pattern is identical. so that the portion of the transmission line leading
from the antenna can use these insulators or guide
loops as a means of preliminary support. These spe-
cial attachments help maintain rigidity in the line and
prevent any unnecessary swaying. Some manufac-
orient the antenna broadside to the signals that are turers suggest that, when passing twin -lead line
weakest, in order to have adequate signal pickup on through these loops that are connected to metal masts,
most stations. the line should be twisted two or three times to main-
The field pattern for a dipole, considered either in tain electrical balance between each wire of the line
and the metal mast and thus minimize noise pickup. respect to their price. They are trying to be as fair as
However, these twists in the line should be made only possible to the public in establishing what they believe
in the vicinity of the metal mast. to be an honest account of what they have put into the
After passing through the insulators or guide loops, antenna in the form of engineering design as well as
material things.
the line should be drawn tight until it is attached to
the receiver. In doing this, the line may be pressed
In Fig. 3-22 is illustrated an antenna known as the
Di -Fan. This antenna offers broadband characteristics
flat against only those types of objects that will not
influence its transmission characteristics. Such things
as wood, brick, and roof shingles (no metal involved)
are commonly used. To secure this tightness in the
line, special guide loops or other such attachments are
often attached to some place in the vicinity where the / /
transmission line is running. Twin -lead line may be '
fastened at a number of points by using small metal
brads (that is, nails without any heads) through the
center of the plastic insulator and into any material
that will take the nail, provided the material is not
_.-
metallic. The brads themselves have a negligible ef-
fect on the transmission of the signal along the line.
However if nails with heads or similar objects are
used, difficulties may arise since a wearing effect
on the insulator may occur and the line may be short-
circuited, or the characteristic impedance of the line Courtesy Andreae Co.
at the point where the nails pass through it may be FIG. 3-22.-A special type of antenna known as the Di -Fan.
changed, causing a mismatch from the antenna to the It offers broadband characteristics over the two television
receiver. bands and the f -m band.
over the complete range of the two television bands

Other Types of Television Antennas
and hence also includes the f-m broadcast band of 88
So far in the analysis of television receiving an- to 108 mc. The nominal impedance of this type of
tennas we have dealt primarily with simple and folded antenna is about 300 ohms and it is therefore suitable
dipole antennas, with and without reflectors, because for use with a 300 -ohm transmission line.
they are the types that are most universally used. The antenna illustrated in Fig. 3-23 is known as
There are, however, numerous other types of antennas the double -V. This type of antenna, when cut to the
used for the reception of television signals. Some of
them are no more than slight variations of the an-
tennas already analyzed. Many of the design prin-
ciples and the theory involved have already been dis-
cussed in general form as applicable to all antennas.
These other types of antennas are usually designed for
one or more of the following reasons : better signal
pickup, omnidirectional signal response, and better
broadband response.
Some of these antennas may be higher priced than
those previously discussed. In other words, if more
time and equipment is put into the design of television
antennas, the price will be high compared with other
Courtesy Technical Appliance Corp.
antennas. Thus the television receiver buyer, or even
FIG. 3-23.-The double -V antenna has a broadband char-
the television manufacturer buying antennas, is often acteristic.
faced with limitations on how much he can economi-
cally afford to spend on antennas. Therefore, most proper frequency in either television band, is said to
antenna manufacturing companies refer to the elec- give a fairly good frequency response over the entire
trical advantages or superiority of their product in band. When designed to be used with the low-fre-
quency television band it can also be used for the f -m and essentially consists of one simple dipole and re-
band. flector arrangement mounted above the other. The
The stacked dipole, the so-called H -type antenna, dipole elements are, however, connected together at
is shown in Fig. 3-24. This type of antenna essen- their center points as in the H -type antenna. The
tially consists of two center -fed dipoles mounted one reflector elements are added to help prevent ghosts
above the other with their center points connected as and to make the antenna unidirectional. The over-all
shown. The transmission line is attached to the parts gain of this antenna is somewhat greater than the
connecting these center points. This antenna, like a H -type because of the addition of the reflector ele-
single dipole, is bidirectional. Mounting one dipole ments. It is recommended by the manufacturer that
over the other allows for greater signal pickup in a this antenna be used in areas where there are a num-
broader horizontal plane. In other words, this system ber of reflected signals and also in noisy areas where
provides greater gain than the simple dipole. Since the signal strength is low.
There are antenna systems appearing on the market
that appear to be a dipole and reflector arrangement,
but the two elements are spaced somewhat closer
together than the usual dipole-reflector arrange-
ment. In most cases these antenna systems use the
element with the transmission line attached as a
director and not a reflector. Director is the name given
to an antenna parasitic element that is oriented in the
direction of the signal to be received. The exact spac-
ing between the dipole and director is determined by
the physical makeup of the antenna system and sur-
rounding territory. In Fig. 3-26 a dipole and director
Courtesy Technical Appliance Corp.

FIG. 3-24.-The stacked dipole, or H -type antenna, is bi-
directional and provides greater gain than does a simple dipole.
higher gain is obtainable, this antenna is recommended

by the manufacturer to be used in localities where the SPACING
VARIABLE
signal strength is low. 0.1 T0
WAVELENGTH
0.15
The stacked dipole of Fig. 3-24 is put to further

use by employing two reflectors with the unit. This is
shown in the antenna drawing of Fig. 3-25. This an-
tenna is called the double-doublet by its manufacturer
Courtesy Shur -Antenna-Mount, Inc.
FIG. 3-26.-The distance between the dipole and director is
variableso that best use of the director can be attained when
the antenna is installed.
system is illustrated ; the distance between the two

elements is variable so that the maximum use of the
director can be attained wherever the antenna is in-
stalled. The director element is made slightly smaller
in length than the dipole element.
The input impedance of many such systems is quite
low, and special impedance matching sections very
often must be used in conjunction with the trans-
mission line so that the maximum amount of eñergy
Courtesy Technical Appliance Corp. transfer from the antenna to the receiver can be at-
FIG. 3-25.-The double -doublet antenna essentially consists tained. Most manufacturers incorporate the necessary
of a simple dipole and reflector arrangement mounted above equipment for correct impedance matching, or include
another one. The reflector elements are added to help prevent
ghosts and to make the antenna unidirectional. information as to how it may be accomplished.
The signal pickup can be further increased by an element is made smaller by this method, thereby
antenna system employing both director and reflector making the antenna frequency response broader. This
elements in conjunction with a dipole. A drawing of special shaping also adds to the rigidity of the antenna
such a system is illustrated in Fig. 3-27, where the
dipole is the center element. Although not readily
Courtesy L. S. Brack Mfg. Corp.

FIG. 3-28.-Each element is a simple dipole, folded back on
itself and shorted at the ends. The longer element is for the
low -frequency television band and the shorter element for The
high -frequency band.
system. The longer antenna is for the low -frequency

television band and the shorter antenna for the high -
f requency band. To prevent interaction between an-
tenna elements, they are separated by special sections
called "dividers." These dividers are the black por-
tions of the vertical rods connecting the two antenna
elements as seen in Fig. 3-28. The antenna elements
and dividers are so designed that by connecting the
Courtesy Consolidated Television Corp. transmission line above the dividers, good reception
Ftc. 3-27.-The director is placed in the path of the incom- on both television bands will be achieved.
ing signal and the reflector is placed behind the dipole. In- All the antennas discussed so far were either bi-
creased signal pickup and great directivity are provided by
the use of this system. directional or unidirectional. There are some an-
tennas however, that are omnidirectional in that they
evident in this drawing, the director element is slightly will respond to signals from all directions. Such an
shorter than the dipole and the reflector element
slightly longer than the dipole. The director is placed DIPOLE
in the direction of the television signal. This type of
antenna system, in addition to providing increased
signal pickup, increases the directivity as compared to
the dipole reflector and dipole -director antennas. That TOTAL EFFECTIVE
is, this antenna is unidirectional and the reflector ele- F IELD PATTERN
ment also helps eliminate ghosts.
There are a number of television antennas on the
market today that are so designed that they contain
two separate dipole elements, one for each television
band. The element designed for the high -frequency DIPOLE
band (channels 7-13) is smaller than the element de-
signed for the low -frequency band (channels 1-6). In
most cases the elements are cut for the center fre-
quency of the band in question.
Fig. 3-28 illustrates one such antenna. Upon first
inspection, this antenna appears as though it consists
of two folded dipoles. This is not the case. Each ele-
ment is a simple dipole that is folded back on itself FIG. 3-29.-The total effective field pattern of two crossed
in a U shape and shorted at the ends. The Q of each dipoles is virtually circular in shape.
antenna is, of course, only suitable in regions where Two crossed dipoles are drawn in Fig. 3-29 in
there is very little chance for reflected signals and conjunction with the field patterns of each individual
thus little possibility of ghosts. Omnidirectional an- dipole. The total effective field pattern is virtually
tennas have been used a great deal with f -m receivers circular in shape due to the addition of the field pat-
with very good results. These antennas essentially terns of the individual dipoles making up the antenna.
consist of two dipole elements both of the same size, Such an antenna is definitely suitable in areas where
at right angles to each other. Their lengths are cut to there are a number of different television signals com-
the center of either band, or whatever other frequency ing from different directions. In such areas, special
may be desired. Their center -fed points are usually orientation of the antenna to the weakest signal, as is
individually connected to transmission lines, which in usual with bi- or unidirectional antennas, is not nec-
turn are combined in a single line for whatever the essary because the antenna will respond equally to all
impedance desired. signals.
TWO OR MORE RECEIVERS ON ONE ANTENNA

FOLDED
DIPOLE
300 - OHM
TWIN LEAD
300(1
°PDT TV SET 30011

BALANCED INPUT
30011
150(1 150(1
(A) (B) EQUIVALENT CIRCUIT

SERIES RESISTANCE FOR
TWO
THREE
FOUR
RECEIVERS
RECEIVERS
RECEIVERS
-
-- -. 150.(
300(1
450(1
7511
TO
TV SETS
WITH 7511
UNBLANCED
INPUT
75-OHM
COAX
LINES
(A) (B) EQUIVALENT CIRCUIT
SERIES RESISTANCE FOR

TWO RECEIVERS
THREE
FOUR
RECEIVERS
RECEIVERS
-+
-e-
7511
1501E
22511
You must know the relative signal strength in your area so you can tell if 2, 3, or 4 receivers can be connected to one antenna,
as shown above. For two sets, the available signal voltage is cut 2 to 1; for three, it is cut 3 to 1, and for four sets, it is 4 to 1.
Use carbon resistors, as wire wound may unbalance the system. Place switches and resistors in a control box. All leads must be
short and direct. If a set is removed from the system, throw switch to dummy load.
CHAPTER 4
R -F AMPLIFIER, OSCILLATOR, AND MIXER CIRCUITS
BY HENRY CHANES
The front end of a television receiver includes vision receivers in the same neighborhood. This is
the r -f amplifier, if any, the oscillator, and the especially objectionable in large cities where there
mixer. Usually these circuits are assembled together are as many as 40 or 50 apartments in a single
as a tuner which supplies the i -f inputs for the building. If each apartment contained a television re-
video and sound i -f amplifiers. The functions of ceiver, the problem of local oscillation could indeed
the r -f circuits in a television receiver are similar be serious.
to the functions of the corresponding circuits in the One of the objections to the use of an r -f ampli-
conventional superheterodyne receiver. The r -f fier is that it increases the number of tuned cir-
amplifier selects the desired signal, amplifies it, and cuits. If a switch is used, this means adding a large
rejects all undesired signals. This amplified signal number of switching elements and the components
is combined in the mixer tube with a locally gen- of the tuned circuits. If continuous tuning is used,
erated signal of a frequency that differs from the in- the problem of tracking is complicated. These com-
coming signal by a fixed amount. When the two plications result in a higher -priced receiver. There
signals are mixed, a beat frequency is produced in is also the fact that the gain of an r -f amplifier in
the output of the mixer. This beat frequency, which a television receiver is comparatively low and that
is equal to the difference between the two signals, much higher gain can be achieved if an extra stage
is known as the intermediate frequency (or i.f.) of intermediate amplification is employed.
and is fed tc the input of the i -f amplifiers. This, Generally the sound and picture carriers are not
essentially, is the function of the front end in the separated before the i.f.'s are obtained. Therefore
television receiver, and so far it is the same as the it is necessary for the r -f amplifier and the mixer
front end of an ordinary sound superheterodyne tuned circuits to have a bandwidth sufficient to pass
receiver. However, because of the higher frequen- the entire television channel of 6 mc. (The bandwidth
cies and bandwidths required by television, the front of a television channel is discussed in chapter 2 and
end design of a television receiver becomes some- illustrated in Fig. 2-4.) This bandwidth is not too
what more complicated. difficult to obtain at the frequencies used for television
channels. Even at the lowest channel, which is from
44-50 mc, the 6 -mc bandwidth is not too great a per-
The R -F Amplifier centage of the carrier frequency. The tuned circuits
As in the case of sound receivers, not all televi- themselves are usually wide enough to take care of
sion receivers employ an r -f amplifier before the this frequency range. If not, resistors can be put
mixer. In the receivers that do not, the signal from across the tuned circuits to load them down and in-
the antenna is fed directly to the input circuit of the crease the bandwidth.
mixer. There are several advantages to be gained Another very important consideration in the de-
by the use of an r -f amplifier. sign of the r -f amplifier is the signal-to-noise ratio.
1. The selectivity of the receiver will be in- At the first stage in the receiver, the signal is at a
creased, resulting in a greater rejection of image very low level, and any noise developed in this stage
frequencies. will be amplified along with the signal and if the
;
2. The signal-to-noise ratio is better when an r -f noise is a large enough percentage of the signal, it
amplifier is used before the mixer stage. This is an may mask out the picture entirely. For a good pic-
important factor in the performance of a television ture, the noise level should be about 40 db below
receiver and will be discussed in detail later on. the signal level. By noise we do not mean the noise
3. The r -f amplifier decreases the amount of sig- picked up by the antenna or transmission line, that
nal that will leak through to the antenna from the is, atmospherics and man-made noise ; that noise
local oscillator. This signal might be radiated from should be taken care of by a suitable antenna and
the antenna and cause interference with other tele- transmission line. We are now considering the
55
noise actually generated in the r -f amplifier or, if The inductance of the primary is usually very small
an r -f amplifier is not used, the noise generated in compared with that of the secondary, and its effect
the mixer. This noise is caused by both the thermal on the tuned circuit will be negligible. The reso -
and shot effects. Noise due to thermal agitation is fiant frequency of this input circuit will be deter-
due to the fact that the electrons in a conductor are mined by the inductance of the secondary L and
always in motion. When no current is flowing the shunting capacitance C. This capacitance is
through the conductor, this motion is random and the made up of the input capacitance of the tube, stray
amount of motion will increase with the tempera- capacitance due to the wiring, and the minimum ca-
ture of the conductor. When current is flowing, the pacitance of a tuning capacitor if one is used. As in
average flow of electrons will be in one direction, any resonant circuit, tuning may be done by vary-
but a random component of motion due to thermal ing either L or C. Either one or both may be a pre-
agitation will still exist. This noise will increase as set control which is adjusted for a particular sta-
the temperature, impedance, or bandwidth is in- tion and switched into the circuit when desired, or
creased. one may be variable for continuous tuning. The reso-
The other component of the noise generated in nant frequency of this circuit is given by the ex-
the r -f amplifier is due to shot effect. When the pression :
electrons leave the cathode in a vacuum tube, they = Eq. 4-1
do not flow in a smooth stream to the plate but fr 2 w N/LC
leave in small groups. These groups cause pulses
of current in the plate circuit that constitute a ran- where :
dom noise. The strength of this noise signal in- = resonant frequency, in cycles
fr
creases with an increase in the impedance in the L = inductance in henrys
plate circuit, the current, or the bandwidth. These C = capacitance in farads
two sources of noise in the r -f amplifier, rather = constant = 3.14 approximately
than the amount of gain that can be achieved in a Usually the losses in the secondary will be great
television receiver, are the limiting factors on the enough to give this circuit sufficient bandwidth if
;
amount of signal necessary at the input of the re- not, resistance can be shunted across the secondary
ceiver in order to produce a good picture. Taking winding. When continuous tuning is employed,
these factors into account, and assuming a mini- several problems are encountered because of the
mum amount of man-made noise, a signal strength range of frequencies that have to be covered. The
of 500 microvolts at the input to the receiver should highest frequency in the television band is almost
provide satisfactory performance. In noisy loca- five times the lowest frequency. If C is made a
tions where the interference is great, larger signals variable, the minimum value of a capacitor with a
than this may be necessary. suitable range will cause a decrease in voltage gain,
and in order to keep a constant gain over the band,
Input Circuits the mutual inductance between the primary and the
secondary will have to be changed as the circuit is
The antenna, or the transmission line if one is tuned to different frequencies. If L is the variable,
used, must be matched to the input of the r -f ampli- there is the problem of getting sufficient change in
fier, or the mixer, if the receiver does not have an inductance to cover the frequency range. These
r -f amplifier. If the line is not properly matched, problems have been solved by the use of special cir-
reflections will occur that will interfere with the cuits that will be discussed.
signal and result in a blurred picture. The simplest If the input is balanced, as in Fig. 4-1, the cen-
type of coupling circuit consists of a transformer ter tap of the primary winding is grounded. An in-
with a tuned secondary connected to the grid of the terfering signal would tend to induce equal volt-
ages in the two wires of a transmission line, since
they are closely spaced. This would cause currents
Fig. 4-I.-The sim- that are equal but opposite in polarity to flow in
plest type of trans-
former coupling is each half of the primary. The effect of the two
the single -tuned cir- would cancel and the interfering signal would not
cuit.
appear on the secondary. The signal due to the an-
tenna would not be canceled because the antenna
tube. The primary, which is untuned, is connected produces a voltage on the transmission line that is
to the transmission line. This is shown in Fig. 4-1. balanced to ground. That is, the instantaneous volt-
R -F AMPLIFIER, OSCILLATOR, AND MIXER CIRCUITS 57
ages on the two wires of the transmission line are through a single -tuned circuit consisting of C1 and
always of opposite polarity. The currents in the pri- L1. This circuit is tuned to the same frequency as
mary of the input transformer due to these voltages the input circuit of the r -f amplifier, and the switch-
would add up and produce a voltage in the second- ing or tuning considerations mentioned will also be
ary. If a coaxial type of transmission line is used, true of this circuit. The signal is then coupled to
the input is not balanced and one end of the pri- the mixer tube by the coupling capacitor Co. If the
mary winding is grounded. In this case, interference circuit were double -tuned, the secondary of the
pickup by the transmission line is prevented by the transformer would be connected to the mixer grid.
shielding provided by a coaxial cable. In this respect, in this particular circuit the output of the local
the coaxial cable is more effective than the open -wire oscillator is coupled to the same grid by the capacitor
line with a balanced input but is also more expensive. Co. In some circuits inductive coupling is used be-
The gain of the circuit shown in Fig. 4-1 may be tween the oscillator and the mixer. The two signals
improved by tuning the primary as well as the are mixed to produce a third frequency which is the
secondary of the transformer. This circuit is shown difference of the two. The primary and the secondary
in Fig. 4-2. Both circuits are tuned to the same res - of the output transformer, COLE and CoLs respec-
tively, are both tuned to the difference frequency. The
output is the intermediate frequency or i.f. and is fed
Fig. 4-2.-The gain to the i -f amplifiers.
of this double -tuned
circuit is improved The local oscillator is operated at a frequency
over that of Fig. 4-1. above the picture and sound carriers. Since the
sound carrier is always 4.5 mc above the picture car-
rier, the sound i.f. will be 4.5 mc below the picture
onant frequency which again may be determined
i.f. Let us consider the second channel 54 to 60 mc,
by Eq. 4-1. While this circuit has greater gain than
as an example. The picture carrier is 55.25 mc and
the single -tuned circuit, it has the disadvantage of
the sound carrier is 59.75 mc. In a typical receiver,
having another tuned circuit, which complicates the
the local oscillator is set to operate at 82 mc when
switching or tuning circuits, whichever is used. To
obtain a voltage gain in the input transformer, the tuned to this channel. This produces a picture i.f. of
26.75 mc and a sound i.f. of 22.25- mc. It is advan-
secondary inductance should be large with a corre-
spondingly low value of shunting capacitance. The tageous to have the picture i.f. at the higher frequency
minimum value of Co is limited by the input capaci- because of the larger bandwidth required for the video
tance of the tube, making tubes with interelectrode signal. At the higher frequency, this bandwidth is a
capacitances desirable. In order to keep the ratio of smaller percentage of the carrier frequency and re-
inductance to capacitance as large as possible, Co is
usually the input capacitance of the tube and strays
due to wiring, and the tuning is accomplished by R -F AMPL. CONY.
varying the inductance. This may be done by hav-
ing taps on the coil, by switching in different coils,
or by changing the position of an iron core in the
coil.
B+ TO O5C.
Frequency Converters
Fig. 4-3.-Simplified schematic of a converter (mixer)
The output of the r -f amplifier is fed to the stage.
mixer. If no r -f amplifier is used, the antenna, or
transmission line, is connected to the mixer through
a suitable input circuit. This input circuit may be of quires less loading of the tuned circuit. This means a
the same type that is used for the r -f amplifier. In higher value shunting resistor and in turn, a higher
Fig. 4-3 we have a typical mixer circuit. For the gain.
sake of clarity, loading resistors across the tuned In almost all television receivers, separate tubes
circuits have been omitted, and the mixer tube has are used for the mixer and local oscillator, as the
been shown as a triode although it usually is a pen- combination tubes available at present are not satis-
tode. The r -f amplifier is coupled to the mixer factory. These tubes do not have a high enough
conversion gain and do not oscillate readily at the electric between them. Since the dielectric is a de-
high frequencies required for television. termining factor in the amount of capacitance, it
is evident that the collection of moisture on the
plates changes the value of capacitance, which in
High -Frequency Oscillators turn changes the frequency of operation of the os-
cillator. The moisture that collects on the coil of
The oscillator used in a television receiver must be
the oscillator tuned circuit changes the dielectric
able to supply sufficient signal over the entire range
properties of the surrounding medium, which in
of frequencies and has to be stable with respect to
turn changes the distributed capacitance of the coil,
voltage and temperature. To cover the entire tele-
resulting in oscillator frequency drift.
vision band, the oscillator will have to work over "a
The bad effects of humidity are usually control-
range of frequencies from about 70 to 240 mc. The
led by adding some unit that will produce a con-
oscillator tuning accuracy is an important considera-
stant high temperature in the vicinity of the oscil-
tion in television receivers, since the oscillator deter- lator, thereby keeping the circuit dry. Another
mines the i.f. for both the picture and sound channel. method is to prevent moisture by placing the oscil-
At the same time that the picture i.f. is at the correct lator tank circuit in some form of impregnated can.
frequency, the audio i.f. has to be placed in the center Still a third method is to coat the active coil and
of the response band for proper reception of the sound. capacitor elements that may be affected by humidi-
Generally, the audio i -f amplifiers will have a band- ty with some moistureproof material.
width of about 250 kilocycles. The f -m sound signal Frequency changes due to humidity do not. occur
only swings 25 kc each side of the carrier, and this very often compared with those due to heating ef-
bandwidth is not necessary just to pass the sound fects. Changes in temperature surrounding the os-
signal. However, by using a larger bandwidth than cillator circuit will cause the inductive and capaci-
necessary, it is possible to build an oscillator that tive tank circuit components to change in value,
will keep the audio i.f. within the pass band. In a re- thereby causing the oscillator to drift in frequency.
ceiver of this type, the audio i.f. will be able to drift It should be remembered that we are dealing with
100 kc above or below the center frequency and still very high frequencies (the oscillator in television
be passed by the sound i.f. amplifier. This means an receivers operating in the vicinity of 70-240 mc)
allowable drift of the oscillator of 100 kc in 240 and that changes in the shape of an inductance or
mc at the highest oscillator frequency, or about 0.04 capacitance, although minute, will cause a definite
percent. This requires a very stable oscillator and also change in oscillator frequency.
an accurate method of adjustment. Increases in temperature cause the windings of
High -frequency oscillators are more subject to the coil and plates of the capacitor to expand, there-
drift from the original frequency of operation, due by inherently increasing the inductive and capaci-
to such factors as heat, humidity, and changes in tive values of the oscillator tank circuit. The amount
the B -supply voltage on the elements of the oscil- of expansion is determined by the material from
lator tube, than are low -frequency oscillators. The which the coil and capacitor are made, and a con-
oscillator drift caused by heat and humidity is a stant called the temperature coefficient is a réady
slow process, wherein the elements of the oscillator means of determining how much the component will
tube and the circuit components change their in- expand. A low temperature coefficient means that
ductance and capacitance values in accordance with the component will have a small amount of expan-
the slow changes in temperature and humidity sur- sion and, therefore, contribute little to oscillator
rounding the complete oscillator circuit. frequency instability. Low -temperature coefficient
The effect of humidity on oscillator drift is not coils and capacitors are desired in high -frequency
readily apparent in locations where the atmosphere oscillator circuits. Since it is more difficult to ob-
is arid and the humidity does not undergo much tain a variable capacitor with a low -temperature
change ; however, it is definitely apparent in places coefficient than a variable inductor with this feature,
having a humid atmosphere which is constantly permeability tuning is preferable to capacitor tun-
changing. The effect of humidity is such that a ing wherever temperature changes are evident in
certain amount of moisture condenses on the coil the oscillator circuit.
and capacitor of the oscillator tank circuit and Oscillator instability due to temperature changes
causes a change in the dielectric surrounding them. is chiefly controlled by the insertion of a negative
As far as a variable air capacitor is concerned, the temperature coefficient capacitor which is placed
moisture collects on the plates and changes the di- across the oscillator tuned circuit. When the tem-
perature increases, the capacitance of this negative coupling method used in the circuit shown in Fig.
coefficient capacitor decreases; offsetting the in- 4-3. Some of the oscillator voltage is tapped off
crease in the values of the oscillator tank circuit part of the oscillator tank coil and fed directly into
components and thereby maintaining stability of the the cathode circuit of the 6AG5 miter tube through
oscillator against temperature changes. The chief the cathode's resistor-capacitor bias combination.
drift in the oscillator is during the initial warm-up Within this 6AG5 tube the r-f signal and oscillator
period of the receiver. signal are mixed together and form the i.f. which
Another cause of oscillator instability is change is selected by the i -f output tuned transformer cir-
in the B -supply voltage on the oscillator tube. A cuit.
change in plate voltage will cause the transconduc-
tance and plate resistance of the tube to change,
which causes drift in the oscillator frequency. This Belmont Model 21A21
change in supply voltage is usually due to line volt-
The front end of the Belmont model 21A21 tele-
age variations. A regulated power supply can be
vision receiver illustrates several of the features that
used to stabilize the plate voltage, but this may not
we have discussed. The television channels are cov-
be necessary if the oscillator is carefully designed so
ered in two bands, the low band covering channels
that the frequency changes very little with moderate
1 to 6 and the high band covering channels 7 to 10.
changes in supply voltage. Different types of oscil-
lators have different degrees of stability with re- Separate tuned circuits for the high and low bands
spect to changes in plate voltage. are used in the r -f amplifier input, the mixer input,
and in the oscillator tank. These tuned circuits are
switched by the switch Si. Within the band, the re-
ceiver is tuned to the different television channelsby
varying the inductance of the tuned circuits. This is
done by changing the position of a powdered iron core
in each coil. When the core, or slug, is entirely with-
in the coil, the inductance will be higher than when
the core is partly removed from the inside of the
coil. The cores are all connected together mechani-
cally and are adjusted simultaneously by a single
tuning control from the front panel. This control
is the tuning control that tunes the receiver to the
desired station.
Let us consider this circuit in the high -band posi-
tion, as shown in Fig. 4-5. The transmission line,
which should be a 300-ohm balanced line, is con-
Fig. 4-4.-A typical oscillator -mixer circuit as used in nected to the transformer primary, T2 -B. The sec-
television receivers.
ondary winding T2 -A has an adjustable core, as
mentioned above, and tunes with the capacitor C2
A typical oscillator and mixer circuit that may be to the desired channel. The trimmer C3 enables this
used in television receivers is shown in Fig. 4-4. circuit to be aligned for proper tracking. The band-
The 6AG5 mixer tube and 6C4 oscillator are minia- width of this tuned circuit is sufficient without a
ture type tubes especially designed for high -fre- loading resistor, because the coil T2 -B covers the
quency work. Their input, output, and interelec- high band of the frequency range, and the 6 -mc
trode capacitances are smaller than those of regu- bandwidth required is not too great a percentage of
lar size tubes, making them readily adaptable to the carrier frequency. However, a loading resistor
high -frequency work. The separate oscillator is a is necessary for the low -frequency band. This is
conventional Hartley circuit in which the coil is resistor R2 which is across the low-band coil Tl -A.
tapped and connected to the cathode. The 6AG5 The low end of the coil T2 -A is connected to the
mixer tube is a pentode with the control grid used avc circuit. The action of this is similar to the avc
for the r -f voltage input and the cathode circuit for circuits used in sound receivers. The gain of the r -f
the injection of the oscillator voltage. The method amplifier is varied by changing the bias on the grid.
used to inject the oscillator voltage into the mixer The amount of avc voltage depends on the strength
tube in Fig. 4-4 is somewhat different from the of the signal that is received. Besides the r -f ampli-
fier, the first picture i -f amplifier, and the first sound the center of the pass -band, but nearer to the edge of
i -f amplifier are controlled by the ave. the response characteristic. This is because of the
In the plate circuit of the 6AK5 r -f amplifier, vestigial sideband type of transmission that is used
the coil T3 is tuned to resonate with the output ca- for the picture channel. This is discussed in chapter 2.
pacitance of the 6AK5 at the same frequency as the The output of the converter is connected to the
input circuit. The trimmer C6 has a similar func- first video i -f amplifier. The audio i.f. is taken off
tion as the trimmer C3 mentioned above. The signal the output of the first video i -f stage. The oscilla-
from this tuned circuit is coupled to the grid of the tor circuit is a modified Colpitts type and uses a
6AK5 mixer tube by the capacitor C11. At the 6C4 tube. At the high frequencies used in a tele-
same time, the signal from the local oscillator is fed vision receiver, the interelectrode capacities are suf-
to the grid of the converter by the coupling capa- ficient to be used as the feedback circuit in the
citor C50. The local oscillator operates at a fre- oscillator. In this circuit the cathode is effectively
quency 26.75 me above the picture carrier fre- connected to the junction of two capacitances that
quency, therefore a picture i.f. of 26.75 me will be consist of the cathode -to-grid and cathode -to -plate
produced in the mixer. Since the sound carrier fre- interelectrode capacitance. On the high -band posi-
quency is 4.5 me above the picture carrier fre- tion, the frequency of the oscillator is determined
quency, a sound i.f. of 22.25 me will also be produced by the tuned circuit made up of the coil T9 and the
in the mixer. The transformer in the output of the capacitors C45 and C46. The inductance of the coil
mixer, T5, is double -tuned to a center frequency of T9 is varied to provide tuning to the television
25.25, is overcoupled, and has the secondary loaded channel desired and is coupled to the other variable
by the resistor R14 in order to provide a response inductance coils. The trimmer C46 serves as an
wide enough to pass the video and sound i.f.'s. This alignment adjustment to secure proper tracking of
is shown at (A) in Fig. 4-5. The sound i -f frequency the oscillator. The choke coil L3 and the resistor
of 22.25 me will come just within the pass -band of the R34 keep the oscillation out of the B supply and
coil T5. The picture i.f. of 26.75 me does not fall at also serve to keep the grid from being grounded
ANT
300 OA.
(BALANCEP) /
COY.. 41,10,1.0 POINT AS
CONVERTER
OUTPUT
3 23.75 MC' L26.75 MC

6C 4
osc. RESPONSE CURVE OF COIL T5
Courtesy of Belmont Radio Corp.

Fig. 4-5.-The front end of the Belmont model 21A21 television receiver. The switches are shown in the high -band
position.
through the B supply. The cathode circuit also Neutralizing an amplifier over the wide frequency
contains an isolating resistor R36 which helps main- range encountered in television would be difficult
tain oscillation throughout the frequency range re- and the grounded -grid amplifier eliminates this ne-
quired by placing the cathode above ground. cessity. Another important advantage is that the
output capacitance is reduced. The output load is
across the plate to grid circuit rather than the plate
General Electric Model 802 to cathode circuit. The interelectrode capacitance
The local oscillator in the GE model 802 television between the plate and grid is much less than the ca-
receiver is used for the 13 television channels, the f-m pacitance between the plate and cathode. This re-
band, and the broadcast band. The r -f amplifier and duced output capacitance enables a larger induc-
mixer used for the television band is also used for the tance to be used to resonate to the desired f re-
f -m band. The schematic in Fig. 4-6 shows the front quency, which in turn simplifies the construction of
end of this receiver when the receiver is operating the tuned circuits, especially at the higher end of
on the television band. For the sake of clarity, all the television band.
switching circuits have been eliminated. Television Referring to the schematic in Fig. 4-6, the antenna
channels are selected by a switch which changes the or transmission line is connected across the choke,
coils LK in the cathode circuit of the r -f amplifier, LK. This coil is changed for the different television
Lp and Ls in the coupling circuit between the r -f channels in order to present a 300 -ohm load to the
amplifier and the mixer, and the coil Lo in the oscil- line for a proper match. In the plate circuit of the
lator circuit. A 6AU6 tube is used as the r -f ampli- r -f amplifier, the autotransformer L,.Ls and the
fier in a grounded -grid amplifier circuit. In this type shunting capacitances form a double -tuned coupling
of circuit, the grid of the amplifier is connected to circuit. The primary Lp is tuned with the shunting
ground, and the input is connected to the cathode capacitance formed by the output capacitance of the
circuit instead of to the grid as in the conventional 6AU6 plus the stray capacitance due to wiring. The
amplifier. The output load is still in the plate cir- secondary Ls is tuned with the shunting capacitance
cuit. There are several advantages to the use of a consisting of the input capacitance of the 7F8 mixer
grounded -grid amplifier at the high frequencies plus the strays. The cir uit is tuned by adjusting
utilized for the reception of television signals. The the inductance of Lp and L$ and by varying the
grid, being grounded, acts as a shield between the coupling between them. The windings are over -
cathode and the plate. This prevents coupling be- coupled to provide the necessary bandwidth. For
tween the input and output circuits due to the inter - proper alignment, the inductance of either Lp or Ls
electrode capacitances that otherwise might cause can be varied by moving a few turns of the coil.
oscillation unless the amplifier were neutralized. The coupling between them can be varied by mov-
6AU6 7F8 TO VIDEO

R -F AMPL. CONV.
I -F AMPL.
C5 C9
Lp
2 7F8
05C.
C103
R92 C105 R93

ANT.
INPUT
C101
cp TZ _ L20
1297 C111
Fig. 4-6.-The r -f amplifier, converter, and oscillator of the GE model 802.

Courtesy General Electric Co.
ing the coils closer together or farther apart on the lator is tuned to the correct frequency by resonance
assembly. On channels 1 and 2, the coupling trans- between the coil Lo and the capacitor C101. The
former is triple -tuned to provide greater selectivity. inductance of the coil Lo can be changed by spread-
This is shown in Fig. 4-7. The three windings of LP, ing or compressing the turns of the coil. A different
Ls, and LT are all inductively coupled to each other. coil is switched into the circuit for each television
The third winding LT is tuned with the shunting channel. Since the plate of the oscillator is ground-
capacitance CT to the same frequency as the primary ed by capacitor C103, the cathode has to be remov-
and secondary windings. The reason greater selec- ed from ground. The choke L20 places the cathode
tivity is necessary on channels 1 and 2 is that above ground for r.f. but provides a path to ground
for d.c.
Fig. 4-7. - The

triple -tuned coupling RCA 621TS
circuit used for
channels 1 and 2 The r -f amplifier, mixer, and oscillator circuits of
in the GE model 802.
the RCA model 621TS are assembled on a separate
subchassis of the receiver and this unit is also used in
there is possibility of image interference from the the RCA model 630TS. Three 6J6 tubes are used for
f -m band which is from 88 to 108 megacycles. the r -f amplifier, mixer and oscillator. This tube is a
When the oscillator is set for channel 1, for ex- twin -triode miniature tube that has been designed for
ample, the oscillator frequency will be 26.4 mc (the operation at high frequencies. These tubes are con-
i.f. for this receiver) above the picture carrier of nected in a push-pull arrangement in all input and
45.25 mc or 71.65 mc. If a station in the f -m band at output circuits except for the output circuit of the
98.05 mc gets through the r -f amplifier, it will beat mixer, where the plates are connected in parallel to
with the oscillator and also form an i.f. of 26.4 mc. provide a single -ended output. Up to this point, all the
This will go through the i -f amplifiers and cause plate and grid circuits are operated push-pull. This
interference with the picture. This type of inter- arrangement is necessary because of the resonant line
ference can be eliminated by improving the selec- tank circuits employed. The series of inductances in
tivity of the tuned circuits before the mixer, and in the plate circuit of the r -f amplifier and oscillator, and
this receiver it is done by using triple -tuning instead in the grid circuit of the mixer make up an artificial
of double -tuning on the television channels 1 and transmission line. These small elements of inductance,
2. For the other channels, the coupling circuit is resistance, and shunting capacitance take the place
as shown in Fig. 4-6, except that loading resistors of the distributed inductance, resistance, and capa-
are placed across the secondary L3 for channels 3 citance of an actual transmission line. They are
to 6 in order to provide the proper bandwidth. properly proportioned so that the tube sees the
From the coupling circuit the signal is fed to the same impedance as though it were working into an
grid of the mixer tube, which is one half of a 7F8, actual transmission line. Of course, an actual trans-
a dual -triode tube. The other half of the 7F8 is mission line of the proper length could be used in the
used as the oscillator. The output of the oscillator is same circuit, but the length would make its con-
also coupled to the grid of the mixer tube because struction impractical. At the lowest picture carrier
the oscillator coil Lo is wound on the same form as frequency, 45.25 mc, a quarter wave -length would
the coils Lp and Ls and is, therefore, inductively be about 4.5 feet long. An artificial line can be
coupled to the other windings. The output of the made much shorter than this. In the RCA model
mixer is the difference between the two signals, 621TS, the resonant line is small enough so that it
the i.f. The oscillator in this receiver is a modified can be constructed around the channel selector
Colpitts. The feedback circuit consists of capacitor switch.
C105 in shunt with the interelectrode capacitance An artificial line made up of series inductances
between the grid and cathode of the 7F8 and the and shunt capacitances is shown in Fig. 4-8 (A).
interelectrode capacitance between the plate and The capacitances shown do not have to be actual
cathode. Since the plate of the oscillator is at capacitors. They can be the capacitances between
ground for r -f frequencies because of the by-pass pairs of inductances. At the right end terminals of
capacitor C103, the plate -to -cathode capacitance ap- this line a certain impedance Z will exist. If the
pears between the cathode and ground and is indi- artificial line is properly designed, this impedance
cated by the capacitance Cp in Fig. 4-6. The oscil- will be the same as though the line were an actual
transmission line as shown at (B). Therefore, as the line was made longer. Thus we can simulate the
far as the circuit connected to the terminals is con- effect of changing L and C in the tuned circuit at
cerned, a line exists. If a shorting bar were placed (C) by merely changing the position of the short-
across the line shown in Fig. 4-8 (B) a quarter ing bar on the line shown at (B). In the artificial
wave -length from the terminals, we would have a line used in the front end of this receiver, the same
resonant line which would have the characteristics effect is achieved by switching the shorting bar to
of a tuned circuit as shown at (C). The resonant different points along the line.
frequency of this tuned circuit would be the fre- Since a resonant line looks like a tuned circuit
quency at which the line is a quarter wave -length at the terminals, the question may be raised as to
long. If the shorting bar can be moved along the why a line is used at all. Why not instead use a
line, we can make the line resonant to different fre- tuned circuit consisting of a coil and capacitor?
quencies. The resonant frequency would become One advantage of the line over an ordinary tuned
higher as the line was made shorter and lower as circuit is that the line has a higher impedance at
resonance, making possible a higher gain in the
amplifier. When used as an oscillator tank, the high-
er Q of the resonant line will result in increased
stability of the oscillator. Since the effect of chang-
r -r
ing the position of the shorting bar on the line is
the same as changing L and C in an ordinary tuned
(A) ARTIFICIAL LINE
circuit, it is possible to keep the L/C ratio fairly
-
SHORTING
BAR
WAVELENGTH -+{ o
constant and thus have a constant bandwidth over
the range of frequencies in the band. The switch-
ing arrangement when using a resonant line is
simpler, since components do not have to be re-
(6) ACTUAL LINE (C) EQUIVALENT CIRCUIT moved from the circuit to be replaced by other
Fig. 4-8.-The artificial resonant lines used as tank cir-

cuits in the RCA model 621TS are shown in (A). The
physical line which (A) replaces, is shown in (B) and the
equivalent circuit in (C). a. TO SOUND I.F.
.-i
VI ANTE
Ri R -F
6J6
AMPUFlER C2
4700 L1 L5 LS 1.7 L5 LII LIS LIS L17 09 L11 LIS L25
-1.5 V 270
.4. +90V. C3
6 7 3 5 10 I I 12 á RIS
52-B 6
ISO
52-A Re
4 5 6 7 8 5 p 11 15
190
Y
11
1
LB LI/. 20 L24 L2c +50v. CI
CI L2 L4 LB 110 L12 1.14 LM L22 C4
1.5 270 179
1500 C 10 C12 LINK C19
2.2 0.59 V2
4.7 COUPLING Cn
54 L27 L29 L58 LS5 L37 1.39 L43 L45 L47 L49
6J6 1500
C1B
270
1000
131 1.41 L51 CONVeRTLR
S 11 7 8 0 1/ 11
M
3
+115V. 7^2 -I
CDNV.TRAN
I>. TO PIX I.F.
59-B LCIG
CNANNEL 55 22.B 6
62
SWITCH
-
IOK MC.
7
! q ! ID
n 2
c 21.2s14c.1r
i
L28 L50 U2 L54 156 156 140 142 144 L46 L48 LS0 L62 vs M4
14
LIN(
6J6 15001 1500
-
CWPLINB
3 R -F 05C. jti
RED
L53 155 L57 L59 L61 163 165 L67 Le L71 L73 175 77
+97V
+1YsV
-a 5IIELD C6
2 3 4 6 7 e ! 0 II I2 ß 4.7
51
4100
54-B CIS
qa5-2.75 T
1 6
R12
A 1000
I
FI1 TVNIN6 e I
GREEN TOCONTRAST
54-A 1 CONTROL
I
CB 4 a 9 A
1500
S G 7 10 I I 12
7 C5
2 47 1500
6-) + v.
= C19 L54 L56 1.413 160 162 1.64 L66 168 170 L72 L74 L76 L78
R8 RIO R9
10
47 100 K 100K Fig. 4-9.-The front end of the
RCA model 621TS. Note the arti-
ficial lines used for tuning.
Cowrtesy RCA
components. All the switch has to do is to place core. The inductance of the coil is decreased when
a short at different places along the line to tune to the core is moved farther inside the coil, and in-
the different television channels. creased when the core is moved out of the coil. The
Referring to the schematic of the front end feedback circuit for the oscillator is from the plate
shown in Fig. 4-9, the transmission line from the of one triode to the grid of the other triode by
antenna, which should be a 300 -ohm line, is con- means of the capacitors C6 and C7. The small var-
nected to the grids of the 6J6 r -f amplifier. The iable capacitor C15 is the fine-tuning adjustment.
resistors R13 and R3 provide a proper match for This capacitor can shift the frequency above or
the 300-ohm transmission line. The coil T1 acts as below the Center frequency by a small amount. This
a bypass for frequencies that are below the televi- range varies from about 300 kc at the lowest tele-
sion band and effectively shorts them to ground to vision channel to about 750 kc at the highest chan-
prevent them from causing interference in the re- nel. The control for C15 is on the front panel and
ceiver. Since a triode amplifier N used, neutraliza- is mounted on a concentric shaft with the station
tion is necessary to cancel the effect of feedback due selector switch.
to the interelectrode capacitance between grid and The signal from the local oscillator and the in-
plate which may causé oscillation or instability. This coming signal are combined in the mixer tube and
is accomplished by the capacitor C3 and C4. In a beat frequency is produced which is equal to the
the resonant line in the pláte circuit, L25 and L26 difference of these two frequencies. This is the i.f.
are adjustable to provide the correct length of line and appears in the output circuit of the mixer. The
for the 13th channel. The following inductances mixer transformer T2 is not only the plate load of
L13 to L23 on the upper side of the line and L14 the mixer but also serves to separate the picture i.f.
to L26 on the lower side are fixed. Physically they from the sound i.f. The primary of this trans-
consist of a strap between the switch contacts. The former is tuned with the output capacitance of the
shorting bar is moved along this strap to tune to 6J6 mixer to 22.8 mc. This frequency is used rather
channels 12 to 7. The next coils in the line are L11 than the center frequency of the picture i -f re-
and L12 which are adjustable to take care of the sponse because the i -f amplifiers in this receiver are
large change in frequency between channel 7 and stagger -tuned. The secondary winding on the mixer
channel 6. The following inductances, L1 to L9 transformer together with capacitor C16 is tuned to
and L2 to L10, are fixed and are switched into the the sound i.f. of 21.25 mc. This circuit acts as a
line for tuning to channels 5 to 1. trap and removes the sound i.f. from the primary
The resonant line in the mixer grid circuit is winding by absorbing that frequency. The sound
similar to the line in the r -f amplifier plate circuit i -f signal that appears on the secondary is coupled
and operates in the same manner. This line is to the sound i -f amplifiers, and the picture i.f. on
coupled to the r -f line by the coupling capacitors the primary, from which the sound i.f. has been
C10, C12, C13 and also by a link. The series com- removed, is coupled to the picture i -f amplifiers
bination of L80 and C14 is tuned to the i.f. to short through the coupling capacitor C18.
out any interference from external signals at the i.f.
and to prevent feedback from the output of the mixer
at the i.f. The grid circuit of the mixer is also coupled Du Mont Model RA -103
to the oscillator by means of a link between the
mixer line and the oscillator line. The oscillator The r -f amplifier, oscillator, and mixer circuits
line is constructed in the same manner as the r -f for the Du Mont model RA -103 are shown in Fig.
and mixer line except that provision is made for 4-10. The r -f amplifier uses a 6J6 twin -triode tube
tuning the line for each television channel. This is with the two triode sections connected in parallel. The
done by varying the inductances L54 to L78 in the input circuit is designed to match a 75 -ohm coaxial
lower leg of the mixer line. L63, L64, L77, and transmission line. The r -f amplifier employs a
L78 are varied by changing the position of an iron grounded -grid type of circuit arrangement, therefore
core within the coil, as are the variable inductances the input signal from the antenna or transmission line
in the r -f and mixer line. The other variable induc- is fed to the cathode circuit of the 6J6. The advantages
tances in the oscillator line are adjusted by chang- of a grounded -grid amplifier have been discussed pre-
ing the position of a brass screw that is placed viously in this chapter in conjunction with the GE
close to the strap sections, L66 to L76, or by mov- model 802. Referring to Fig. 4-10, the coil L106 across
ing a brass core within the coils L54 to L62. The the input circuit provides a bypass to ground for any
effect of the brass core is the opposite of an iron interference below the television band. The grids of
the r -f amplifier are returned to ground through the the grid of the 6AK5 mixer tube by means of the
parallel combination of C116 and Rlll rather than coupling capacitor C112. The two signals beat to-
directly in order to suppress parasitic oscillations. gether in the mixer and produce the i.f. in the output
The 6J6 r -f amplifier is coupled to the grid of of the 6AK5. This frequency is equal to the difference
the 6AK5 mixer by means of a double -tuned coup- in frequency of the local oscillator and the incoming
ling network. The coils L101, L102A and L104- signal to which the receiver is tuned. In this receiver
L102B are tuned with the shunting capacitances the picture i.f. is 26.4 mc and the sound i.f. is 21.9 mc.
due to the output capacitance of the 6J6, the input Both these frequencies are fed to the input of the
capacitance of the 6AK5 and the coupling capaci- first i -f amplifier, to be removed farther on in the
tors C105, C106, and C107. The inductance of coils circuit.
L 102A, L 102B and oscillator coil L 102C is con- The oscillator circuit is a modified Colpitts and
tinuously variable by means of a sliding contact on uses one half of a twin -triode 6J6. The feedback
each coil. These are part of a unit known as the consists of the interelectrode capacitance between
Inductuner. The inductance of these coils can be plate and cathode and between grid and cathode of
varied sufficiently to cover the entire television band the 6J6. The frequency of the oscillator is deter-
from 44 to 216 megacycles. The inductance varia- mined by the resonant frequency of the parallel
tion of each coil is approximately from 0.02 to LC tank in the grid -plate circuit. The inductance of
1.0 microhenry, which is spread over 10 turns of L102C is continuously variable and is mechanically
the Inductuner. This system of tuning has sev- coupled to the other coils in the Inductuner. The
eral advantages. The entire television band is cov- trimmer capacitor C111 and the inductance L103
ered by one control without the use of separate provide proper tracking of the oscillator. Bias on
tuned circuits for each channel. The bandwidth is the oscillator tube is provided by the combination
fairly constant over the entire range of frequencies. of C114 and R109. The plate and grid of the second
The f -m band of 88-108 mc is covered in the con- section of the 6J6 are grounded since this section
tinuous tuning range of the Inductuner which sim- of the tube is not used. The capacitor C115 across
plifies the construction of combination television and the heater of the oscillator places the heater at
f -m receivers. ground for r.f. Notice that the r -f amplifier and
The resistors R110 and R104 load down the converter heaters are also bypassed with a similar
tuned circuits in order to provide sufficient band- capacitor. This prevents coupling between these
width, which should be about 6 mc in order to pass three stages through the common heater supply and
both the picture and sound signal. The capacitors also prevents interfering signals from the rest of
C105 and C106 can be adjusted for proper alignment. receiver. The plate voltage supplies are also sepa-
The output of the local oscillator is also coupled to rately bypassed at each tube for the same reason.
CI04 0105 C106 C108

ISy01 4-30001 4-30yy1 ISyyl
NPO IOx N500 N500 N20 IOx MO V.DG. "A" TO LT TRANS.
MIXER ON RECEIVER CHASSIS
RI10
12K
C1
>S20yyOT 1 I1/2
8104 V102
GARS
1101 ARRROR L104 2R
IOW
1
R102
IOR C110
2W 4TOyy1
8105
L10 2A LIO2B IMEO
1/2W
RIOT
3308
1/2W
C109
470yy1
6.1VA.G. "8" TO HTR. TRANS.
ON RECEIVER CHASSIS
MALLORY
L T
1
3 GANG 9CIICTLBER J
OSG. 25VD.G. "G" TO 8
ON RECEIVER CHASSIS
CIIS
47080
LIO
6114
SRyI
N-030
Fig. 4-10.-The r -f amplifier, oscillator, and mixer circuits of the Du Mont model RA -103.
Courtesy of Allen B. Du Mont Laboratories, Inc.
Westinghouse Model H-181 plates from being grounded for r.f. through the B+
supply. The two cathodes are tied together and
The local oscillator circuit used in the Westing- cathode bias is obtained by the voltage divider con-
house model H-181 is shown in Fig. 4-11. The fre- sisting of resistors R1 and R2. This bias holds the re-
quency stability of this oscillator is improved by the actance tube at the correct operating point. The oscil-
use of an automatic frequency control (afc) circuit. lator uses grid-leak bias supplied by the combination
In Fig. 4-11, the first half of the 6J6 tube is used as of C3 and R3 and is not affected by the positive volt-
the local oscillator, and the second half of the 6J6 is age on the cathode because the grid of the oscillator
used as a reactance tube in the afc circuit. The oscil- is returned to the cathode rather than ground by the
lator circuit is a modified Colpitts that employs the resistor R3. The grid of the reactance tube is coupled
interelectrode capacitances of the tube as the feedback to the oscillator coil through the network consisting
circuit. The cathode of the oscillator tube V1 is of R4, C5, Ce, and the input capacitance Ci of tube V,.
grounded for r.f. by capacitor C4, thereby effectively The cold grid -to -cathode capacitance of the 6J6 is
placing the cathode at the junction of the plate -to - only about 2 µµf, but due to the gain of the tube, the
cathode and grid -to -cathode capacitances of the tube. input capacitance of the reactance tube will appear to
The frequency of oscillation is determined by the reso- be much larger than this (Miller effect). The capaci-
nant frequency of the tank circuit consisting of the tor C5 will have very little reactance at the operating
oscillator coil L1i capacitors C2 and C3, the interelec- frequencies of the oscillator (approximately from 70
trode capacitances of V1, and the shunting capacitance to 240 megacycles) and resistor R4 may be consid-
C. of the reactance tube V7. The oscillator is set for ered to be in parallel with capacitor Ce. This capacitor
the correct frequency for each television channel by enables the reactance tube to operate properly over
switching in different values of inductance for L1. The this large range of frequencies.
frequency of the oscillator can also be changed by The current flowing through Ci will be determined
changing the value of any capacitance in the tank cir- mainly by resistor R4. This current will therefore be
cuit. If this could be done automatically as the oscil- in phase with the voltage producing it, that voltage
lator drifts, due to changes in voltage or temperature, on the grid of V1. The voltage across C4 will lag the
and if the change in capacitance were sufficient and in current by 90° and, therefore, lag the grid voltage of
the right direction, the frequency of the oscillator V1 by 90°. The plate current of V2, which is in phase
could be kept constant. This is the function of the re- with the grid voltage on V2, will also lag the voltage
actance tube V2. The shunting capacitance C. supplied on the grid of V1 by 90°. Since V1 is an oscillator, the
by this tube varies in such a manner as to keep the voltage on the plate will be 180° out of phase with the
frequency of the oscillator constant. voltage on the grid, and since the two plates are tied
To analyze the operation of the reactance tube, let together, the plate voltage of V$ will also be 180° out
us first examine the circuit shown in Fig. 4-11. The of phase with the grid voltage of V1. We have already
plate of the reactance tube is connected to the plate shown that the plate current of Vg lags the grid volt-
of the oscillator tube. The choke L2 keeps the two age of V1 by 90°. Therefore, the plate current of V7
a. 13+
Rl 6J6
1470µµf
Ci 22K 05C.? OSC. Fig. 4-11.-The local oscillator and auto-
CONTROL matic frequency control circuit of the West-
inghouse model H-181.
Vi V1
Courtesy of Westinghouse Electric Corp.
äK TO
DISCRIMI
C3 Rg THROUGH
22pi.if 22K FILTER
Cl
470pµf
R2 _C6
220 Cq
220µµf
220µµF
will lead the plate voltage of Vs by 90°. This is the produce a sound i.f. of 22.1 mc. This will correspond
same relationship that would exist with a capacitance to point B on the response characteristic in Fig. 4-12.
in the circuit, and because of this, tube Vs will appear The output of the discriminator will be zero, and no
as a capacitive reactance, or an equivalent capacitance correction will be applied to the oscillator. The second
C8, across the plate circuit of the oscillator V1. The case is that where the oscillator has drifted to a higher
value of this capacitance can be varied in the follow- frequency due to some change in voltage or tempera-
ing manner. If the grid of V2 is made more positive, ture. This results in an audio i.f. which is too high and
the plate current of this tube will be increased. This corresponds to point C. The output of the discrimi-
would make the tube appear as a smaller reactance, nator is then positive, and this positive voltage applied
and since the reactance is capacitive, the equivalent to the grid of the reactance tube makes the capacitance
capacitance C8 appears larger. Conversely, if the grid Cs larger, as explained previously. This larger capaci-
were made more negative, the capacitance Cs would tance makes tube Vs oscillate at a lower frequency,
be decreased. bringing the audio i.f. back to the proper frequency
at point B. In the third case, the oscillator drifts to a
frequency below 71.85 mc. This would result in a
sound i.f. which is below 22.1 mc, as shown by point
C E
A on the discriminator characteristic. The negative
voltage produced by the discriminator and applied to
the grid of the reactance tube would decrease the
capacitance Cs. This would make the oscillator work
at a higher frequency and bring the sound i.f. up to
point B. From the foregoing, it can be seen that the
afc system will tend to keep the oscillator at the cor-
rect frequency for the particular channel being re-
ceived at all times, despite any -tendency of the oscil-
lator to drift due to changes in voltages or tempera-
21.35 21.85 22.1 2235 22.85 ture. However, the range of control of this system is
FREQUENCY IN MC limited to the frequency range between the peaks of
the discriminator characteristic. This is about 500 kc
Fig. 4-12.-The discriminator characteristic of the West- in this receiver. If, for any reason, the oscillator
inghouse model H-181.
should drift past the peak on either side, for example,
point D or E, the afc system would lose control and
The grid of reactance tube V2 is connected to the not restore the oscillator to the proper operating fre-
output of the discriminator through a filter which quency. This happens because the control voltage at
removes the audio component but allows the d -c volt- these points decreases with increasing drift in fre-
age to be applied to the grid. The discriminator char- quency, instead of increasing as it does at points A
acteristic of this receiver is shown in Fig. 4-12. At and C.
22.1 mc, which is the sound i.f., the output of the dis- To get the maximum advantage from the afc sys-
criminator is zero. For frequencies below this value, tem, it is important that the initial operating condition
the output will be negative, and for frequencies above, of the oscillator be at point B. This adjustment is
the output will be positive. made in the alignment of the receiver by disabling the
Let us consider three possible conditions that may afc system and adjusting the frequency of the oscil-
arise in the operation of this system. Assuming that lator until the output of the discriminator is zero. This
the receiver is tuned to channel one, the local oscil adjustment should not be made until the receiver has
lator will be operating at a frequency of 71.85 mc. reached normal operating temperature, which usually
This will beat with the audio carrier of 49.75 mc to' takes at least 20 minutes.
CHAPTER 5
THE F -M SOUND CHANNEL

BY SEYMOUR D. USLAN
In the first type of commercial television receivers Frequency Deviation

the sound signals were amplitude modulated and
the sound channel contained circuits similar to the In regular f -m broadcasting the maximum band-
a -m superheterodyne receiver. Today the sound width involved, as defined by the FCC, is equal to
signals used with television are frequency modu- 100 kc on either side of the center frequency. This
lated. There are many reasons why f -m sound is due to the 75-kc peak frequency deviation for
broadcasting is superior to amplitude modulation 100 -percent modulation plus the 25-kc guard band.
in this field. Reduction in interference, better high In television the sound transmitter according to
fidelity sound, possibly better signal-to-noise ratio FCC standards should be designed to operate satis-
are just a few of the advantages of f.m. factorily with a peak frequency deviation (fre-
Due to the nature of f -m and video signals, they quency swing) of 25 kc for 100 -percent modula-
can both be received by the same antenna. The cir- tion. In other words we have a 150-kc peak -to -
cuits of the f -m sound section of a television re- peak frequency deviation in regular f -m broadcast-
ceiver are very similar to those of the regular f -m ing and a 50-kc peak -to-peak frequency deviation
receivers that now appear on the market. The cir- in television sound transmission, both for 100-per-
cuits that we are essentially interested in are the cent modulation. Consequently we find that the f -m
i -f .section, the detector section, and the audio sec-
sound signal in television has 1/3 of the frequency
tion. These three topics of the f -m sound channel deviation of a regular f -m broadcasting transmitter
will now be considered in detail. for 100 -percent modulation.
This reduced deviation and, hence, bandwidth
does not put any limitation on the range of audio
THE I -F SYSTEM frequencies that can be passed. As in regular f -m
broadcasting, the f -m sound transmitter in television
The output signal from the converter or mixer has to be designed to pass audio frequencies from
from the front eigd of the set is a combined video 50 to 15,000. cycles. Therefore high fidelity in the
and sound i -f signal. In some television receivers television sound section is still possible even though
the sound i.f. is taken from the output of the mixer the peak frequency deviation of output signal is at
or converter tube in the front end of the set. In a maximum of 25 kc. The primary difference be-
other receivers the combined i -f signal output from tween a wide bandwidth and a smaller bandwidth
the converter or mixer is impressed across an ampli- is that in the latter the signal-to-noise ratio will be
fier. This was already discussed in connection with somewhat lower, considering the same amount of
the front end of the set. This i -f amplifier is usually amplification in the receiving systems. This can be
called the first video or pix i -f amplifier. In this remedied, if necessary, by increasing the strength
case the sound i.f. is taken from the output of this of the signal in the i -f stages of the sound channel
tube through a sound i -f tuned circuit. of the television receiver.
In either type of receiver the sòund i -f signal The FCC recommends that the television sound
is impressed across the grid of the first sound i -f transmitter be designed to operate satisfactorily with
amplifier. From this point on, the analysis of the a peak frequency deviation of at least 40 kc. This
f -m sound section of the television receiver is very allows for slight amounts of overmodulation in the
similar to a regular f -m receiver. The only major transmitter.
difference between them lies in the frequencies in-
volved. The I -F's and Image Frequency Interference
NOTE: For a more detailed explanation of frequency The intermediate frequencies of the sound chan-
modulation, see "FM Transmission and Reception" by Rider
and Uslan. nel of some of the first television receivers employ-
68
THE F -M SOUND CHANNEL 69
ing f -m sound was in the vicinity o.f 8 mc. Today i.f. will be detected by the video detector and passed
many of these frequencies are much higher-some- along with the video signal to the picture tube, in-
where around 21 and 22 mc. Some of the common terfering with the pattern on the picture tube.
sound i.f.'s used today are 21.25 mc, 21.9 mc, and The fourth harmonic of this same sound i.f. can
22.25 mc. The i.f. used in most of the f-m receivers cause interference on channel 6 whose frequency
on the market is 10.7 mc. This intermediate fre- range is 82 to 88 mc. This is so because the fourth
quency, in accordance with the commercial f -m harmonic of 21.25 mc is 85 mc which falls in the
band of 88 to 108 mc (20 -mc bandwidth), does not center of the 6th television channel. In similar fash-
allow image frequency interference. In other words, ion if the sound i.f. is 22.25 mc, its second harmonic
if twice the i.f. employed is greater than the band- can cause interference on the second television chan-
width of the frequency range in question, image nel and its third harmonic on the fourth channel.
frequency interference cannot occur.
The bandwidth of frequencies involved in the 13
television channels is somewhat different. In this I -F Stages
case we can consider these channels as divided into The primary function of the i -f amplifiers and
two groups-those channels from 1 to 6 in one transformers in the sound section of a television
group and those those from 7 to 13 in the other. receiver is the same as in any ordinary receiver.
The television sound carrier frequency of the first The i -f amplifiers provide a large part of the neces-
channel (44 to 50 mc) is equal to 49.75 mc and that sary gain and selectivity needed for the proper op-
of the sixth channel (82-88 mc) is equal to 87.75 eration of the sound system. Even though the peak -
mc. This means that the frequency range of the to -peak frequency deviation for 100 -percent modula-
sound carrier signals in the first group, between the tion in television sound broadcasting is 50 kc, the
first and sixth channels, is 87.75 less 49.75 mc or 38 bandpass characteristics of many of the i -f trans-
mc. The sound carrier frequency of the seventh formers are greater than the 50 kc. One of the
channel (174-180 mc) is 179.75 mc and that of the reasons advanced for this large bandwidth is so the
13th channel (210-216 mc) is equal to 215.75 mc. transformer will accept the i.f. even when the r -f
Thus the frequency range of sound carrier signals oscillator of the set drifts in frequency also causing
in the second group, between the seventh and 13th the i.f. to drift. In other words, this 50-kc bandpass
channels, is 215.75 less 179.75 mc or 36 mc. is the miminum bandwidth requirement for the i -f
From this analysis we find that the frequency transformer.
range of the sound carrier signals is 38 mc in the
first group .and 36 mc in the second group. Conse-
quently, if twice the sound i.f. is greater than the
largest of these frequency ranges, then image fre-
quency interference will not result. Thus if the
sound i.f. is around 20 mc (as it is today), image
frequency interference occurring in either channel
fr fb
grouping is not possible. The 86 -mc separation be- Q
FREQUENCY
tween the sixth and seventh channel does away with
the possibility of image frequency interference be- Fig. 5-1.-An ideal selectivity curve that is flat-topped
tween channels. would pass all frequencies between the limits fa and fr,
with equal amplitude ; such a curve cannot be realized in
This same analysis also holds true for the video practice with the use of a single i -f transformer network.
i -f system regarding the picture carriers causing
image frequency interference within the video band.

If harmonics of the sound i.f. are strong enough, The nature of a true f -m signal is such that it
they may cause interference in the video channel varies only in frequency but is constant in ampli-
if picked up by the r -f section of the receiver. For tude. The f -m signal input to the i -f stages for
instance, if a television receiver with a sound i.f. proper operating conditions should be constant in
of 21.25 mc is tuned to channel 3 whose frequency amplitude and it is desired that the output also be
range is 60 to 66 mc, then the third harmonic of constant in amplitude. For the output f -m signal
this i.f., which is 63.75 .mc, can be picked up by from the i -f system to be constant in amplitude, the
the r -f section and passed on through the video response of these i -f stages should be as broad and
system. If this happens, the third harmonic of the flat as possible. A broad selectivity curve with a
somewhat level or flat characteristic can be achieved of the response curve determines the shape of the
but the curve is not perfectly flat-topped and will amplitude variations of the output f -m signal.
cause amplitude variations in the f -m output signal In f -m broadcast receivers, the desired selectivity
from the i -f stages. of the i -f stages is much broader than that of the
A response curve is, after all, a curve that shows television sound channel. To obtain the necessary
the selectivity characteristics of a particular reso- selectivity in the i -f section of f -m broadcast re-
nant circuit, but these curves are not restricted to ceivers, various methods are used to broaden the
operate on a -m or f -m signals. A particular reso- selectivity. Some methods used are to employ over -
nant circuit, like a simple parallel LC circuit, will coupled i -f transformers or a series of stagger -tuned
respond to both a -m and f -m signals. This is the i -f transformers. This latter method is employed
case with all i -f transformer networks. The mean in the video i -f sections of many television receivers
resonant frequency and not the type of modulation and will be discussed in the video section later on.
determines the center frequency of the signal that In the sound i. f section of television receivers,
can be received. the usual method is to employ two i -f amplifiers in
The ideal selectivity characteristic for an f -m conjunction with three i -f transformers where the
signal would be a flat -top response curve as illus- transformers are all single peaked to the same res-
trated in Fig. 5-1, where only frequencies between onant frequency. The single peaking is usually
fa and fb would be accepted, and all would have somewhat under critical coupling. By the use of
the same amplitude. However, such a curve cannot two amplifiers and sometimes more, the necessary
be realized in practice with the use of a single i -f gain for the particular set is thus obtained. Even
transformer network. Methods which help approach though the higher i.f. used in television sound
the shape of such a curve entail the use of more systems causes a lower gain per stage, the type and
than one i -f transformer network, and preferably number of tubes used today make it possible for
of three. receivers to withstand this loss and still produce
Manufacturers are cognizant of the fact that such the necessary gain.
UNITY RESONANT
FREQUENCY Detection of the F -M Signal
rö
<-
<
The f -m sound channel of a television receiver is
similar to a regular f -m or a -m broadcast receiver
FREQUENCY in that the i -f stages are followed by a system of
detection which is followed by an audio system.
Fig. 5-2.-Typical response characteristic of a sound i -f
stage in a television receiver. The primary difference between a -m and f -m re-
ceivers is in the methods of detection. The detectors
for a -m and f -m differ considerably in circuit con-
a curve is only ideal and is extremely difficult to struction and performance but their purposes are
obtain. They are faced with the necessity of getting essentially the same-to detect or demodulate their
a final output i -f signal that is high enough in gain input modulated signal to obtain the audio modula-
and broad enough to pass at least 50 kc at the half - tion component.
power points (also called the 3 -db points) of the The f -m detector system must respond only to
i -f response without any discrimination. They are frequency variations and not to amplitude varia-
also concerned with the fact that a constant ampli- tions. This is essential because the signal resulting
tude signal is preferable, but since limiters and from detection should contain only those audio
special detection circuits that effectively do away characteristics of the modulating signal present at
with the amplitude variations are available, such the sound transmitter. Television receivers today
design is of a secondary nature. A picture of a employ two main types of f -m detectors in the sound
typical response characteristic of i -f stages as used channel-the Foster -Seeley, or phase discrimina-
in the sound section of television receivers is shown tor, and the ratio detector. The latter type of circuit
in Fig. 5-2. From the shape of this curve we can is newer and only one tribe is employed to accom-
readily see that the f -m i -f signal, even though plish the process of f -m detection and amplitude
constant in amplitude at the input to the stage, limitation. The former type circuit employs one
will at the output be an f -m i -f signal that is vary- tube for amplitude limitation and another one for
ing in amplitude as well as in frequency. The shape detection. A separate tube is used because the phase
discriminator will respond to amplitude variations Ideally the function of the limiter is shown in
as well as frequency variations. These amplitude Fig. 5-3 with respect to an f -m signal. The input
variations, whether small or large, are caused pri- to the limiter is an f -m signal that is varying in
marily by the response characteristics of the i -f amplitude as well as in frequency. These amplitude
transformers and also by superimposed noise. variations are undesired and, by the action of the
These two types of detector systems are so im- limiter, are "clipped off" and the output is an f -m
portant to the proper operation of the television signal of constant amplitude.
sound channel that they will be considered in detail For an amplifier to act as a limiter, the potentials
here. We will analyze their circuits from a general on the tube are so chosen that the tube will over-
viewpoint and then show possible modifications o.f load easily with a small amount of signal input. A
such circuits.
The first type of f -m detector system that was
Fig. 5-4.
limiter plate
- The
cur-
rent-grid voltage
used commercially in f -m receivers was the phase characteristic differs
discriminator where a separate amplitude limiter from the usual char-
acteristic in that it
was employed. In this system the limiter tube, which flattens out in the
precedes the discriminator detector, levels off prac- positive region (be-
yond point 2). Am-
tically all of the a -m variations by clipping off the plification will occur
upper and lower portion of the f -m signal so that between 1 and 2;
it will be constant in amplitude when it reaches the
beyond 2, limiting
action will occur. 4- GRI D
o
VOLTS -->
4-
discriminator network.
We will discuss the limiter-discriminator system number of special operating conditions are com-
first and then the ratio detector systems. bined to make the limiter function properly. First,
the amplifier tube used is usually of the sharp cut-
off type like the 6SJ7 or 6SH7 tubes or the minia-
THE LIMITER SYSTEM ture type 6AU6 tube. Low values of screen and
The limiter stage of the sound channel of a tele- plate voltage and little or no initial control grid
vision receiver immediately follows the final i -f bias are applied to the tube, so that it will quickly
stage and in circuit arrangement appears very simi- overload and plate current cutoff will be rapidly
lar to an i -f stage. In fact, this limiter system may reached.
be considered as the last i -f stage of the receiver, To understand the operation of the limiter fully,
although the purpose is one of amplitude limitation let us first examine a typical plate current -grid volt-
rather than amplification. age (ib-e,,) curve, as illustrated in Fig. 5-4. In
It was stated that for the i -f response curve not general, ib-e,, curves show the amount of plate cur-
to contribute amplitude variations to the f -m signal, rent that will flow for a given value of grid voltage.
it would have to be flat topped with straight sides Most ib-e,, curves take on the shape shown between
as in Fig. 5-1. It is, therefore, seen that the f -m points 1 and 2 in Fig. 5-4., but between points 2
signal, although devoid of a -m variations as it and 3 this curve differs from most curves typifying
leaves the transmitting antenna, will contain ampli- amplifier action. A great deal of this difference is
tude variations at the output of the last i -f stage. brought about by "clipping" parts of the positive
It is the purpose of the limiter to eliminate these halves of the input signal due to the grid and
amplitude variations. cathode of the limiter acting as a diode at this part
OUTPUT F M
SIGNAL CONSTANT
IN AMPLITUDE
Fig. 5-3.-The action of the limiter

in the sound system of a television
receiver is to clip off the amplitude
variations of the input signal, making
its output constant in amplitude.
INPUT FM
SIGNAL VARYING
IN AMPLITUDE
of the input signal. This is commonly known as the relative frequencies present in the frequency
diode clipping and will be discussed in detail later deviation. An instantaneous frequency of say 21.25
on. If diode clipping did not occur, the curve in- me does not undergo any change if it is increased
stead of taking on the shape between points 2 and from 10 volts to 50 volts or reduced from 50 volts
3, would continue onward as shown by the dashed to 30 volts. Any harmonics introduced by the clip-
line. ping section are of no importance, because the fre-
In the region of the curve between points 1 and quencies representing these harmonics are outside
2, the tube, by virtue of the shape of this curve, will the range of the resonant circuit in the plate cir-
act as an amplifier to any input voltage that has cuit of the limiter, a typical circuit of which ap-
values lying in between these two points. Beyond pears in Fig. 5-5. This circuit responds only to the
point 2, the curve is seen to level off and, no matter range of frequencies representing the frequency de-
how high the input grid voltage, the plate current viation on both sides of the carrier, and perhaps a
will be virtually constant in value. Beyond this point little beyond these limits. The harmonic frequencies
on the curve the tube will function as a limiter. are filtered out of the circuit by the tuned trans-
To make sure that the tube functions as a limiter former which couples the limiter to the discrimina-
in ideally producing a constant output signal, the tor. .
instantaneous input grid signal must always rise at The RC network in the grid circuit in conjunction
least to the value at point 2. If an input signal is with the low potentials on the tube help provide
such that it does not rise beyond point 2, any ampli- for proper limiting action of the tube. This will now
tude variations in the input signal will be retained be discussed in detail.
in the output. Similarly, if the input signal swings
below point 1, its negative peaks will be clipped, be- Analysis of Limiting Action
cause below point 1 the plate current is substantial-
ly zero. To understand the true function of a limiter stage
Further analysis of Fig. 5-4 brings a few perti- we should first know something about its circuit
nent facts to light. To get limiting action at both arrangement. Most limiter circuits are very much
extremes of the input signal, this signal must have alike with the chief difference in their grid circuit
a swing that falls outside points 1 and 2 on the arrangements. In most grid circuits, however, a
curve. Any input signal that has amplitude varia- grid -leak resistor and capacitor arrangement is used.
tions not exceeding points 1 or 2 in voltage will, as This RC combination makes it possible for the
a consequence, be reproduced in the plate circuit bias on the tube to change in accordance with the
with these amplitude variations and this is unde- diode clipping action of the input signal. The time
sired. This simple analysis thus reveals that a cer- constant of the RC network determines how quickly
tain threshold of input voltage to the limiter grid is the grid bias change will occur. (The time constant
needed for limiting action at both extremes of the in seconds is determined by the product of R in
input cycle. ohms and C in farads.) In the typical limiter circuit
Further examination of Fig. 5-4 must lead to the illustrated in Fig. 5-5 resistor R and capacitor C
conclusion that such action of the limiter tube re- represent the grid bias network. In other receivers
sults in the development of distortion in the plate the grid resistor R is shunted across C and both are
placed in the grid return circuit. Numerous ver-
sions of these two circuits exist, but the analysis to
follow with respect to the limiter of Fig. 5-5 will
Fig. 5-5-The RC hold for limiter circuits in general.
combination determines The grid arrangement, in conjunction with low
how quickly grid bias
on the limiter tube plate and screen voltage and no fixed bias, causes
will occur, thus caus- automatic bias regulation of the limiter tube. At any
ing automatic bias reg-
ulation. instant of time there is a bias on the grid, but this
bias is changing during the first few cycles of in-
put signal. The instantaneous value of the grid
voltage determines the operating bias at each in-
circuit. Plate -current variations are enlarged repro- stant of time during these first few cycles. After a
ductions of grid -voltage variations. The distortion certain amount of time has elapsed, a point will be
does exist, but is of no consequence because, when reached where the bias on the tube will remain con-
we clip the amplitude of the wave, we do not change stant with constant input signal voltage. The con-
trol grid and cathode of the tube, in conjunction The moment that grid current begins to flow, a
with the RC network, produce the bias. The actual charge is stored on the grid capacitor C, and, as
plate current -grid voltage curve for this circuit is the signal becomes more positive, the charge on the
illustrated in Fig. 5-6. The bias on the tube without
any input signal to the control grid is zero volt. In CLIPPING
accordance with the potentials on the plate and DUE
DIODE
TO
ACTION
screen, a certain amount of d -c plate current flows
through the circuit. As seen by the curve of Fig.
5-6, plate current cutoff will occur when the grid o
has a bias of -8 volts. GRID
VOLTS
Since there is zero fixed bias on the tube, the con-
trol grid will be driven positive during the positive
half of the input signal with respect to the cathode,
which is at ground potential. As soon as the grid EFFECTIVE
BIAS
is driven slightly positive, the grid and cathode act CURVE
as a diode rectifier where the grid takes the place INPUT
SIGNAL
of the diode plate and grid current starts to flow.
As the grid draws current, the coupling trans- Fig. 5-7.-The solid curves above the zero axis show a
former from the preceding stage is effectively load- uniform amplitude due to clipping; the dotted portions above
the zero axis show what the signal would be if no clipping
ed so that signal voltage applied directly tó the occurred.
grid can become only slightly positive. This is shown
in Fig. 5-7 where the positive parts of the input
above the zero -volt line are small due to clipping capacitor increases. On the negative half cycle of
action. Contrast these true small positive swings input signal, the grid no longer will draw current
shown by the solid lines above the zero grid -voltage and the capacitor C will begin to discharge through
axis with the dashed curve. This latter curve shows the resistor R as shown by the current arrow
what the signal voltage would be if there were no through R in Fig. 5-5. An automatic bias will be
clipping due to diode action. developed on the tube, as seen in Fig. 5-7, due to
POSITIVE & NEGATIVE

PEAKS ARE CLIPPED
SO AS TO PROVIDE
IMITING ACTION
L
r
PLATE
CURRENT
(ma.)
.r
-16 -14 -12 -10 -B -6 -4 -2 0 +2 +4
NEGATIVE PEAKS
GRID VOLTAGE ARE CLIPPED IN
INPUT
1 I
THE PLATE
SIGNAL_ CIRCUIT
I
POSITIVE PEAKS ARE

7-1 CLIPPED IN THE GRID
CIRCUIT
r.-----POINT OF BIAS BALANCE
Fig. 5-6.-The clipping action of the input signal to the limiter circuit of Fig.. 5-5 is illustrated in the characteristic
curve shown here.
the voltage drop across R. The "effective bias On the positive half of the third input cycle, a
curve" (idealized curve) represents the average bias. still greater bias has to be overcome before the grid
The true curve would be a bit irregular, because will draw current. This means that a larger negative
the diode current flows in pulses. voltage on the capacitor has to be overcome before
In practice, the values of R and C are so chosen the capacitor can become charged. During the part
that they represent a time constant somewhere in of the cycle when the grid is not drawing current,
the vicinity of a few microseconds. This time con-
. the capacitor is discharging through R. Since a
stant means capacitor C will take so many seconds larger initial voltage existed across the capacitor at
to lose about 63 percent of its charge through the the beginning of this discharge, a still greater volt-
resistor R. If C is, say 50 µµf and R 50,000 ohms, age drop appears across the resistance and, there-
then the time constant will be R X C = 50 X 10-32 fore, a higher operating bias. This process of a
X 50,000 = 2.5 X 10-6 seconds or 2.5 microseconds. continually higher positive signal needed for grid
This time constant is considered high compared with current to flow for each succeeding cycle of input
the time it takes the input i -f signal to complete one signal continues up to a certain point. Operating
cycle. With a 21 -mc i -f signal, the time it will take bias also increases with each successive input cycle
the signal to complete one cycle is equal. to 1/21 me until a constant bias is reached. The increase in
or about 0.05 microsecond. Comparing this value signal input bias reaches a point where a balance
with the time constant of 2.5 microseconds, it is or equilibrium occurs in the system. After a certain
seen that one cycle of the 21 -mc input signal will number of input cycles the grid will reach a point
be completed 50 times as fast as that time repre- where constant operating bias on the tube, and the
sented by R X C, and a half cycle of the input sig- charge and discharge of the capacitor during each
nal will be completed 100 times as fast. By the time succeeding cycle after this point has been reached,
one half cycle of input voltage is completed, the will remain the same. The applied d -c voltages on
capacitor has discharged only about one percent of the tube elements, the RC time constant, and the
its voltage. Upon this very significant fact the oper- strength of the input signal determine this balance
ation of the limiter is primarily based, because it point.
means that during the first negative half cycle not Coming back to Fig. 5-6, the operating bias on
all of the voltage is discharged from C. this ib-e0 curve is seen to be balanced at -6
This means that after equilibrium is reached the
volts.
At the start of the second cycle of input signal,
the grid voltage is different. A residual voltage re- input signal will swing about the operating bias of
mains on the capacitor, because the capacitor keeps -6 volts. A glance at the curve will show that-8
a large portion of its charge during the negative volts on the grid (whether fixed or instantaneous)
half cycle of the signal input. Consequently, no grid will drive the plate current to cutoff. Thus the neg-
current will be drawn during the positive half of ative half of the input signal need swing only 2
the second cycle until this half cycle reaches a value volts to result in plate current cutoff. Consequently,
whereby it will surmount the negative voltage re- the rest of the negative cycle is clipped, or cut off
maining on the capacitor. This will occur at some in the plate circuit. This clipping action of the neg-
time after the start of the positive half of the second ative peak is shown in the diagram. Clipping of the
cycle. During the negative half of the first cycle an positive peaks as previously discussed takes place
operating bias was established due to the drop in the grid circuit as a result of the diode action.
across R, and this bias voltage must be overcome On these positive peaks the operating bias controls
by the positive half of the second cycle before grid grid current flow to a point where the grid is barely
current can be drawn. When this point is reached, driven positive. This is readily noticed in the dia-
grid current will start to flow again, and the capaci- gram where more than 6 volts swing on the positive
tor begin to charge. Since the input signal is con- half cycle is needed to drive the grid into the posi-
tinually varying in a negative and positive direction, tive region for grid current flow. In this manner
the capacitor is always being charged or discharged. the grid acts as a diode rectifier in conjunction with
Since grid current only flows during part of this the cathode, and clipping of the positive peaks re-
positive half of the second input cycle, in which case sults.
capacitor C is. being charged, then during the rest The time constant indicates how quickly the grid
of the second cycle the capacitor will discharge bias will change with change in level. of the input
through R, and a greater operating voltage will be signal to produce amplitude limitation. The ampli-
developed than during the first cycle. tude variations in the final output i -f signal due to
the response characteristics of the i -f circuit are and i -f stages determines the signal input to the
not radical or sudden changes but gradual ones. As limiter system, it is evident that, for a given re-
a result the time constant of the network need not ceiver, the variable factor of signal pickup deter-
be very low to limit such amplitude variations, and mines whether or not there will be enough voltage
time constants of 10 microseconds would be suffi- to drive the limiter.
cient. But one finds that in practice time constants From the foregoing factors we see that for tiffe
between 1 and 5 microseconds are commonly used sound channel of a television receiver with a fixed
so that the operating bias can follow any quick rise amount of r -f and i -f gain, the f -m input signal
in amplitude of the input signal resulting from sud- from the antenna to the first stage of the television
den noise impulses or other interference. If the time receiver has to be above a fixed value. For instance,
constant were too high for the bias to follow this if the required threshold voltage of the limiter tube
impulse, plate current would flow in accordance is 2.5. volts and if the over-all r -f and i -f gain is
with the impulse variation, and the noise impulses equal to 100,000 times, the necessary f -m input sig-
would be passed on to the discriminator to be de- nal to the first stage of the receiver is 2.5 volts/100,000
tected and, eventually, reproduced by the audio or 25 microvolts. This means that any f -m signal input
system. to the receiver below 25 microvolts will not drive
the limiter sufficiently to cause limitation.
AVC From Limiters From this analysis it is seen that the limiter
stage has a very decided influence on the design of
In a -m receivers when automatic volume control the other stages and definitely influences the gain,
(avc) is employed it is taken from the diode de- sensitivity, and selectivity of the receiver.
tector. When avc is used in the sound channel of a The selectivity of the i -f stages helps determine
television receiver employing the limiter -discrimi- the necessary amount of f -m input signal to the
nator method of detection, it is often taken from the antenna. This is best explained in terms of the i.f.'s
grid circuit of the limiter. It will be recalled that response characteristics. The amplitude of à signal
due to the discharging of the capacitor of the time passed through the i -f network is not constant for
constant network, a bias is developed across the the complete range of frequencies the network is
grid -leak resistor. Also, the stronger the input sig- designed to pass. Since all these frequencies must
nal, the greater will be the bias developed across be limited, that part of the i -f curve that produces
the resistor. Another glance at Fig. 5-5 will reveal the lowest amplitude must be the criterion in the
that, due to the current flow through R, the grid design of the gain of the receiver. In other words,
side of the resistor is made negative with respect with the proper r -f and i -f gain, this i -f band-pass
to ground. Since the bias on this tube is a function signal must exceed the threshold voltage required
of the instantaneous input signal and becomes more for the limiting tube at the smallest amplitude with-
negative with increase in signal input, we have a in its band.
point from which avc voltage can be obtained. As mentioned, the limiter tube has a certain
amount of amplification, but beyond some value of
Input Voltage Considerations signal input the plate current of the limiter tube
will not increase but will remain constant and a
One of the rudiments of correctly using a limiter constant output voltage is the result.
system is to know the minimum amount of input
signal required to have the system limit properly THE DISCRIMINATOR SYSTEM
and, in addition, produce an output signal strong
enough for detection. As explained before, a cer- Immediately following the limiter is the discrimi-
tain threshold voltage has to be met to have the nator stage. The many different types of discrimi-
stage following the last i -f stage act as a limiter. nator networks all have the primary function of
Therefore, the output of the last i -f stage must be demodulating the incoming f -m signal. It should be
of sufficient strength to drive the limiter tube be- remembered that this input f -m signal is constant
yond the threshold point of operation. Since the in amplitude due to the action of the preceding
last i -f stage has to drive the limiter tube, all the limiter stage. The type of discriminator circuit used
previous stages together must have produced enough in the sound channel of television receivers is simi-
signal output to cause the last i -f tube to drive the lar to that used in f -m receivers. The general form
limiter tube properly. Since the input signal to the of this circuit appears in Fig. 5-8. There are some
receiver times the fixed amount of gain of the r -f slight modifications of this circuit which appear
in certain sound channels of television receivers and replaced by a resistor and in others no coil or re-
these will be discussed later on. This type of cir- sistor is used. However, these changes do not
cuit, known as the Foster -Seeley discriminator, is change the operation of the discriminator circuit.
also known as the phase discriminator, the center - This will be seen later when modifications of the dis-
tapped secondary discriminator or the center -tuned criminator circuit are analyzed. Let us now discuss
type of discriminator. In many instances this type the discriminator circuit of Fig. 5-8 with respect to
of circuit is referred to just as the discriminator. the voltages illustrated therein.
The purpose of the discriminator is to demodu- Voltage E, is the i -f signal voltage developed
late the f -m input signal so that in its output dr - across the tuned primary circuit. Examining the
cuit will appear an audio signal that is varying in secondary of this i -f transformer, we note certain
frequency in accordance with the audio modulation significant details. It consists of two windings L,
of the f -m signal and is proportional in amplitude and L, in series, resonated to the i -f peak by means
to the amplitude of this audio modulation. of C,. The center tap on the secondary winding is
connected to a coupling capacitor C and, also, to an
Circuit Analysis r -f choke L.
Associated with the two circuits and the r -f choke
From Fig. 5-8 we notice that the discriminator L are three voltages, designated as E,, E,, and E,
consists of two diodes which can be separate tubes respectively, the latter being virtually identical to
but are usually a single duo -diode tube. Associated E, across the i -f transformer primary. To explain
with the discriminator is the tuned discriminator these designations, it is necessary to discuss the
transformer whose primary coil L, is in the plate *coupling between the primary and secondary cir-
circuit of the limiter tube and the center -tapped cuits of this transformer, as well as what happens
secondary coil L, is connected to the plates of the in a transformer when the secondary is tapped at
discriminator. The primary tuned circuit L,C, and the midpoint. How does this type of discriminator
the secondary tuned circuit consisting of inductance operate ? In brief the operation can be divided into
L, + L, and capacitor C, are both resonant to the three major actions, although more conditions than
i. f.
just three are, actually involved.
The two diodes are connected in the form of a In the first place, although a single tuned winding
differential rectifier system where the two resistors is used for the secondary circuit, the center tap on
R, and R, are the respective loads for diodes D, this winding causes a division of the signal voltage
and D,. The d -c paths through these diode circuits developed in the tuned circuit across the two halves
are completed through coil L and the respective half of the secondary winding, that is across L, and L,.
of the center -tapped coil associated with each cir- .The signal voltages across these two halves are
cuit. Besides providing the d -c paths between the always equal to each other, irrespective of the f re-
diode plates and their respective cathodes, this com- quency of the signal voltage fed into this circuit
mon coil has another function which will be shown from the primary.
later. In some discriminator circuits the coil L is
The second major consideration is that the signal
voltage present across the primary winding L, is
also present across winding L, which is common to
both halves of the secondary circuit with respect
to the signal voltages eventually applied to the two
diodes D, and D,.
The final major action is the phase relation which
exists between the signal voltage across L,, which
we can call E,, and the signal voltage across L
which, because it is the same as that across L,, is
also identified as E,; also the phase relation between
the signal voltage across L,, or E,, and the signal
voltage across L, or E,. The function of this dis-
criminator network with particular reference to
these three actions will now be discussed in detail.
Fig. 5-8.-Schematic of a conventional phase discrimina-
Two methods of coupling the signal from the
tor detector. primary to the secondary circuit are used in this
system. The resonant primary is inductively coupled tion of the discriminator. That is one of the reasons
to the resonant secondary winding; at the same time why the circuit is often known as the phase dis-
the signal voltage E, across the primary is fed to criminator.
the r -f winding L via the coupling capacitor C. If
the circuit of C, L, and C5 is traced, it will be seen
Resonance Conditions in the
that L is in shunt with the tuned primary grounded
Phase Discriminator
through C,. None of the quantities C, L, C, or C5
is of a magnitude to alter the resonant conditions Let us first examine this circuit at resonance. At
of C, and L,, the resonant primary. Thus, with re- resonance, the frequency of the applied signal and
spect to magnitude and phase, whatever signal volt- the resonant frequencies of the tuned circuits are
age exists across C1 -L1, also exists across L. The both the same. Since the inductances and capaci-
direct connection between the coupling capacitor C tances of a tuned circuit effectively cancel each other
and the mid -point of the secondary winding is of no at resonance, the circuit behaves like a resistance.
consequence to the signal transfer between the pri- In a resistive circuit the current is in phase with
mary and the secondary tuned circuits ; however, it the voltage, so in the secondary tuned circuit the
is the point to which the choke L must be connected induced current, call it I, caused to flow by the in-
to complete the differential rectifier circuit. Thus, duced voltage, call it E, is in phase with this in-
the secondary system receives signal voltages in two duced voltage. It should be remembered that this
ways the resonant secondary receives its signal
: induced voltage is effectively in series with the in-
voltage by inductive coupling, and the r -f choke de- ductance and capacitance of the secondary tuned
rives its signal voltage by means of capacitive coup- circuit.
ling through the fixed capacitor C. The in -phase relationships between E and I are
The equal voltages across each half of the second- indicated in the vector diagram of Fig. 5-9 where
ary winding are obtained in the following manner. vectors OI and 0E, the respective induced voltage
When a winding is tapped at the mid -point and a and current vectors, are seen to be in phase with
voltage is induced in that winding by means of a each other. The voltage across the primary circuit,
varying magnetic field, the total voltage developed designated as E, is 180° out of phase with the volt-
across the entire ,winding divides between the two age induced into the secondary circuit. This voltage
halves. This is logical in view of the fact that half E, is the main voltage upon which all the other
the total number of turns exists between the center voltages are based. Consequently, this voltage is
tap and one end, and half between the center tap drawn as vector OE, along the 0° reference line.
and the other end. So, whatever the nature of the
signal voltage which will be developed across the E I E,
tuned secondary circuit C5 -L2 -L,, it is possible to Fig. 5-9.-Vector 0
diagram of the re-
show that this voltage divides into two parts, that lations at resonance
is, across each half of the winding. These voltages between the current
and voltages in the
are designated as E, and E,. phase discriminator.
The above analysis of the discriminator circuit EE,
will enable us to understand the discussion that
follows. We will see how the input f -m signal to In the vector diagram of Fig. 5-9, vector 0E, is
the discriminator becomes demodulated resulting in 180° out of phase with the induced voltage vector
an audio signal at the output which is the audio 0E. Since the voltage across a pure inductance
modulation of the input f -m signal. In order to leads the current through it by 90°, the voltage
show how this occurs we have to consider the input drops E, and E, across the secondary coil (called
f -m signal (which is varying in frequency above reactive voltage drops because the inductance is
and below its center value) at three different in- considered to be a pure inductance containing negli-
stants of time. We will investigate how the circuit gible resistance) lead the current I flowing through
functions at the time when the f -m signal is exactly it. This is indicated where vectors OE, and OE,
equal to the tuned i.f. of the discriminator. trans- are leading the induced current by 90°. This also
former (i.e. at resonance) and at the time when its means that the reactive voltage drop across the
instantaneous frequency is above and below the secondary coil is lagging the primary voltage E, by
center i.f. Phase relationships among the various 90°, and vector OE, is seen to be leading vectors
voltages are the criterion in establishing the opera- OE, and 0E, by 90°. The 90° phase difference be-
tween these voltages is very important to the oper- it is readily seen from Fig. 5-11 that voltages E,
ation of the discriminator. It should still be re- and E, are active on diode D, and voltages E, and
membered that this voltage E, also exists across E, are active on diode D,.
coil L of Fig. 5-8 in the same phase and magnitude Further examination of this simplified circuit re-
as that existing across the primary tuned circuit. veals that the rectified current flowing through the in-
dividual diode circuits puts certain polarities on their
Fig. 5-10. - Re-
vised vector diagram
load resistors. Since the external current in a diode
rectifier circuit flows from plate to cathode, the
of Fig. 5-9. Note
that the vector 0E3 currents in the diode load resistors will be flowing
E is shifted 180° in in opposite directions, and the polarities across the
phase from its posi-
tion in the previous individual load resistors will be bucking each other.
E: figure. Thus between points A and B a voltage will exist
which will be the difference between the voltage
That the secondary is center tapped means that drops across resistors R, and R,.
it is in a push-pull arrangement, and hence voltages If the voltages E, and E, have the same phase
E, and E, are equal in magnitude but 180° out of angle with respect to the voltage E1, both diode cur-
phase with each other, as referred to the center tap. rents will be equal in value, and the same voltage
However, the same current flows through both parts drop will appear across each load resistor R, and
of the secondary coil, so that a 90° phase relation R,. Since each resistance voltage drop is opposite
must still exist between each voltage and the cur- in polarity to the other but equal in value, the total
rent but in one case one of the voltages is effec-
;
voltage measured between points A and B will be
tively leading the current and in the other case the zero. Under these circumstances the output of the
voltage is effectively lagging the current; by 90°.
This means too that one half of the secondary volt- E4
age drop is leading voltage E, by 90° and the other
half lagging voltage E, by 90°. All of this is indi-
cated in the revised vector diagram of Fig. 5-9 as
shown in Fig. 5-10, in which vector 0E, of the
previous vector diagram has been shifted 180°.
To demonstrate how all these voltages affect the
duo -diode circuit we have redrawn that part of Fig.
5-8 appearing to the right of the secondary of the Fig. 5-12.-Vector diagram of the current and voltage
transformer in simple form in Fig. 5-11. In this relationships of the phase discriminator of Fig. 5-8 at
resonance when both secondary voltages E2 and E3 have a
figure we have made two separate circuits of the 90° phase difference from E1.
diodes, showing the respective' voltages that act
upon each diode. These two simple circuits then
differential rectifier circuit is zero. If, however, the
phase relationships between E, and E, differ, a
Fig. 5-11. - Sim-
plified schematics of
differential voltage will exist between points A and
B, because the voltage drops across R, and R, no
the diodes of Fig. longer will be equal to each other as different cur-
5-8 and their respec-
tive load resistors. rents flow through each diode.
These two schemat- Zero voltage exists across points A and B when
ics are combined on
the right. This is the resonant frequency of the tuned discriminator
permissible since the i -f transformer is exactly equal to the applied fre-
the voltage El is
common to both quency. This is simply indicated by the vector dia-
diodes. gram of Fig. 5-12 where both secondary voltages
have the same phase difference, namely 90°, from
voltage E,. This diagram is nothing more than a
are combined to show how they actually work to- duplicate of that diagram of Fig. 5-10 with the ex-
gether. This figure and Fig. 5-8 show that voltage ception that the two voltages active on each diode
E, is common to both diodes since it exists across are added vectorially. Thus, in Fig. 5-12, vector
the inductance L. Also, since voltage E, is active 0E4 represents the resultant vector of the vector
on diode D, and voltage E, is active on diode D,, addition of voltages E, and E, across diode D, and
vector 0E5 represents the resultant vector of the no longer will be in phase with the induced voltage
vector addition of E, and E, across diode D,. Re- E but rather will lag this voltage by a certain
sultant vectors 0E4 and 0E5 are shown to be equal amount, depending upon the extent to which the in-
in magnitude, causing the same current to flow in stantaneous f -m signal is greater than the tuned f re-
each diode circuit. Thus, equal but opposite volt- quency of the transformer. This is all indicated in
ages are developed across diode load resistors R, the vector diagram of Fig. 5-13 for the off -reso-
and R, producing zero voltage between points A nance condition now being discussed. Voltages E
and B. and E, are still seen to be 180° out of phase but the
phase relationships of the other component voltages
differ somewhat from those of the vector diagram
Applied Frequency Higher Than Resonance of Fig. 5-12.
For the sake of argument let us say that the dif-
When the instantaneous value of the f -m signal ference in frequency between the instantaneous fre-
is equal to its center frequency component, we have
quency of the f -m signal and the tuned i -f trans-
the frequency applied to the discriminator transformer is such that the amount of inductive react-
former equaling the resonant frequency. The situa- ance remaining is sufficient to cause the induced
tion for this was discussed in the preceding sec- current I to lag the induced voltage E by 35°.
tion. At either side of the center frequency com-
No matter what the phase relationship between
ponent of the f -m signal the instantaneous frequen- the induced voltage and induced current, the two
cy is different from the resonant frequency of the
voltages E, and E, across the individual halves of
i -f transformers. Under these conditions the dis-
the secondary are still 180° out of phase with each
criminator transformer is tuned below or above other and equal in magnitude. The induced current
the incoming i -f signal.
flowing through this secondary still bears the same
Let us now consider an instantaneous value of
phase relationship to these secondary voltages. Re-
the f -m signal greater than the center i.f. The dis- gardless of the phase difference between E and I,
criminator transformer then is tuned below the secondary voltage E, will still lag current I by 90°,
incoming i.f. The circuit is still the same as in Fig.
and secondary voltage E5 will still lead current 1 by
5-8, and the nonresonant conditions do not alter the
90°. This is indicated in the .vector diagram of Fig.
fundamental rules of the action of the primary
5-13; and if this vector diagram and that of Fig.
circuit, so that the voltage E, that exists across this
5-12 are compared, these phase relations will be
circuit also exists across L in both the same phase
seen to hold.
and magnitude.
Induced voltage E in the secondary remains 180° E3 E
out of phase with the primary signal, for this too
is a fundamental condition which is not altered by
nonresonance conditions. However, the phase re-
lationship betwen the induced voltage E and the
current I, which it causes to appear in the secondary
circuit, is affected by the state of resonance, and in
turn alters related conditions.
When the applied frequency is higher than the Fig. 5-13.-Vector diagram of the current and voltage
resonant frequency, the reactance of the secondary relationships in the phase discriminator when the instan-
coil becomes greater than the reactance of the ca- taneous frequency of the signal is higher than the reso-
nant frequency of the transformer.
pacitor. This accords with the fundamental law that
inductive reactance varies directly with frequency,
and capacitive reactance varies inversely with f re- Let us further compare these two vector dia-
quency. Accordingly, a portion of the inductive grams. To keep constant the 90° phase relations
reactance will be offset by the capacitive reactance, between voltages E2, E,, and current I, then when
but a certain amount of inductive reactance will current I lags induced voltage E by 35°, voltage
remain to exert a control on the induced current. vectors 0E, and 0E, are both shifted 35° clock-
The circuit as a whole now appears as an inductance wise to keep these 90° relationships intact. The
and resistance in series, rather than as a resistance complete vector line E2 -0-E, is shifted 35° in a
alone, which is the case at resonance. negative direction. When the respective voltages ap-
Under this circumstance the induced current I plied to the individual diodes are added under these
circumstances, it will be seen from the vector dia- when the individual i -f voltages across the diode
gram of Fig. 5-13 that the resultant vector 0E4 circuits are combined vectorially, it will be seen that
representing that voltage across diode Ds and re- resultant vector 0E, applied to diode D, is greater
sultant vector 0E, representing that across diode in magnitude than resultant vector 0E4 applied to
D, are no longer equal, but that vector 0E4 is great- diode D2. Therefore, the current in the circuit of
er than vector 0E5. In this instance diode D, will diode D, is greater than the other diode current. This
draw the greater current, and in Fig. 5-11 load re- means a greater voltage drop exists across R,, the
sistor Rs will have a greater voltage drop than re- load resistor of diode D,, than across R2, the load
sistor Rl; hence a differential voltage will exist of D2, and a differential voltage exists between
across point A to B, with point B being more posi- points A and B of the diode circuit of Fig. 5-11.
tive than point A. This is the same as saying point However, under these conditions the polarity of
A is negative with respect to point B. point A will be more positive with respect to the
junction of R, and R, than point B. That is, point
B is negative with respect to point A.
Applied Frequency Lower Than Resonance
When the instantaneous value of the f -m signal Summary
input to the discriminator circuit is such that it is
Jess than the resonant frequency of the discrimina- Summarizing the action described, it is evident
tor transformer, the differential voltage will still that if a varying frequency input signal (one which
exist across the diode loads, but the polarities will varies in frequency around a mean) is applied to
be reversed. Let us see how this happens. the discriminator network-provided that the range
We still are at off -resonance conditions, even of frequencies covered is not beyond the acceptance
though we are on the lower side of the resonant bandwidth of the discriminator transformer-an
frequency, and the same 180° phase relationship output signal which changes in amplitude and polar-
between E and E, exists. When the applied f re- ity will be obtained. The frequency swing of the
f -m signal is determined by the amplitude of the
quency is lower than -that of the resonant frequency
of the i -f transformer, the impedance of the second- audio modulating signal. The greater the amplitude,
ary of the i -f transformer is such that the capaci- the wider the frequency swing, and the smaller the
tive reactance more than balances out the inductive amplitude, the lesser the frequency swing. The fre-
reactance, and the secondary is primarily capacitive. quency of the audio modulating signal determines
Since this circuit is capacitive, the induced current the time rate change of the frequency swing. This
I leads the induced voltage E. If the off-resonance frequency swing, usually called frequency devia-
conditions are such that a phase angle of 35° again tion, varies an equal amount on either side of the
exists between I and E, I will be leading E by 35°, center or carrier frequency of the f -m signal for
as seen in the vector diagram of Fig. 5-14. Since each cycle of audio modulating signal. If the audio
the 90° phase relations between voltage E, and cur- modulation is a sine wave of constant amplitude or
rent I, and voltage E, and current I must still exist, any similar signal of constant amplitude, the fre-
these two voltages are effectively shifted in phase quency deviation will be the same for every cycle
35° in a counterclockwise or positive direction. This of modulation. However, in speech or music the
is indicated in Fig. 5-14 where vectors OE, and audio modulation is anything but constant in ampli-
0E, are still 180° out of phase with each other, but tude and, therefore, the frequency deviation for
no longer 90° out of phase with vector OE,. Now each cycle is usually different.
The output signal from the discriminator is de-
termined by the amount of frequency deviation ;
the less the frequency deviation, the less the de-
parture from a 90° phase relationship between the
reactive voltages E, and E,, and also E, and E,.
The greater the frequency deviation, the greater is
the difference in angular displacement between Es
and E,, and E, and E,; so that the differential volt-
age obtained from the diodes is greater. When
Fig. 5 -14. --When the instantaneous frequency of the f -m viewed from the angle of audio intensity, the great-
signal is lower than the resonant frequency of the trans- er the differential voltage from the rectifiers, the
former, the current I will lead the induced voltage E.
Compare with Figs. 5-12 and 5-13. louder the audio signal, since the extent of .devia-
tion at the transmitter is a function of modulating 50 kc wide. In fact, the greater the linearity (with-
voltage level. The greater the modulating voltage in reasonable limits) the better, and an over-all
level within prescribed limits, the greater the f re- 200-kc linearity is considered very desirable. This
quency deviation. accomplishes two things : First it means that the
In brief, then, the differential output voltage is a receiver does not have to be tuned very accurately
function of the rate of deviation of the f -m signal for the resting frequency to fall in the middle of
as well as the amount of frequency deviation, Since the discriminator characteristic. If the receiver is
the amplitude of the audio modulating signal deter- mistuned somewhat, no distortion will result pro-
mines the amount of deviation and since the fre- vided the receiver is not so badly mistuned that the
quency of the audio determines the rate of change frequency variations in the signal extend into the
of the deviation of the f -m signal, it becomes readi- nonlinear or curved portions of the discriminator
ly apparent that the differential voltage across the characteristic. Secondly, the fact that the character-
two diodes will be an audio signal equal in fre- istic of the discriminator is linear over a greater
quency and proportional in amplitude to the audio range than that actually required means that the
modulation signal. linearity will be more nearly perfect over the center
portion which is actually used in reception. The
high degree of linearity obtained in this way makes
The Discriminator Output Curve demodulation of the signal possible with practically
If the output voltage of the discriminator is plot- no distortion.
ted against the instantaneous values of frequency
input, we obtain the well-known S-shaped response Pre -emphasis and De -emphasis
curve. A typical S-shaped discriminator response
curve is illustrated in Fig. 5-15. The center point of In the audio input to a transmitter the amplitudes
the curve is at zero voltage output and at the center of the higher audio frequencies are relatively low
frequency of the f -m signal, namely the i.f. Above the as compared with the rest of the audio -frequency
zero point on the vertical axis, the output voltage is spectrum. This is due to the natural distribution of
positive, and below the output voltage is negative. In sound in radio program material, not to any char-
similar fashion, to the right of zero on the horizontal acteristic of a transmitter. As the programs pass
axis the input f -m signal has instantaneous values of through the transmitter to the receiver and then
frequency greater than the i.f. and to the left the in- through the receiver, noise is unavoidably added to
stantaneous frequencies will be lower than the i.f. the desired audio signal. This noise is predominant-
ly high -frequency audio, and this condition tends
The actual shape of the curve is not determined
by the frequency deviations of the input f -m signal. to produce a low signal-to-noise ratio at the high
The shape of the S curve is determined by the de- audio frequencies because the signal is relatively
sign characteristics of the discriminator circuit. In weak and the noise relatively strong. In a communi-
order to have the output audio signal free from
distortion, the input f -m signal to the discriminator
should have frequency deviations that do not fall A
outside the linear range of the S characteristic. The
linear range in the curve of Fig. 5-15 is indicated
by points A and B. If the input f -m signal has f re- FREQUENCY
quency deviations that fall outside this linear range, ABOVE RESONANCE
they will operate in part over the nonlinear portion of FREQUENCY
BELOW RESONANCE
the S curve and distortion in the audio output signal
will result.
The curvature of this response characteristic is B
introduced by the nonlinearity of the detector out-
NEGATIVE
side the band of frequencies for which it is de-
signed to operate. 11111m.-
We know that 50-kc peak -to -peak deviation is FREQUENCY

that for 100 -percent modulation for f -m sound Fig. 5-15.-The S curve represents the typical output
transmission for television, and therefore the linear characteristic of the discriminator circuit with the output
a -f voltage plotted against the instantaneous values of in-
portion of the S characteristic should be at least put frequency.
82 RIDER'S-"HOW IT WORKS
cations system, such as commercial f -m broadcast- scribed in the discussion of limiters, in that it is equal
ing, where the audible spectrum up to 15,000 cycles to the product of R in ohms by C in farads.
is utilized, this low signal-to-noise ratio at the high In Fig. 5-8 the R, and CB combination is a de -
audio frequencies is most undesirable. However, emphasis that appears in the output of the detector
it is possible to overcome this condition to a large network before the audio signal is applied to the
extent by the use of pre -emphasis which means in- preceding audio amplifiers.
creasing the relative strength of the high -frequency De -emphasis networks also appear in the output
components of the audio signal. However, at the of the ratio detector as well as the discriminator.
same time that pre -emphasis is beneficial in improv-
ing the signal-to-noise ratio, it introduces a defect
in that it upsets the natural balance between the Modification of the Discriminator Network
high- and low -frequency tones in the program ma- We mentioned awhile back that there exist mod-
terial. ifications of the discriminator network of Fig. 5-8.
This defect is compensated for at the television These modifications vary somewhat but their func-
receiver by means of a de -emphasis circuit at the tions are all the same. That is, the voltage existing
input to the audio amplifier. The de -emphasis cir- across the primary of the discriminator transformer
cuit reduces the high -frequency audio exactly as the essentially exists across each diode. In some of the
pre -emphasis increases it. However, it operates on modified circuits the method of coupling this volt-
both the high -frequency program material and the age is readily noticed, while in others it is not.
high -frequency noise. Thus, it does not change the
improved high -frequency audio signal-to-noise ratio,
TUNING METER
which is obtained by means of pre -emphasis, while TO ,SWND DISCRIMINATOR
PLATE OF
it does re-establish the tonal balance of the program 6A U6
LIMITER TUBE
material, which is lost in pre -emphasis.
This de -emphasis network is usually inserted
between the detector and audio amplifier in the
sound channel and has a frequency characteristic
just opposite to that of the pre -emphasis network.
The FCC has set a standard of a 75 -microsecond
time constant for the pre -emphasis network in the
transmitter. Consequently, the de -emphasis network
in the receiver also should have a time constant
equal to 75 microseconds. Fig. 5-17.-Sound discriminator circuit of the Du Mont
model RA -102. Note that the coil L in Fig. 5-8 has been
A typical de -emphasis network is shown in Fig. replaced by resistor R.
5-16. The effective impedance offered to the audio Courtesy Allen B. Du Mont Laboratories, Inc.
voltage is the series combination of C and R. As
the frequency of the audio signal increases, the re-
Most of the discriminator circuits of today as
actance of the capacitor C decreases, or as the f re -
used in television sound channels do not employ
a coil .(such as L in Fig. 5-8) for directly coupling
the signal voltage across the discriminator primary
to the two diodes. The modified discriminator cir-
Fig. 5-16.-A typ- cuits discussed below are the more common types
ical de -emphasis net-
work as used in the found in use today.
sound section of tel- The sound discriminator circuit of the Du Mont
evision receivers.
television receiver model RA -102 is reproduced
in Fig. 5-17. The only difference between this cir-
cuit and that of Fig. 5-8 is that coil L in the latter
is replaced by resistor R in the former. This resis-
quency increases, the reactive voltage drop de- tor R performs the same function as coil L in that
creases. The audio voltage to the grid of the tube, the signal voltage drop across the primary of the
therefore, decreases with increase in frequency and discriminator transformer also appears across re-
the reverse of pre -emphasis is obtained. The time sistor R effectively in the same phase and magni-
constant is determined similarly to the method de - tude as across coil L. The rest of the discriminator
circuit of Fig. 5-17 appears and functions the same primary and capacitor C200 to ground. The 6500
as that of Fig. 5-8. The tuning meter MI in Fig. µµf value of C200 has a reactance a little over one
5-17 is used as an indicating device to show when the ohm at 20 mc, so practically all of the limiter out-
station is in tune. When properly tuned, the needle put voltage appears across the primary coil. Across
of this meter is oriented in the center of the scale. this coil and C200 capacitor arrangement is capaci-
The R56 and C56 combination represents the de - tor C202 and the two parallel resistor networks as
emphasis network. shown in Fig. 5-:9. Resistor R219 is in parallel
with resistor R220 and capacitor C203. All the com-
6AL5 ponents to the right of the discriminator primary
AUDIO D15CR.
TO PLATE OF
6 AU6 LIMITER TUBE V107
R236
22K
have an amount of the i -f signal voltage across
T11)
7
C205 them, proportional to their respective reactance or
ID 0025/1F
resistance. However, at an i.f. of about 20 mc, the
C206
4 Zl
270
TC203
01UF
reactance of capacitors C202 and C203 is very
R220
100K R22
small (that of C202 being about 150 ohms and that
R221
I MEG of C203 being about 30 ohms) so that practically
B+ 27K
C200 SOUND
all the signal voltage appearing across the discrim-
- 6500
R 219
C204
VOL.
CONT.
inator primary also appears across both load re-
100K .09µf
- sistors R219 and R220. Since resistor R219 is the
load for one diode and resistor R220 the load for
Fig. 5-18.-Discriminator detector used in the Crosley
models 307TA, 307TA-50. the other diode, it is readily seen how the signal
Courtesy Crosley Div. AVCO Mfg. Corp.
voltage which appears across the discriminator pri-
mary also appears across each diode.
Another modification of the discriminator ap-
pears in Crosley models 307TA, 307TA-50. Com-
V7A
paring the circuit of Fig. 5-18 with that of Fig. I-
T24
DISCRIMINATOR
5-8, we find that coil L is missing and that only TO PLATE I
C81
-6AQ76T
OF 6SV7 I
one capacitor, C203, appears across the discrimina- TUEE 47 1
tor load resistors. It may not be readily apparent I
how the signal voltage across the discriminator

primary of Fig. 5-18 is coupled to both diodes, al-
though the transformer -coupled voltages that exist
across each individual half of the discriminator
secondary of Fig. 5-18 effectively appear across
their respective diodes in similar fashion to that C82
.006
discussed for Fig. 5-8. However, it may not be read- MF
ily evident how the signal voltage across the discrim-
inator primary is directly coupled to both diodes, AUDIO OUTPUT
C65
since coil L of Fig. 5-8 or resistor R of Fig. 5-17 82
is not present.
Fig. 5-20.-The discriminator circuit used in the General
C202 56µµi
Fig. 5-19.
two paths of Fig.
-
The
Electric model 802.
Courtesy General Electric Co.
DISC. 5-18 from the top

6220
Tq ANSF.
PRIMARY
1006 of the discriminator
6219 primary coil to Another type of modified discriminator uses the
1 100K
L ground are redrawn
C200
C203
270 for the sake of the diode 'sections of a duo -diode high -mu triode. The
6500 Nrf discussion. 6AQ7-GT, as used in the General Electric model
µ!F
802, is such a tube. The way the discriminator cir-
If we trace the circuit of Fig. 5-.18 from the cuit appears in the- over-all schematic is illustrated
top of the discriminator primary coil to ground. in Fig. 5-20. The diodes of this tube can be used
we can take two paths after passing through coup- as a discriminator, because one cathode is used for
ling capacitor C202. These two paths are redrawn the diodes and a separate one for the triode section of
in Fig. 5-19. The output voltage from the limiter the tube. A schematic of the 6AQ7-GT tube ap-
is effectively across the discriminator transformer pears in Fig. 5-21. From this drawing and that of
Fig. 5-20 we can see how the two diode sections Tracing the d -c path for each diode, we find that
are utilized. In order to understand better how this resistor R, is the load for D, and resistor R, is the
circuit functions we have redrawn it in Fig. 5-22. load for D,. The reference voltage E, is capacitance
The voltages, coils, diodes, and load resistors are coupled through C, to the common cathode of the
diodes. Capacitor C, is a bypass capacitor for the
6AQ7 GT primary tuned circuit and completes the r -f path
to ground. The reactance of C5 at the i.f. is so small
Fig. 5-21 (left) -
Schematic
of the 6AQ7-GT duo - diode
that negligible i -f voltage appears across it.
If we trace the i -f path from the top of L,, we
high -mu triode tube.
find that there are essentially two parallel paths ,
which appear across the primary circuit to ground.

3 DISCRIMINATOR This circuit is shown in simplified form in Fig.
5-23. Going from the top of L, we pass through
C,, and find two paths available one through re-
:
Fig. 5-22 (right) sistor R, to ground and the other through resistor
-The discriminator R, and capacitor C4 to ground. The capacitances of
circuit of Fig. 5-20
has been redrawn to C, and C4 are so chosen that they will offer a low
show the use of the reactance at the i.f. compared with the resistance of
6AQ7-GT more R, and R,. This means that practically all of E,
clearly. AUDIO
OUTPUT also appears across R, and R,. So far as the high -
f requency i.f. is concerned, R, and R, are both ef-
given designations to conform to those used in the fectively in parallel with L,, and the reference volt-
conventional discriminator circuit of Fig. 5-8 so age E, also appears across the load resistors R, and
that we can compare them. R,.
The chief difference between the circuit of Fig. Since R, is the load resistor for D,, both voltages
5-22 and the conventional discriminator is the E, and E, act on diode D,; and, since R, is the load
method of applying reference voltage E, to both resistor for D,, both E, and E, act on diode D,.
diodes D, and D,. Coil L,, and thus voltage E,, are The on- and off-resonance conditions function as
common to the upper diode D,, and coil L, and volt- in the conventional discriminator. The audio output
age E, are common to the lower diode D,. Capacitor appears across C4 or between the high side of R.
C4 connecting L, and L, is of high enough capaci-
and ground. In this circuit, resistor R, and capaci-
tance, so that both coils are effectively in series to tor C, represent a de -emphasis network, and the de-
the i.f. As with the conventional discriminator, ca- emphasized audio is taken across C,. The triode
pacitor C, is shunted across these two coils and section of the tube is used as the first audio voltage
with them forms the secondary tuned circuit. This amplifier.
In the discriminator circuit of the Consolidated
Television model 2315, a resistor is. used in place of
the usual coupling coil L. This modified circuit is very
similar to that of the Du Mont model RA -102 dis-
Fig. 5-23.--A sim- cussed previously. The interesting and unique thing
R2 plified drawing of about this network is the method of detection used.
the i -f paths that
appear across the For the detection process two diodes are still ene
transformer primary ployed, but they are two germanium crystal diodes
to ground in Fig.
5-22. instead of the conventional duo -diode tube.
The circuit for this arrangement appears in Fig.
5-24. The two 1N34 diodes are connected in a sim-
ilar fashion to the usual duo -diode tubes. Although
1N34 crystal diodes are more expensive than any
analysis reveals how the respective induced voltages duo -diode tube such as the 6H6 or 6AL5, they offer
E, and E, are applied across the individual diodes a number of advantages as compared with the tubes.
D, and D,, but the method of obtaining E, across A 1N34 crystal diode occupies much less space than
both diodes is not readily evident. the conventional duo -diode tube. With these crys-
tais, sockets are not necessary and they do not re- From this diagram we notice one thing that is com-
quire any heater voltage. mon to all ratio detector circuits, namely, that the
Aside from these physical characteristics, crystal two diodes used are wired in series aiding with re-
diodes such as the 1N34 offer greater electrical spect to the load instead of series opposing as in
the discriminator detector circuit. By tracing the
circuit it will be found that the plate of one diode
TO PLATE is connected to the cathode of the other diode
Of 6AC7
through the secondary of the transformer. The
TUBE
-
VL.
L,
N
plate of the latter in turn is connected to the cathode
of the former diode through a battery. Compare
these connections with that of the discriminator de-
tector in Fig. 5-8, and the difference will be im-
mediately apparent. This difference in circuit ar-
rangement of these two detectors is one quick
method of telling them apart.
R41-33KIL Coming back to Fig. 5-25, we find that the trans-
former network in conjunction with C and L is
Fig. 5-24.-The discriminator circuit used in the Con- similar to the discriminator detector arrangement
solidated Television model 2315. Note that a resistor is used
in place of the usual coupling coil. in that the voltage across the primary, L,, is also
Courtesy Consolidated Television Corp. across L. This voltage in conjunction with the indi-
vidual voltages across Ls and L, is effectively on
each respective diode. Across the output of this
stability and thus a better balance is obtained in the
circuit appear two capacitors C. and C, and in par-
discriminator circuit. The total shunt capacitances
allel with them a battery of fixed voltage, EB. In the
such as the input and output tube capacitances and
discriminator detector, besides two capacitors, two
stray capacitances are less when using germanium
resistors also existed, across which the differential
crystal diodes than with conventional duo -diode tubes.
output voltage was developed. However, when any
This is another feature which helps toward better dis-
amplitude changes occurred in the input signal to
criminator balance. The positive or plus terminal of
the discriminator detector, the output voltages across
the 1N34 crystal is usually the plate of the diode
each resistor changed, making the differential out-
and the negative or minus terminal the cathode.
put voltage different, indicating that the detector
was responsive to a -m signals as well as to f.m.
THE RATIO DETECTOR In the simplified ratio detector circuit of Fig.
5-25, the voltage from diode to diode in the output
In the previous discussion on the discriminator side of the circuit is maintained constant at voltage
detector circuits, it was indicated that a limiter was EB by the battery. Consequently, the total voltage
necessary to produce an input f -m signal to the dis-
criminator that was constant in amplitude. The lim-
iter is needed because the discriminator responds to
-
amplitude changes in the signal as well as to f re-
quency changes. To dispense with a separate limiter
C2
tube, the ratio detector, was developed. This detec-
tor responds to frequency changes only and not to
amplitude changes in the input system. Although its INPUT
E _
-F
operation is different from that of the discriminator
I
SIGNAL
detector, it is nevertheless somewhat similar in cir-
cuit analysis. This will be seen as we progress with
the analysis of the detector.
iAUDiO
OUTPUT
Simplified Ratio Detector
To understand fully the operation of a typical Fig. 5-25.-Simplified schematic of a ratio detector cir-
ratio detector circuit let us first study a simplified cuit. Note that the diodes are connected in series aiding
rather than series opposing as in the discriminator de-
version of such a circuit as illustrated in Fig. 5-25. tector.
across capacitors C, and C, will always be equal to and also since the effective strengths of the individ-
EB. According to the polarity of the battery con- ual carriers at the receiver change, it was found that
nection, no current will flow in the circuit until a the best thing possible was to have a relatively con-
signal is applied. The d -c path of this output cir- stant voltage across the output determined by the
cuit is through the battery, and the a -c path through average value of the incoming signals. This was
the two capacitors C, and C,. When an f -m signal is accomplished by placing a parallel RC network in
detected by this arrangement, the individual volt- the output of the circuit instead of the battery, as
ages Ex and Ey across capacitors C, and C, re- shown in the typical ratio detector circuit of Fig.
spectively, will be constantly changing due to the 5-26. Let us now study this circuit in conjunction
change of deviation of the f -m signal, but their sum with the new RC output circuit, and the over-all
will be constant due to the battery voltage EB. At action of this detector will become apparent.
all times EB = Ex + Ey. Since the individual values This new circuit is very similar to the discrimina-
but not the sum of Ex and Ey can change, it is their tor detector arrangement in many ways. For in-
ratio which will be constantly changing, and by plac- stance, the voltage E, can be considered as effective-
ing a potentiometer across C, the audio modula- ly the same, even though it is not obtained from the
tion of the f -m signal can be tapped off this resist- limiter. In both circuits the signal is of the i.f. This
ance. This will all be clearer when a typical ratio voltage E, is coupled to the rest of the circuit in
detector circuit is analyzed. two ways by induction into L, and L, and by capa-
:
Any amplitude variations in the input signal will citive coupling across C, to L and C4. The react-
not appear across either C, or C, as changes in volt- ances of C, and C4 at the i.f. are negligible com-
age due to the constant voltage across the output pared with the reactance of the choke L. Conse-
due to the battery. Only frequency changes will ap- quently, the voltage drop across L is likewise E, as
pear across both capacitors. This is different from the seen in Fig. 5-26. Depending on the degree of coup-
discriminator detector circuit, where both amplitude ling between the primary tuned circuit and the sec-
and frequency changes were recorded as voltage ondary tuned circuit, a certain amount of voltage.is
changes across the load resistors. This simplified form coupled across each half of the center -tapped sec-
of the ratio detector was analyzed first because ondary. So far as voltages E,, E,, and voltage E,
through the use of the battery we were easily able to across L are concerned, they all function in exactly
show the fixed voltage across the output and why the same manner as the corresponding voltages in
this detector responds only to frequency variations. the discriminator detector circuit of Fig. 5-8. This
is true, too, for the vector diagrams of Figs. 5-9
and 5-10 and Figs. 5-12 through 5-14 with respect
Practical Ratio Detector to the applied voltages to the individual diodes dur-
The use of a battery for constant voltage output ing the constant frequency changing of the incom-
in the ratio detector for f -m receivers is not practi- ing f -m signal. Thus, the same phase shifting proc-
cal because, due to the nature of the incoming f -m ess is employed in the ratio detector circuit.
signals, we desire to have the output voltage con- Examining Fig. 5-26 once more, it is evident that
stant only at the average strength of the incoming since the diodes are connected in series aiding they
signal. Since the sound carriers of the different draw current in the same direction relative to R,
television stations are not all of the same strength which is also in series with them. Consequently,
using the convention for the flow of electrons from
AVC
.cathode to plate, the current I will follow the path
LAST
-FI indicated by the arrow, and the top part of resistor
TUBE
R will become negative with respect to its bottom
or grounded end. If the primary and secondary
tuned circuits are both resonant to the i.f. and if an
unmodulated i -f carrier signal (of the same fre-
quency) is injected into the circuit from the i -f
AUDIO
amplifier, the two capacitors C, and C4 both will be
OUTPUT
charged to the same voltage due to the symmetry
of the circuit.
Fig. 5-26.-A typical ratio detector circuit. A parallel RC Now if the i -f carrier were frequency modulated,
network has replaced the battery of Fig. 5-25 in the output
of the circuit. the voltages appearing across capacitors C, and C4
would vary according to the modulation of the i -f the f -m signal. Since the f -m signal is changing, or
carrier. This happens as follows : It was mentioned is being deviated, at an audio rate (that is, at the
that rectified current would flow in the direction rate of its modulating signal), the output voltage
shown in Fig. 5-26 and that the top portion of the across C4 will be varying at an audio rate. In this
resistor R would have a negative potential on it. type of circuit the voltage appearing across C, will
The values of the resistor R and capacitor C are be larger than that across C4 at frequencies below the
so chosen that they represent a long time -constant i.f., and above the i.f., the voltage across C4 will be
network. Usually the value of this time-constant larger than that across C,.
network can vary anywhere between one -tenth and The basic part of the ratio detector that removes
one -quarter second and still be effective to the de- any a.m. appearing in the input is the RC time
sired degree. Consequently, with a long time con- constant network of Fig. 5-26. It is the constant
stant, it will take the capacitor C quite some time to voltage across resistor R and capacitor C that plays
discharge fully through R. Therefore, the negative the primary role in the removal of a.m. Let us sup-
voltage at the top of resistor R will remain practi- pose that an a -m signal appears at the input of the
cally constant over the range of the lowest audio ratio detector and see what happens :
frequency desired to be reproduced in the output Any a -m signal will tend to increase the voltages
of the set. In other words, a time constant of one - across capacitors C, and C4. However, the voltage
tenth of a second corresponds to the period of a across the RC network cannot change rapidly
frequency of 10 cycles per second ; therefore, for enough to folî.ow the a.m., due to the nature of the
frequencies above 10 cycles per second, the dura- long time constant, and the a.m. therefore, cannot
tion of one cycle would be shorter than the time change the voltage across C, and C4. In other
constant, and so the voltage across the RC combi- words, the capacitor C charges or discharges so
nation will remain practically constant. (The higher slowly through R that the potential at the top of re-
the audio frequency, the shorter the duration of one sistor R (or the plate of diode D,) remains nearly
cycle.) constant, and any a.m. cannot change the voltage
across capacitor C in step with this a.m. Conse-
Since the voltage across R and C is constant, the
quently, sudden increases in amplitude of the f -m
sum of the voltages across C, and C4 must remain,
constant. However, if the carrier frequency falls carrier will have no effect in the output audio circuit,
because these sudden increases of amplitude cannot
below, or rises above, the i.f., the voltages appear-
ing across C, and C4 will differ in value according appear across either C, or C4 as a change in voltage.
to the degree of off -resonance condition of the i -f
signal. No matter what the difference between these AVC From Ratio Detectors
voltages, their sum always remains the same, but
their ratio will be varying at the rate of the devia- In the limiter discriminator arrangement, avc volt-
tion of the f -m signal, and it is this change in ratio age was available in the grid circuit of the limiter
which is detected. If the i -f signal is frequency due to grid rectification action. In the ratio detector
modulated, the i.f. will vary above and below its system, since limiters are not employed, the avc is
resonant frequency according to the degree of f.m. obtained from some other place. In the ratio detec-
This accordingly will vary the voltages appearing tor circuit of Fig. 5-26 it is noticed that the voltage
across C, and C4, but in a certain proportion deter- across resistor R serves as a means of obtaining avc
mined by the potential across the RC combination. voltage. Since the time constant network of RC is
Consequently, it can be said that the voltage across made to produce a constant output voltage at the
C4 varies, at an audio rate (due to the degree of average strength of the incoming signal, it is readily
f.m.) Therefore, the a -f output may conveniently evident that this output voltage will change in ac-
be taken off across C4, because one side is grounded. cordance with the varying average strength of the
and applied to the audio section of the set. The in- incoming signal.
stant when the incoming f -m i -f signal is at the What this means is that the capacitor C in con-
exact resonant frequency of the tuned circuits of junction with resistor R averages these signal
Fig. 5-26, the a -f voltage across C4 will be zero. strength changes appearing across R. The time con-
At instantaneous frequencies of the incoming f=m stant is not considered large in this instance as com-
i -f signal above and below the resonant frequency pared with the length of time required for changes
of the tuned circuits, the voltage across C4 will vary in average signal strength. However, the effect of
at a rate determined by the changing frequency of the time constant is sufficiently large to produce
L
effective removal of sudden changes in a.m., in- from this drawing how the signal voltage across the
cluding that brought about by the response charac- transformer primary appears across each diode.
teristics of the i -f stages. This is possible because Unless something is known about the construc-
the input signal does not change in strength as tion of the transformer comprising L, L1, L2, and
rapidly as these other amplitude variations, and the L,, it is somewhat difficult to understand this cir-
RC combination permits slow changes in voltage cuit. It should be known first that, due to inductive
in accordance with slow changes in the received coupling, voltages E, and E, appear across L, and
signal. Therefore, the negative voltage at the top L,, respectively, as in other types of detectors. The
part of resistor R serves as a source of avc voltage. important thing that should be noted is that coil L,
which has only a few turns, is a separate winding
usually closely wound around or near the bottom
Ratio Detector Modifications
or B plus side of coil L,. In this manner the coup-
The circuit of Fig. 5-26 was one of the first com- ling between these two coils is a maximum, and
mercial types of ratio detector circuits to be used. practically all of voltage E, appears across this coil
However, most of the ratio detector circuits that and the series capacitor C5 to ground. Since L is
are used in the sound channel of today's television untuned, the voltage induced into it from L, is 180°
receivers are modifications of the one just discuss- out of phase with E,. The voltages E, and E,, as
ed. The circuits of Figs. 5-25 and 5-26 were both just pointed out are in quadrature with El. There-
analyzed in order to illustrate the similarity to the fore, these two voltages are also in quadrature
discriminator circuit. with the voltage across L, as required.
The most difficult part in the design of the ratio Examining the circuit of Fig. 5-27 a little more
detector circuit is the transformer, namely that closely it will be seen that one end of coil L is tied
comprising coils L1, L,, and L, in Fig. 5-26. It is to the junction of the two 35 µµf capacitors, Clll
beyond the scope of this book to go into such de - and C112, and the other end to the center tap of
sigh work, but it should be remembered that such the transformer secondary. With this in mind we
factors as the proper coupling between windings, can readily see that coil L is common to both
the respective Q's of the coils both during the diode diodes. Consequently voltage E2 and that across
unloaded and loaded conditions, and the gain of the L are applied across diode D, and voltage E3 and that
last i -fi stage are important in this transformer de- across L are applied across diode D2. The vector dia-
sign. grams illustrating the operation of this circuit at
In the following are discussed a number of differ- resonance and off -resonance conditions are very
ent types of ratio detector circuits. The modified ratio much the same as those discussed with the discrim-
detector circuit that appears in Fig. 5-27 is from the inator circuit. As far as the voltage vectors acting
Andrea chassis VJ12. It may not be readily evident
--,ºº --------71
on the two diodes are concerned, the voltage vec-
TO
PLATE` T27 D
21.7 MC
6BÁ6 35±5(2) 15Kt5(2)
4T"I'F v aE +
L 1+ bl
E2 p
Ó I f 11 1
º ;i+
;15-
(
M I I ix)
_
AMPL 4 M CII3
L
= Y
+
,3pL3 ó= 11
I
I
DC
0
24 25V.
Fig. 5-27.-Modified
La 5 2 ratio detector circuit
0 C109 D2 AVC as used in the Andrea
television chassis
VJ12.
Courtesy Andrea Radio Corp.
Y23
R103
6AL5 33K±10
IOK±IO R107
RATIO DETECTOR
B+ AUDIO
OUTPUT C I14
=1000 MICA
tor representing that voltage drop cross L is 180° cuit of which is shown in Fig. 5-28. This circuit
out -of -phase from that normally indicated in the is radically different because coils L, L,, L2, and L3
previously discussed detector circuits. However, are not encased in one shield can as is usual. Part
nothing is radically changed because all this means number T7 -A contains only the primary (L,) and
is reversing the two horizontal axes of the previous tertiary (L) coils while T7 -B contains only the so-
vector diagrams and in this process the phase recalled secondary. Actually there is little or no in-
lationships between the voltages and current remain ductive coupling between these two units because
the same. they are separated at quite some distance on the
We know from previous ratio detector analysis -chassis. Therefore the voltages that appear accross
that at the center frequency of the i -f signal, the L, and L, are not due to transformer action as in
voltages across the balanced output capacitors are the previously discussed circuit. Coil L, is closely
equal. We also know that for frequencies different coupled to coil L and by transformer action volt-
age E, appears across L but of opposite phase, sim-
ilar to the circuit of Fig. 5-27. From the diagram
DI of Fig. 5-28 it may not be readily evident how volt-
ages E2 and E3 appear across their respective in-
ductances L2 and L3. In order to see this more
clearly that part of Fig. 5-28 to the right of and
including coil L has been redrawn in Fig. 5-29.
o- We know that when a signal is being received, a
voltage will be induced into coil L from coil L,
which is approximately of the same magnitude as
that across L, because of the close coupling. Since
coil L has one end tied to the center connection of
ww
eloo coils L2 and L3 and the other end grounded, the
Fig. 5-28.-The modified ratio detector circuit appearing voltage appearing across it also effectively appears
in the Belmont model 22A21.
Courtesy Belmont Radio Corp. across each diode. The completed circuit is through
the ground connection between the balanced capa-
citors C30 and C32.
from the center i.f. (i.e. off -resonance conditions) Let us analyze the action of the circuit on only
the voltages across the capacitors will no longer be one half cycle of the input signal at a time. When
equal but will vary in accordance with the varying the voltage across L is on the positive half cycle,
frequency of the f -m i -f signal. Consequently, the the plate of diode D2 and the cathode of diode D1
junction point between these two capacitors varies will both have a positive potential and the cathode
in potential with respect to the off -resonance condi- of diode D2 and plate of diode D, will both have a
tions and is a convenient place to obtain the audio respective negative potential. Thus diode D, will not
output signal. conduct because its plate is less positive than its
Looking at Fig. 5-27 once again, we see that the cathode. However, diode D2 will conduct because
two balanced output capacitors are Clll and C112. its plate is more positive than its cathode. Conse-
Across these two capacitors is the output load re- quently a current will flow in coil L3 and a voltage
sistance composed of two separate but equal re- drop E3 will appear across this coil. This voltage
sistors, namely R104 and R105, with their junction drop is due to conduction current flowing through it
point grounded. This means that when the incom- and not to induction as was usually the case. How-
ing f -m signal is equal to the resonant frequency 6AL5
of the transformer, the junction of capacitors C111 Di
and C112 is effectively at ground potential. By con- fE¿w
necting a suitable coupling network from the junc-
tion of C111 and C112 we have a means of obtain-
ing the audio output signal. In Fig. 5-27 this coup-
AUDIO
ling network is seen to consist of resistor R107 and OUTPUT
capacitor C114 with the audio voltage signal out-
put taken off the capacitor.
Another modified form of the ratio detector ap- Fig. 5-29.-The conduction current flowing in coil L3
will induce a voltage in coil L2. This induced voltage in
pears in Belmont model 22A21, the detector cir- turn causes an induced current to flow in L2.
ever, coil L2 and L, are usually so closely wound to- the a -m circuit to one of so-called f.m. was a proc-
gether on one form (it usually is single center -tap- ess of detuning the i -f transformers in the set.
ped coil) that there is effectively unity coupling be- This process of detuning was very important for
tween them. The conduction current flowing through the correct operation of the sound channel on f -m
L3 will thus set up magnetic lines of force which will signals. This process of detuning was done to such
link coil L2 and induce a voltage in coil L2 of the television receivers as the RCA models TRK-9
same magnitude as that across L,. Thus voltages and TRK-12 and the General Electric models HM -
E2 and E3 equal each other in magnitude. However, 171 and HM -185.
due to the method of connection between L2 and Detuning of the i -f transformers changed the
L3, the two equal voltages E2 and E3 are 180° out - response curve of the circuits in question. The de -
of -phase with each other. Therefore we have the tuning was such that either one or both of the sides
same phase conditions between the three voltages, of the response curve became quite linear. The
(that across L and E2 and E,) as in previous cir- more linear the sloping characteristic of the re-
cuits. sponse curve, the less the amount of distortion in
The induced voltage across E2 will then cause an the output. The amount of detuning was such that
induced current to flow through its circuit. Conse- the center frequency of the incoming f -m i -f sig-
quently on the positive half cycle we see that equal nal fell on the linear portion of the sloping charac-
but 180° out-of -phase voltages appear across sec- teristic of the response curve. (Either the slope at
ondary coil L2 and L3. On the negative half cycle the high or low frequency end of the response curve
of the signal across L, the potentials across each can be used. Of course the more linear the slope,
diode are interchanged from what they were on the the better the detection). The incoming f -m i -f sig-
positive half cycle. This means that diode D, will nal will be changing in frequency an equal amount
conduct first and coil L2 will have conductive cur- on either side of its center frequency, and the i -f
rent flowing through it and thus coil L, will have an transformers are detuned to the point where the
inductive current flowing through it. The voltages frequency swing of this signal will fall upon the
across L2 and L, are the same as on the positive linear slope of the response curve. By the f -m i -f
half cycle because each half cycle is of the same signal falling on the sloping characteristic of the i -f
amplitude. It is thus seen that during each cycle response curve, detection of the f-m signal is possi-
of input signal equal but 180° out -of -phase volt- ble. This type of detection is commonly known as
ages appear across L2 and L,. slope detection.
The phase relationships between the voltage across In order to show how an f -m wave can be detect-
L and that across L2 and L, are the same as in the ed by this method, let us refer to the response curve
circuits previously discussed. of Fig. 5-30. This curve approximates the selectivi-
ty curve of an i -f transformer with somewhat
broadband characteristics. In this drawing we as-
Slope Detection sume that the frequency represented by point A on
the left slope of the curve and the center frequency
In the first commercial television receivers the of the incoming f -m i -f signal are the same. The
sound channel employed a.m. However, when the curve on the bottom of the drawing is just a sine
FCC changed the sound television transmission wave depicting the frequency variations of the in-
from a.m. to f.m, the sound channel of the tele- put f -m signal. This curve is not an f -m wave.
vision receivers likewise had to be changed in order Since we know that an f -m wave varies in fre-
to receive f -m sound signals. The only commercial- quency we can illustrate this frequency variation by
ly used f -m detector network at the time of the a sine wave that is contant in amplitude. Thus the
changeover was the limiter -discriminator circuit. peak amplitude of the sine wave at the bottom of
To convert the a -m sound channel to one capable the drawing illustrates the peak deviation of the in-
of detecting f.m. would require many changes, espe- coming f -m i -f signal.
cially in the detector circuit. In order to avoid re- From this drawing it is readily noticed that the
wiring and adding and deleting components, the peak -to-peak frequency deviation of the f -m input
circuits already in the sound channel were uti'ized signal is well within the so-called linear range of
but were detuned so that detection of an f -m signal the left hand portion of the curve. Since the curve
was possible. is sloping, conversion of the frequency variations
In other words, the only essential change from of the f -m signal into one that also varies in ampli-
simplicity we showed the frequency variation as

I -F PEAR constant, but for speech and music it is anything
but constant. The output wave at the left hand side
of the drawing illustrates the amplitude change of
the f -m signal brought about by the slope of the
6
A
response curve. It is shown as being constant in
I -r amplitude because the incoming f -m signal is shown
SELECTIVITY
S S
CURVE
as being constant in frequency deviation over a
OUTPUT WAVE
INDICATING short period of time.
VARIATION IN
> AMPLITUDE OF The output signal from the i -f transformer net-
FM SIGNAL
work will still be an f -m signal but this f -m signal
will also be varying in amplitude as well as in f re-
quency. The conversion of the input signal is such
that the output f -m signal varies in amplitude in
FREQUENCY'
accordance with the ratio and strength of the audio
modulating signal.
SINE WAVE DEPICTING
FREQUENCY VARIATION The usual diode detector following the i -f stages
LIMITS OF INPUT FM
SIGNAL of a -m systems will not detect true f -m waves.
However, with the detuning method just described,
Fig. 5-30.-When an f -m i -f signal falls on the sloping the input signal to the diode detector will vary in
characteristic of the i -f response curve, detection of an f -m
signal is possible. amplitude as well as frequency. Consequently the
diode detector will respond to the amplitude varia-
tions and detection will be complete. The audio sig-
tude is brought about. The frequency variation of nal output from the detector will thus be varying
the f -m signal changes in accordance with the am- at the frequency of the audio modulation of the
plitude of the audio modulation. For the sake of f -m signal.
CHAPTER 6
THE VIDEO I -F AND DETECTOR SECTION

BY SEYMOUR D. USLAN
In the regular a -m and f -m receivers, the i -f stages to the high frequencies involved and also due to the
provide the set with the necessary selectivity and gain required polarity of the video signal on the picture
for proper reception. The selectivity of the i -f stages tube.
is primarily determined by the i -f transformers and
the gain of these stages by the i -f amplifiers. In a -in The Video I -F System
broadcasting, the bandwidth is very narrow compared
to that for f.m. In either case the tuned i -f circuits of The output signal from the mixer tube is a com-
the receivers are designed to pass only those fre- bined video and sound i -f signal. One of the functions
quencies within the bandwidth of the desired signal. of the i -f system is to deliver the sound modulat-
By bandwidth is meant those frequencies involved in ing frequencies at the input to the audio amplifiers
the transmitted sidebands. In a -m broadcasting (to and the video modulating frequencies at the input to
1500 kc) the bandwidth of the signal can be no the video amplifiers. The methods of obtaining this,
greater than 10 kc, or 5 kc on either side of the car- however, differ in many of the receivers.
rier. In f -in broadcasting the over-all bandwidth is For instance, some receivers -tap off the sound i -f
150 kc, or 75 kc on either side of the carrier fre- signal at the output of the mixer stage and apply it
quency. In the f-m sound channel of television, the directly to the f -m i -f amplifiers ; the video i -f signal
over-all bandwidth is 50 kc (see chapter 5). For the is also sent through its i -f system. In other receivers
video part of the television signal, as previously men- the combined i -f signals at the output of the mixer
tioned, the bandwidth far exceeds any of the other are first sent through one or two i -f amplifiers which
types of signals due to the very high frequencies in- amplify both signals before they are separated for
volved in the video signal. their respective channels. These tubes are sometimes
The video i -f system of television receivers has called video i -f amplifiers ; however, they are also
to perform the same function of selectivity and gain sound i -f amplifiers and help increase the gain of the
as in the other sets. However, due to the bandwidth sound i -f signal. Still other receivers send both i -f
of the signal involved, the design of the i -f circuits signals through a complete series of amplifiers, but the
greatly departs from that used in regular a -m re- reason in this case is not just to amplify both i -f sig-
ceivers and even differs appreciably from f -m re- nals but also to produce a new sound i -f. All these
ceivers. The video i -f amplifier must handle frequen- circuits will be discussed later on.
cies extending over 5 mc. Not only must the i -f am- The terminology of the video i-f stages is applied
plifiers and associated circuits have a flat response to all stages that have to pass the video signal, whether
over this range, but they must also reject interfering or not the sound i.f. is also involved. In other words,
signals close to the edges of the pass band. The coup- it is common practice to refer to those i -f stages not
ling circuits between the video i -f amplifiers are usu- used solely for the sound i.f.'s as video i -f stages. It
ally not as simple as the i -f transformers in the a -m is in these video .i -f stages that we are interested. Be-
and f -m receivers. Many different i -f circuits will be sides providing the desired gain and broad selectivity,
discussed in this chapter in conjunction with the the over-all flat response curve of the video i -f system
methods of obtaining the required bandwidth and in- must have a special sloping shape around the video
terfering signal rejection. i -f carrier part of the curve. In general the shape of
The detector stage of the video section of a tele- the correct over-all i -f response curve is quite critical.
vision receiver follows the i -f stages. Its function is The necessary broad bandwidth characteristic of the
the same as that of a detector employed in a -m re- video i -f stages is obtained primarily by two methods
ceivers and similar type circuits are used. The diode -overcoupled i -f stages or stagger-tuned i -f stages.
detector is a common form of detector circuit. Slight Both types are used today. Before we discuss them
modifications in circuit arrangement are required due we will study the bandwidth requirements.
92
THE VIDEO I -F AND DETECTOR SECTION 93
Bandwidth Requirements If television receivers are to be selective for all

In order to reproduce all the video frequencies, the video frequencies and nonselective to others, the
bandwidth of the video i -f circuits has to be quite shape of the curve of Fig. 6-1 should be adhered to
wide. With vestigal-sideband transmission and with as closely as possible. If this bandwidth is not strictly
a maximum of 4-mc video frequency for one sideband. adhered to, the low video and high video frequencies
the maximum video i -f bandwidth should be about will not be amplified equally. The low video frequen-
5.25 mc. This was indicated in the over-all trans- cies contribute to the background of the picture and to
mitted bandwidth picture of Fig. 2-4. The receiver the resolution of large objects. If the low frequencies
bandwidth for the video modulated signal was illus- are discriminated against, the background and large
trated in Fig. 2-6. Due to the use of vestigal-sideband objects will not appear clear on the picture tube. This
transmission, the lower video frequencies are em- will be noticed no matter what the size of the pattern
phasized, and to avoid this overemphasis, the selec- on the picture tube. Consequently, it is very impor-
tivity characteristic of the receiver for the video sig- tant that the low video frequencies be reproduced
nals should have a sloping characteristic such as illus- according to the curve of Fig. 6-1.
trated in Fig. 2-6. This sloping characteristic is in the If the bandwidth of the response curve of Fig. 6-1
neighborhood of the video carrier. The de -emphasis were decreased by reducing the high video frequency
of the lower video signals is brought about by the end of the band, not as much trouble would be en-
video i -f system. Due to the process of frequency con- countered as in the case of low video frequency dis-
version, the video receiver response characteristic is crimination.
reversed from that shown in Fig. 2-6. The high viideo frequencies are not too critical be-
The ideal over-all frequency response characteristic cause they add to the finer detail of the picture. A
of the video i -f stages is illustrated in Fig. 6-1. The loss in the finer detail of the reproduced picture will
sharp cutoff characteristic on the left-hand side of the not be noticed as readily as the loss due to low video
response is primarily due to the sound traps, which we frequencies. This is even more pronounced when the
will discuss later on. The right-hand side of this re- size of the screen used is small, because the smaller
sponse should essentially be a mirror image of the the picture, the more difficult it is to note the fine
left-hand side of the response shown in Fig. 2-6. Note details. For small picture tubes a high -frequency
the change in position of the video and sound car- bandwidth characteristic as wide as shown in Fig. 6-1
riers. In order to understand this change, let us ana- is not absolutely necessary for good picture repro-
lyze the system of frequency conversion. duction. There are television receivers where the
In the transmitted television signal, the sound car- bandwidth of the video i -f system is not as wide as
rier is 4.5 mc higher than the video carrier. That is
the reason that the sloping characteristic around the FIG. 6-1.- The ideal }
vioc°
video carrier of Fig. 2-6 is on the low -frequency side over-all frequency re- SO%
agglit
of the curve. The oscillator in a television receiver is sponse characteristic of the
video i -f stages : The sharp
tuned above the incoming television signal. Hence, cutoff characteristic on the
after frequency conversion, the center frequency com- left results from the sound
traps ; the right-hand side of the response should essentially
ponents of the produced i -f signals are reversed in fre- be a mirror image of the response shown in Fig. 2-6.
quency spectrum relationship as compared to their
carrier frequencies when transmitted. illustrated in Fig. 6-1, and the response at the high
For example on the first channel (44 to 50 mc), the video frequencies is decreased. Such receivers em-
picture carrier is 45.25 mc and the. sound carrier is ploy small size picture tubes and the loss of the high
4.5 mc higher at 49.75 mc. Assuming that the re- video frequencies is not easily noticeable upon the re-
ceiver oscillator is tuned to 71 mc, the i.f.'s produced produced picture.
will be 71 less 45.25 mc or 25.75 mc for the video The wider the bandwidth of the i -f stages the
signal, and 71 less 49.75 inc or 21.25 mc for the sound greater will be the gain required and, consequently,
carrier of the saine channel. Consequently we see that the more video i -f amplifiers needed. Thus the amount
due to conversion action, the new carrier frequencies of video gain, and hence the number of video ampli-
now termed the i.f.'s are such that the video i.f. is fiers needed, depends upon the bandwidth of the i -f
higher than the sound i.f. The frequency difference stages.
between them is still, however, equal to 4.5 mc. Thus In the following sections we will study the different
in Fig. 6-1 we see the reason for the shape of the methods used in television receivers in order to ob-
over-all i -f selectivity curve of the video i -f system. tain the broad bandwidth characteristic.
Overcoupled I -F Transformers Resistance Loading
The use of overcoupled itransformers to produce
-f In the video i -f circuits of both the GE model 802
the necessary wide band is quite common in many of and Farnsworth model GV-260, which are discussed
today's television receivers. For sharply selective cir- ater on, loading resistors are used across the sec-
cuits, the response curve is usually a single -peaked ondary of the video i -f transformers. These loading
curve as shown by curve A in Fig. 6-2, where the resistors reduce the over-all gain of the stage but at
peak is at the resonant frequency of the circuit. Such the sante time make for a flatter and somewhat broader
circuits have a low coefficient of coupling. When the response.
coefficient of coupling is increased, a point will he It is known that decreasing the Q of a circuit flat-
reached where any further increase will result in a tens out the frequency response of that circuit. Plac-
double -peaked curve. This point is called the point of ing a resistance across a tuned circuit will reduce the
critical coupling and curve B in Fig. 6-2 illustrates a O of that circuit. The Q of á circuit (or coil) is de-
critically coupled curve. Note that it has a somewhat fined as the ratio of inductive reactance to series re-
broader peak than curve A. When the coefficient of sistance. If the series resistance is increased or the
coupling is increased beyond its critical value, the inductive reactance decreased, the 0 decreases. Plac-
response curve will become much broader, with ing a resistor across a parallel LC tuned circuit will
double peaks appearing instead of a single peak. effectively increase the series resistance in the induc-
Curve C in Fig. 6-2 illustrates greater than critical tive or capacitive branch. Since the coil always con-
coupling. The dip in the center of the curve occurs at tains sonie inherent (series) resistance as compared
the resonant frequency of the circuit, and the higher to the negligible amount contained in the capacitor,
the coefficient of coupling, the deeper will be the dip
the resistance placed across the parallel LC circuit is
and the greater will be the distance between the peaks.
said to increase the series resistance of the inductive
Such curves as C in Fig. 6-2 are desired for broad-
band characteristics and are often acquired by means branch. The lower the value of the parallel resistance,
of overcoupling the i -f transformers. the greater will be the effective series resistance and
In some systems the coupling between the coils hence the lower the O.
(both tuned to the same resonant frequency) remains The parallel resistance, however, cannot be too low
constant and the circuit Q's are varied to effectively because the gain per stage decreases with a decrease in
result in a double -peaked broad -band curve. Another Q and a certain amount of gain is required for proper
method of obtaining a wide band curve is to detune operation of the video i-f system. The resonant fre-
both the primary and secondary of a coupled circuit, quency of `a parallel LC circuit also depends upon the
with the coupling remaining constant. Overcoupling series resistance in the circuit. By lowering the 0 of
the circuit is primarily used in video i -f circuits.
such a circuit by the insertion of a parallel resistor,
It had sonietinies been found that in order to ob-
the resonant frequency is also changed. This change,
tain the necessary bandwidth, the overcoupled circuit
however, depends upon the magnitude of the parallel
produced a double -peaked curve that was wide
enough, but the dip in the center of the curve was too resistance and the value of inductance in the circuit.
low. If the transformer were loaded down and the Q If the effective series resistance becomes appreciable,
reduced without appreciably changing the bandwidth the resonant frequency will change.
In the design of circuits with resistance loading,
Fic. 6-2.-Response curves for the manufacturer usually takes this into account, so
different coefficients of coupling:
ic A low coefficient produces the
single -peaked curve A. The
that with parallel resistance loading, the desired reso-
nant frequency is still attained. Whenever resistance
ié broader curve B has reached the
.« critical point beyond which any loading is used to broaden out the response charac-
fr increase in the coefficient pro-
duces a still broader but double - teristic of a tuned circuit where the resonant fre-
FREQUENCY
peaked curve C. quency is high, it is always advisable when the re-
and resonant frequency of the circuit, then a flatter sistor needs to be replaced that an exact duplicate be
response characteristic would result. In many re- substituted ; special emphasis should be placed on
ceivers this effect is obtained by loading down the trying to keep the leads and wiring similar to what
transformers with resistance. they were before.
Typical Overcoupled Circuits the Q of the secondary circuit and hence flattens and
Employing Resistance Loading broadens the response characteristic. Normally when
We will now investigate two typical overcoupled a circuit is overcoupled, two peaks of the sanie ampli-
video i -f circuits that employ resistance loading across tude will appear equally spaced on either side of the
the secondary of their i -f transformers to flatten the resonant frequency of the response carve. If the 0's
response characteristic further. We will not show of lwth the primary and secondary resonant trans-
the complete video i -f circuits of each set but only one former circuits are reduced by the same amount, the
coupling network from each, because the other video double-peaked response characteristic will be flattened
i -f stages are pretty much the same. out evenly.
Fig. 6-3 illustrates the video i -f transformer coup- However, if the Q of only one circuit (the sec-
ling network used between the second and third video ondary in the circuit of Fig. 6-3) is reduced, the
i -f amplifiers of the GE model 802. This circuit has
peaks of the overcoupled curve will not be reduced
been drawn in a simplified form for the sake of the
evenly. The degree of overcoupling, the amount of
discussion. The primary resonant circuit consists of
reduction in Q, and the change in the impedance that
inductance Lh and the parallel combination of capaci -
the secondary reflects into the primary due to the re-
V12 V13
duction in secondary Q, determines how much each
Z". V10t0 I-F AML.
AC7
3" VIDEOIFAMPL. peak will be reduced. The change in Q of the sec-
6AC7
ondary circuit clue to the resistance loading injects a
different reflected impedance into the primary circuit.
This new reflected impedance changes the Q of the
primary circuit, but this change of Q is not the same
as that of the secondary circuit.
Courtesy GE For most practical purposes, those television re-
FIG.6-3.-Simplified circuit diagram of the video i -f trans-
former network used between the second and third video i -f ceivers using overcoupled video i -f transformers with
amplifiers of the GE model 802. resistance damped secondaries are usually designed
so that there is very little change between the damp-
tances Cp and C,5. The secondary resonant circuit ing of the two peaks of the overcoupled response
consists of inductance Ls, the parallel combination of curve. Not all the overcoupled video i -f stages in any
capacitors Cs and C26 and resistor RIG. Capacitors one receiver produce exactly the same response curve.
C25 and C26 are a physical part of T-17, the third
They have to vary slightly in order to help produce an
video i -f transformer, and are variable. (The video over-all video i -f response that resembles curve C of
i -f transformers are often called pi.v i -f transformers
Fig. 6-2. Each video i -f stage contributes to the gain
as in this model under discussion. In fact, the word
and frequency response to produce the necessary re-
f'ix is used quite often in place of the word video.)
sponse curve for proper operation of the receiver. It
Capacitance Cr represents the total shunting capaci- should be remembered that resistance loading reduces
tances common to the primary circuit of the trans- the gain of a stage, and the video amplifiers must in-
former and essentially consists of the output capac- crease the gain to the necessary amount for correct
itance of the second video amplifier, the distributed operation of the receiver.
capacitance of coil Lr, and the stray wiring capaci- If all the video i -f circuits produced the same over -
tances in the vicinity of the primary circuit. In similar coupled response curve, the over-all i -f curve would
fashion, capacitance Cs represents the total shunting be of an undesirable nature. The curves are additive,
capacitances common to the secondary circuit of the resulting in a greater difference between the ampli-
transformer and essentially consists of the input tudes at the peaks and the trough ; thus the over-all
capacitance of the third video i -f amplifier, the dis- curve will be less flat than any one of the original
tributed capacitance of coil Ls, and the stray wiring curves. In order to avoid this, four video i -f stages are
capacitances in the vicinity of the secondary circuit. employed in the GE model 802 where three are over -
Coil Lp is closely coupled to coil Ls so that the coupled stages each resulting in a double -peaked curve
transformer circuit is effectively overcoupled and pro- and one is a single -peaked stage. The first, third, and
duces a broad response. The response is made still fourth video i -f transformers are overcoupled, double -
broader by the use of a 3900 -olmi loading resistor R16 tuned circuits with resistance loading in their sec-
across the secondary of the transformer. This reduces ondaries. The second i -f stage uses a single parallel
LC resonant circuit, effectively contributing a re- shaping the final resultant curve for the correct fre-
sponse curve that has only one peak. quency response. Curve D of Fig. 6-4 is the over-all
The curves of Fig. 6-4 will make this somewhat video i -f response curve of the receiver under discus-
clearer. These curves represent the appearance of the sion. The frequency markers on the curves indicate
video i -f signal of the GE model 802 at different points how the relative shape progresses from one video i -f
in the i -f system. Curve A represents only that con- stage to the other. The shaping of the left side of the
tributed by the last video i -f stage which is over - curves is clue to the 21.9 -mc sound i -f traps of the
coupled and resistance damped, resulting in the double - receiver.
peaked curve. Curve B is the result of the additive
responses of both the third and fourth (last) video 6AC7
2ND VIDEO I-F AMPL.
6AC7
i -f stages. The third i -f transformer is likewise over -
V5
3RD VIDEO I -F AMPL.
V6
coupled and, together with the fourth i -f stage, helps
73
produce a resultant curve that has a deeper trough 3RD VIDEO I -F
than the previous curve A. This means that curve B R24
will not allow those video frequencies that fall within 2200
011M3 -a- C
-T - 5
the center of the curve to be amplified as well as the
frequencies on either side of the trough. To obtain B*
equal amplification, the response curve should be flat.
This is accomplished by the use of the third i -f stage, Courtesy Farnsworth Tel. and Radio Corp.
FIG. 6-5.-A simplified diagram of the third video i -f stage
which is a single -tuned stage, essentially producing a of the Farnsworth model GV-260.
single peak in the region of the frequency where the
trough of curve B appears. When all the response In the video i -f section of the Farnsworth model
curves of the second, third, and fourth video i -f stages GV-260 (first production), overcoupled i -f trans-
are added together, they result in a response that has formers are used with resistance loading across the
a fairly flat top characteristic as indicated by curve C secondaries. A simplified drawing of the third video
in Fig. 6-4. The response of the second video i -f stage, i -f stage of this receiver appears in Fig. 6-5. It will
when added to the response of the third and fourth be noticed that in this circuit there is no separate
video i -f stages (curve B) , increases the gain at the capacitor and that the transformer is not tunable. The
mid -frequency region, thereby resulting in curve C. circuit is so designed that coil Lp is parallel resonant
The first video i -f stage is also overcoupled and re- with capacitance Cp which represents the total shunt-
sistance damped, but the amount of overcoupling and ing capacitances common to the transformer primary.
damping is so designed that it chiefly contributes to Likewise coil Ls is in parallel resonance with Cs the
total shunting capacitances common to the transformer
23MC 26.4 MC
secondary. Coils Lp and Ls are overcoupled so that a
3 MC 26.4 MC
double -peaked response curve results and the 2200 -
ohm resistance R24 across the secondary reduces the
Q of the circuit and broadens the response.
As in the GE model previously discussed, the total
shunting capacitances represented by Cp and Cs in-
clude the stray wiring capacitance and the distributed
capacitances of the coils, in addition to the tube capaci-
23MC 23.6304C 23MC 23.83MC
tances. Since this transformer is untunable, these
26.4 MC shunting capacitances are very important in deter-
244 t4C
mining the resonant frequency of the transformer cir-
O sox, cuit. When leaving the factory, the resonant fre-
quency of these transformer circuits is already set.
2L1 MC 21t "IC This means that anything that tends to change the
Courtesy GE
values of Cp or Cs will tend to change the resonant
FIG. 6-4.-Curves representing the video i -f signal of the frequency of the transformer primary and/or sec-
GE model 802 at different points in the i-f system. The double - ondary.
peaked curves A and B represent, respectively, the response
of the fourth i -f stage and the additive signal of the third and Consequently if this section of the receiver is ever
fourth stages. The single -peaked curve C represents the addi- serviced, the person performing the servicing should
tion of the second, third, and fourth stages, and D the over-all
video i -f response curve of the receiver. be careful to replace the exact component parts that
THE VIDEO I-F AND DETECTOR SECTION 97
are removed (if any) and also to make sure that the Stagger -Tuned I -F Transformers
wiring is in the sanie position as before servicing. In The other method of producing the necessary video
replacing parts, not only must they be of the same
i -f response makes use of stagger -tuned video i -f
value and of the same size and make, if possible, but
stages. In stagger tuning, the cascaded i -f stages are
the length of the connecting leads and the position of
each tuned to a different resonant frequency. In the
the part should be the same as before. These precau-
stagger tuning employed in the video i -f stages of
tions are necessary because only a minute change in today's television receivers, each i -f stage produces a
the shunting capacitance will change the resonant
single -peaked response curve, so that when all the in-
frequency of the circuit.
dividual response curves are combined, the resultant
If the video i -f amplifier tubes are changed, there
video i -f curve will be that required for the proper
is the likely possibility that the input and/or output
operation of the receiver.
capacitances of the new tube will be different from the The over-all video i -f response that results from
one removed, in which case the over-all shunting
stagger -tuned systems should approach the curve of
capacitance will change and hence the resonant fre-
Fig. 6-1 as closely as possible for faithful reproduc-
quency of the circuit will be different. The manufac-
tion of all the video frequencies. It is possible to pro-
SOUND 50%
V1DEp
CARRIER
26.25 MC
FIG. 6-6. -
The over-all
video i -f response curve of
duce a resultant curve similar to that of Fig. 6-1 by
adding together a number of curves, each single -
CARRIER
21.75 MC
L- the overcoupled i -f trans- peaked at a different frequency.
former stages of the Farns-
worth model GV-260. There are various methods by which the video i -f
Courtesy Farnsworth Tel. and Radio Corp. stages may be tuned. It is beyond the scope of this
book to discuss every method in use today. Conse-
turer states that the absence of tuning adjustments in quently the circuits analyzed in this section are by no
the third video i -f transformer (Fig. 6-5) and also in means the only available types but are chosen as typi-
the second i -f transformer makes the gaih of the stage cal examples.
higher. It also makes the over-all video i -f alignment In the RCA model 630TS four video i -f amplifiers
simpler. Although all the video i -f stages in this re- and five i -f coupling stages are employed. The coup-
ceiver are overcoupled, the amount of overcoupling ling between stages is not by transformer action but
and resistance loading in each circuit is so designed rather by simple inductive -capacitive coupling. Fig.
that the required bandwidth is obtained without dis- 6-7 is a simplified schematic of the video i -f section of
crimination to any video frequencies. The over-all the RCA model 630TS. Only those parts of the cir-
video i -f response curve of these overcoupled stages cuit and the connections of interest now have been
is illustrated in Fig. 6-6. drawn, with the traps being omitted.
Although not shown in the video i -f circuits illus- In this circuit, the coils (with the exception of the
trated in this section, sound i -f traps are also incorpo- primary of T105) are all slug tuned, so that they reso-
rated in these circuits to prevent interference from nate with the distributed capacitance common to each
the sound i -f signals. These traps also help shape the coil and the 270 µµf coupling capacitors (C18, C116,
over-all video i -f response curve. C123, C128, or C134). The video i -f signal voltage de-
6AG5 GAGS 6AG5 GAGS

1ST PIX I -F AMPL. 2ND PIX I -F AMPL 3RD PIX I -F AMPL. 4TH PIX I -F AMPL.
C 11 C123 C128 C13 4
270 270 270 270
135V 135V 135 V 99 V TO
CATHODE
OF GALS
TO PLATES PRIMARY VIDEO
OF 6J6 OF DETECTOR
MIXER T103
C135
PRIMARY 1SOOT
OF T2 1185
PRIMARY
T-
R14ÿ
1000
396 - OF
T10 5 8+ -
_ R133 0131
C9 50 82
15001
_ B+
Courtesy RCA
FIG. 6-7.-Simplified schematic diagram of the video i -f section of RCA model 630TS in which the coupling between stages
is simple inductive -capacitive coupling.
veloped across each coil is coupled to the following The response curve for this first video i -f stage is
video amplifiers through these coupling capacitors. shown in Fig. 6-9. It is peaked at a frequency of
In order to understand how these video i -f stages con- 21.8 inc.
tribute to the final resultant curve, we will study the In this receiver, the oscillator is so designed that it
circuit from the input to the first video i -f amplifier. will track above the incoming r -f signal hence a dif-
;
The signal output from the 6J6 mixer tube consists ference frequency of 25.75 mc results for the video
of a number of different signals. Besides the differ- i -f carrier and 21.25 me for the sound i -f carrier.
Since the mixing process produces an i -f video car-

Fic. 6-8.-In this simple network rier of 25.75 mc, how can the 21.8 -mc resonant cir-
representing the primary of T2, C is cuit be selective enough to pass this frequency to the
the over-all capacitance in the circuit ;
R115 is the grid -leak resistor and to- video amplifiers? Although the primary of T2 is
gether with R14 the load resistor, re- peaked to 21.8 mc. its response curve is broad enough
duces the Q and makes the response
less sharp than it otherwise would be. to pass signals up to 25.25 mc. and somewhat beyond
this frequency, to the following stages for amplifica-
ence frequency between the local oscillator and in- tion. This is a very important fact about stagger -
coming r -f signal, there is also their sum, the indi- tuned circuits and must be understood in order to
vidual r -f and oscillator signals themselves, and nu-
merous harmonics of all these signals. The primary
of T2 is made selective to the difference frequency be-
tween the incoming r -f and oscillator signal. The re-
sponse of this circuit is single peaked, but broad
enough to pass the complete band of the produced i -f 21.8 MC 25.3 MC
T2 T103
signal. However, Clue to the single-peaked effect, not
all signals are equally amplified. The primary of T2 is >
loaded down by the 1000-ohm plate load resistor R14 4
-J
and 10,000 -ohm grid leak resistor R115. This coil cir-
cuit, and some of the others, can be represented by
the simple network of Fig. 6-8. Capacitor C repre-
sents the over-all capacitance in the circuit (including FREQUENCY
the total shunting capacitances, the distributed capaci-

tance of the coil, and the effect of the coupling capaci- Courtesy RCA
tors) an(1 it is essentially in resonance with the coil. FIG. 6-10.-The response curve of T103 is shown to be es-
sentially the same as that of T2. T103 is also loaded down by
The other components are the same as in Fig. 6-7. plate and grid -leak resistors, but is tuned to a peak resonant
The loading down of this resonant circuit by the re- frequency of 25.3 mc.
sistors reduces the Q and makes the response less
sharp than it would be without this loading effect. comprehend fully how the shaping of the final over-all
response curve is possible.
The signal input to the first video 6AG5 i -f ampli-
fier (also called a pix i -f amplifier) contains all the
frequencies necessary for proper reception. However,
these frequencies are not of equal amplitude due to the
21.8 MC response characteristic of the primary of T2. This
T2 first 6AG5 tube does not change the relative ampli-
W
tude of the signal input.
1
The primary of T103 is loaded down by plate and
4-J
grid -leak resistors of the same value as before, and so
K
the shape of its response curve is essentially the same
as that of Fig. 6-9. However, this coil is tuned to a
FREQUENCY peak resonant frequency of 25.3 mc. This response
curve and that of T2 are drawn in Fig. 6-10. The
Courtesy RCA amplitude of this curve is essentially determined by
Fic. 6-9.-Response curve for the first video i -f siage which
is peaked at a frequency of 21.8 mc. the gain of the second video i -f amplifier. The output
of the second video i -f amplifier is impressed across circuit is primarily brought about by the 2700-ohm
the next tuned circuit. plate load resistor R127 of the third video i -f amplifier.
The primary of T104 is tuned to a frequency of The plate load resistor is effectively in parallel with
22.3 mc and it, also, has a single -peaked response L183 through the coupling capacitor C128. However,
curve. The shape of this response curve differs from due to the absence of a grid -leak resistor and to the
value of R127, the Q of the coil is reduced to a greater
degree than in the other circuits. This is a necessity
in order to obtain the correct over-all video i -f re-
sponse curve. The resonant frequency of coil L183 is
21.8 MC 2 5. 3 MC 25.2 mc, only 0.1 me lower than that for the primary
12 22.3MC 1103 of T103. The fourth video i -f amplifier contributes to
W
the gain of the response curve for coil L183. However
the gain cannot be too high or too low, but should be
4
of the proper amount to insure the correct over-all
video i -f response. The plate and screen voltages on
this fourth video i -f tube are so adjusted that the tube
FREQUENCY contributes the necessary amount of gain for the
proper over-all i -f response. For comparison purposes,
Courtesy RCA the response of the L183 circuit is illustrated in Fig.
Fu.. 6-11.-The primary of T104 has the saine single - 6-12 together with the response curves of the three
peaked response of T2 and T103; however, the shape of the
curve differs from the others because it is loaded down still previous stages.
further. Its much smaller grid -leak resistor R125 is only 4700
ohms compared with the 10.000 -ohm resistors used with the The output from the last or fourth video i -f ampli-
other primaries. fier is impressed across L185, the last tunable video
i -f stage. Coil L185 is not actually in the plate circuit
the previous curves because the circuit is still further
of the fourth video amplifier as it would seem in Fig.
loaded clown. Even though the value of the plate -load 6-7 but instead is in the cathode circuit of the video
resistance has not changed, the 4700 -ohm grid -leak
detector which follows. The, resonant frequency of
resistor R125 is much smaller than the 10,000 -ohm
this last stage is 23.4 mc, and it is loaded down by two
resistor used in the previous instances. Since the grid-
components, the 5600 -ohm plate load resistor R134
leak resistor is effectively in shunt with the tunable
and the following detector circuit. Both components
coils, then the greater the resistance, the higher the
are essentially in parallel with coil L185; the detector
effective coil Q, and the lower the resistance, the
circuit represents a very low resistive load to this
lower the coil Q. This means that the 4700 -ohm grid -
coil, so that in conjunction with R134, it appreciably
leak resistor 8125 effectively lowers the Q of the cir-
reduces the effective Q of the coil. This results in
cuit, reducing the gain of this i -f stage but at the same
quite a broad -band single -peaked response curve.
time making this stage much broader than the others.
The peak of the curve is not very well defined as in
The amount of reduction in gain and increase in band-
width is readily noticeable when the response of this
network is compared to that of the previous two. All
three curves are illustrated in Fig. 6-11, where it is
seen that the 22.3 -mc peak response curve is reduced
in amplitude but broader than the other two. The 21. 8 MC 25.3 MC
third 6AG5 video amplifier contributes to the gain of 22.3 MC T103
this stage. The output of this amplifier is effectively 25.2
impressed across the L183 network as seen in L183
Fig. 6-7.
The resonant circuits that follow the third video i -f
amplifier are arranged somewhat differently from the
previous ones. Coil L183 is placed in the grid circuit FREQUENCY
of the fourth 6AG5 video i -f amplifier and thus no
grid -leak resistor is used in this circuit. This coil pro- Courtesy RCA
vides the d -c return path for the grid in this fourth FIG. 6-12.-The response curve of coil L183 of the fourth
video i -f amplifier can here be compared with the curves of
video i -f amplifier. The reduction in coil Q of this the three previous stages.
100 RIDER'S -- "HOW IT WORKS"
the other curves, and the decrease in coil Q also re- circuit of the mixer tube, first video i -f amplifier, and
duces the gain of the L185 stage quite a lot. However, second video amplifier respectively, and thus d.c.
the reduction in gain and the broad response obtained (plate current) flows through each of these coils. The
is so designed that this last video i -f response curve last two video i -f coils (L183 and L185) are not in
contributes the final touch to the required shaping of any B -plus circuit and thus d.c. does not flow through
the over-all video i -f response. The response curve of these coils.
the 23.4 -mc circuit is illustrated in Fig. 6-13 in con-
junction with the other four response curves. Note The Need for Sound and Video I -F Traps
how broad this latter curve is compared to the others.
In this chapter we have mentioned the importance
All the five video i -f response curves illustrated in
of the sound and video i -f traps in shaping the final
Fig. 6-13 contribute their necessary share of band- over-all video i -f response curve. These traps, al-
width and gain so that the over-all video i -f response though helping shape the over-all video i -f response
that results from the multiplication of the curves ap- curve, are essentially inserted to prevent interference
proaches the ideal curve of Fig. 6-1. The over-all from undesired sound and video signals. The possi-
bility of interference from different signals will be
considered before we analyze the different traps as
used in present-day receivers.
If any sound i -f signal passes through the video i -f
21.8 MC 25.3 MC stages, it will probably be detected and thus cause
\
T2 22.3 MC 1103
interference in the reproduced picture. One type of
1104
sound interference may be recognized by a series of
23.4MC 25.2 horizontal bars across the screen of the picture tube.
L183
1185
. It may not be readily evident how a sound f -m i -f
signal can be detected by the video detector, which is
usually a simple diode detector and as such responds
FREQUENCY only to a -m changes in its input signal. This detection
is possible because of the position of the interfering
Courtesy RCA sound f -m signals on the video i -f response curve.
Fm. 6-I3.-The response curve for L185, the last tunable
i -f stage, is shown in conjunction with the foregoing four re-
These signals lie on the outskirts of the response
sponse curves. Note the broadness of this last curve compared curve, where the shape of the curve has a definite
with the others.
slope. Due to the shape of the slope, the f -m signal
curve for the entire stagger -tuned system is illus- will be changed to one that also varies in amplitude ;
trated in Fig. 6-14, this curve is fairly flat and this amplitude variation is essentially ät the rate of
covers the necessary bandwidth. The 25.75 -mc mark the audio signal modulating the f -m wave. Since the
indicates the position of the video i -f carrier. The new signal is varying in amplitude, it can thus be de-
broadness of the response of the last video i -f stage tected. This process of changing the f -m signal to one
is the most helpful in making the over-all curve fairly that also varies in amplitude is known as slope de-
flat. 'Without this last video i -f response, the center of tection and was discussed in greater detail in chap-
the over-all response curve would have quite a clip in ter 5.
its center. The sharpness of the left-hand side of the Besides sound i -f interference, there is also the pos-
sibility that the video signals adjacent to the one
FIG. 6 -14. ----The over- which is desired may be passed and detected, causing
all curve for the entire a blurred and incoherent pattern on the screen of the
stagger -tuned system is
25.75 MC
shown to be fairly flat picture tube. These signals can easily cause inter-
and capable of covering ference, because without the use of special traps it is
the necessary band-
NOTE: THE EFFECTS OF THE width. The 25.75 mark quite difficult to obtain the desired shape of the over-
TRAPS ONTHE INDIVIDUAL indicates the position of
RESPONSE CURVES ARE NOT SHOWN. the video i -f carrier. all video i -f response curve, particularly at the ex-
treme ends of this curve. The shaping of the sloping
curve in Fig. 6-14 is clue to the traps employed in the right-hand side of the curve of Fig. 6-1 with the video
video i -f stages. i -f carrier frequency approximately in the center of
It is interesting to note that in the first three video the slope is accomplished chiefly by the video i -f trans-
i -f stages the coils are in the B -plus lead of the plate formers and not by the traps. However, the sharp
response on the left-hand side of the curve of Fig. 6-1 sound, point D, and video, point A, i -f carriers are
is due to the traps employed. If no traps were em- calculated on the basis of the frequencies assumed for
ployed, the shape of the response curve might be points B and C.
broad enough (although not at the same level) to fall If traps were not employed, the response curve of
outside its desired bandwidth. This means that un- the over-all video i -f network of Fig. 6-15 would not
desired signals falling inside this oversize response take on the shape indicated by the solid line part of
can easily be passed through the video i -f circuits. the curve but would be different at the ends, probably
Since each television channel is only 6 me wide, taking on a shape at these extreme points as indicated
the sound i -f carrier of the next lower channel is only by the dashed lines. Upon a quick inspection of the
1.5 mc away from the video i -f carrier to which the
curve, with regard to the shaping due to dashed lines,
set is tuned, and only 0.25 mc away from the limits of it is readily seen that points A, B, and D all fall within
the channel. In similar fashion, the adjacent video i -f the dashed portion of this curve. All the signals rep-
carrier of the next higher channel is 1.5 mc away from resented by these three points are undesired, but they
the sound i -f carrier to which the receiver is tuned would he detected because all fall within the region of
and 1.25 mc away from the upper end of the channel response curve. If the amplitude of the dashed lines
limit. If any of these adjacent -channel signals pass could be considerably reduced, little or no interference
through the r -f and mixer stages, they will be con- would result. This reduction is accomplished by the
verted to i -f signals similar to the desired signal. The use of special traps which absorb or attenuate these
frequency separation between the carriers will remain undesired signals. The shape of the response curve
the same even after they are converted into i -f signals. with traps employed at frequencies represented by
points A, B, and D will be such that, insteadof follow-
Remember that before frequency conversion the
ing the dashed line, it will take on the shape of the
video r -f carrier is 4.5 mc lower than the sound r -f
solid line as indicated in Fig. 6-15. The amplitude of
carrier. After frequency conversion the video i -f car-
the solid line curve at points A, B, and C is seen to
rier is 4.5 mc higher than the sound i -f carrier, due
to the heterodyne oscillator in the receiver tracking 6MC
CHANNELLIMITS
above the incoming signal. Thus the video i -f carrier
will fall on the high end of the over-all i -f response, VIDE0(1-F)
CARRIER
and the sound i -f carrier will fall on the low side of
the over-all i -f response. In a similar fashion, the ad-
jacent video i -f carrier of the next higher channel will ADJACENT VIDEO I
L25
MC
I I
I
r ---------
ADJACENT
I -F CARRIER I I SOUND -F CARRIERI
appear on the low side of the over-all i -f response, ILSM I 4.5 MC 'LSMC¡
19.5 21 MC 25.5 27 MC
near the position of the sound i -f carrier.
In order to visualize how undesired signals can pass FIG. 6-15.-A normal over-all video i -f response curve with
through, let us refer to Fig. 6-15. The sólid line curve an ideal flat top is shown by the solid -line part of this curve.
If traps were not used, the curve would probably end as indi-
represents a normal over-all video i -f response curve cated by the dashed lines, allowing undesired signals at A, B,
after proper alignment of the video i -f stages. The and D to pass. Reduced in amplitude through the use of traps,
their interference becomes negligible.
shape of this curve may differ somewhat for various
receivers due to the type of coupling between stages be negligible, compared with the over-all height of
and the number of different traps employed. For the the curve, and the signals at these points will there-
sake of simplicity, it is shown as having an ideal flat fore not be able to pass through the video i -f net-
top. Points A through D represent different frequen- works.
cies. Point C represents the video i -f carrier to which
the receiver is tuned. Point B represents the accom-
Practical I -F Traps
panying sound i -f carrier to which the receiver is also
tuned. If the video i -f carrier, point C, is at 25.5 mc, A trap is usually a resonant circuit sharply tuned
the sound i -f carrier point B, being 4.5 mc away, is at the frequency to be suppressed. The resonant trap
at a frequency of 21 mc. Point A represents the adja- effectively either absorbs energy or causes a loss in
cent video i -f carrier and point D the adjacent sound gain at its resonant frequency. Either a series or a
i -f carrier. The frequency separation between these i -f parallel resonant circuit can be used as a trap. In the
carriers is indicated on the drawing. The long solid following analysis of different video and sound i -f
vertical lines indicate the 6 -mc limits of channel band- traps, it should be remembered that a series resonant
width. The frequency indications of the adjacent circuit offers a minimum impedance (which is purely
resistive) at resonance and a parallel resonant circuit sound i -f carrier, being 4.5 mc less, is equal to 21.9
offers a maximum impedance (also purely resistive) mc. Thus the four traps to suppress the accompanying
at resonance. sound i -f are tuned to 21.9 mc. Since the adjacent
Not all the television receivers employ the same sound i -f carrier is 6 mc higher (the distance of a
number of i -f traps. The number of traps employed channel width) than the accompanying sound i -f car-
(for different frequencies) is determined by how wide rier, or 1.5 mc away from the video i -f carrier, it is at
the desired video i -f pass band mast be. In those re- a frequency of 27.9 mc. Thus the single adjacent
ceivers that use a small picture tube, where loss of sound i -f trap is tuned to 27.9 mc.
high video frequencies is not too critical, the pass band This adjacent sound i -f trap appears in the grid
is often small with consequent reduction in bandwidth circuit of the first video i -f amplifier (tube V3) and
at the high video frequency end. A narrower band- is a series tuned circuit. It consists of capacitor C126
width may make unnecessary the use of a trap for in series with coil L32 and variable capacitor C127
the adjacent video i -f carrier or even, perhaps, for the which helps tune the trap to the 27.9 -mc resonant fre-
sound i -f carrier of the channel in question. In most quency. This trap, as all traps, is sharply tuned ; that
cases, however, two or three traps are employed and is, it is highly selective to its resonant frequency. A
in special cases four traps may be employed. The response curve of a trap circuit will show a very sharp
number of traps as referred to in the discussion so far peak with steep sides. This is necessary to insure that
means those traps that have different resonant fre- only the resonant frequency is suppressed and not
quencies and not to the actual numerical amount of other frequencies that might fall within its response
traps used. Some receivers may employ four indi- curve if the curve were broad. For such highly selec-
vidual traps but two may be resonant to the adjacent tive circuits, high Q coils are needed. This adjacent
sound i -f and the other two to the accompanying sound trap offers a low impedance (due to series reso-
sound i -f. Two traps of the same frequency are used nance) to the 27.9 -mc frequency and, since it is effec-
to give better insurance of rejection at the frequencies tively wired across the first video i -f amplifier, it will
to which they are tuned. offer a ready path to ground for 27.9 -mc signals and
In the GE model 802 five individual traps are em- they will thus not be amplified by tube V3.
ployed, four are tuned to the accompanying sound i -f The other series resonant circuit, which is tuned to
signal and one to the adjacent sound i -f signal. A 21.9 mc, the accompanying sound i.f., is situated in
schematic of the video i -f section of this receiver, the cathode circuit of the video detector V 15A. This
including all these traps, is shown in Fig. 6-16. Two trap consists of capacitor C36 in series with the paral-
of the traps are series resonant circuits and the other lel combination of L36 and variable capacitor C34. This
three are parallel tuned circuits. The video i -f carrier circuit, although perhaps this is not readily evident, is
of this set is equal to 26.4 Inc, and the accompanying essentially in series resonance at 21.9 mc. The com-
V3 V12 V13 VISA

IST VIDEO I F 250 VIDEO LF 350 VIDEO IF
C21 VIDEO DETECTOR
6AC 100 6AC 7 64C7 1.2-6946
TO SECONDARY
T 16 TIT T IS
OF FIRST
256 PIO IF 315D PI% 1 F TO PIA 1-F
VIDEO I -F
TRANSFORMER
C16
R17
5600
C19
IOOO
R7
3900
C50 . .J1 56
RIS 21.9MC
3900 TO AP
R6
tlOR
C126 55
] 1000
N\Az TOCÖÑTRÓLT
L32 }.-.__ 27 9 MC
TRY
0127
TO FIRST Bt
C6S
SOUND IF AMPL.
13300
Courtesy GE
FIG. 6-16.-Schematic diagram of the video i -f section of the GE model 802 showing the five individual traps employed in this
receiver.
bination of the two capacitors C36 and C34 in con- sound i -f amplifier and the low side is grounded. At
junction with L36 makes the circuit somewhat sharper 21.9 me this trap offers a maximum impedance and
in its tuning. Its impedance is a minimum and offers thus. besides causing a loss in the video i -f response
a ready path to ground to the 21.9 -mc accompanying at this frequency. a maximum voltage at a frequency
sound i -f signal that may enter the detector stage. of 21.9 mc appears across this circuit and is applied to
Thus we see that this trap, as well as the one dis- the first sound i -f amplifier. The selectivity of this
cussed previously, reduces the gain of the over-all latter 21.9 -mc trap circuit is such that it will pass the
video i -f response at the frequencies to which the sound i -f signal with its complete bandwidth so that
traps are tuned. proper reproduction of the audio modulating fre-
The 21.9 -mc trap used in conjunction with the third quencies will be obtained.
video i -f transformer T17 works on the principle of Four traps are employed in the RCA model 630TS.
absorption. This trap is a parallel resonant circuit Two traps are employed for the accompanying sound
which consists of coil La in parallel with the fixed i -f signal, one trap for the adjacent sound i -f signal,
capacitor C37 and variable capacitor C29. The latter and the fourth trap for the adjacent video i -f signal.
capacitor is used to tune the trap to the proper reso- Since the desired video i -f carrier is 25.75 mc, the
nant frequency of 21.9 mc. Coil La is a tertiary wind- accompanying sound i -f carrier, being 4.5 me less,
ing on transformer T17 and together with its parallel is equal to 21.25 mc. The adjacent sound i -f carrier
capacitors presents a highly selective tuned circuit at is therefore 27.25 nic, and the adjacent video i -f car-
its tuned 21.9 mc. Due to its high selectivity and coup- rier is 19.75 uic. The video i -f circuit for the RCA
ling to T17 it absorbs energy at 21.9 mc from the video 630TS receiver is illustrated in Fig. 6-17. The sec-
i -f signal coupled through 7'17. The gain of the video ondary circuit of the converter transformer T2 is a
i -f response curve is thus effectively decreased at the parallel resonant circuit tuned to 21.25 mc, the ac-
tuned frequency of this trap, that -of the accompanying companying sound i -f signal. This resonant circuit
sound i.f. The absorbed energy is dissipated in the acts as an absorption trap to the accompanying sound
inherent resistance of the coil. i -f signal, similar to the L,,C22 trap circuit of Fig.
Another such absorption trap for 21.9 mc is used 6-16. This tram coil is tapped at point B, and some of
in conjunction with the video i -f transformer T16. the sound i -f signal is taken from the circuit and fed
This trap consists of coil Lb in parallel with trimmer to the control grid of the 6BA6 first sound i -f am-
capacitor C20. It is also tuned to the accompanying pli fier.
sound i.f., namely 21.9 mc, and functions in the same The parallel tuned secondary circuits of trans-
manner as the other absorption trap. Below this 21.9 - formers T103 and T104 both act as sharply tuned
mc trap is another 21.9 -mc trap consisting of coil Le absorption traps to the signal to which they are timed.
in parallel with trimmer capacitor C22. The high side The secondary of transformer T103 is tuned to 27.25
Of the 'coil is connected to the control grid of the first mc and acts as a trap to the adjacent sound i -f signal.
61e5cite
26i1GSlF 6 RG5 6;11.5
_,
11
1ST PPIXX IF 3 4TH PI X 1.1 PIX.2Nt0 DET.
TO CONTROL GRID Vllo 25.3 vili 22 V112. IF 7Ó zñ V113 23.4
Nc
V 114-A
OF 6BA6 FIRST _ I1 1
R127 0134
/y,16 270 (+,16 270 ,.2700 2 70
Tf cle
j
L183
TO PLATES CONvetrlR T104
OF 6J6 TRANS
270 r103 r ---, C135
MIXER
TUBE
rIAa p -1c16
- --f- â ßRN 2(130
1500
1500
wIC1¡ N LIES
43
C113 -'-(11_761
C127
1500
C132
75
JLC '1500 "ti= .50o,
il NC
,,8
C
21.251-1C 4-
,The 0114
TRA. ,f
Tat..
4
1565 R135
C12o- 111123 .R1ze R13
1000
}
P R119
39Ç 500 1000 :1000 ISO3 e2
1000
1/w
tpus 8114
150
-C115 R120 C121

VN/V.
R124 R125 C129
W
R'30 C133
lì+
ISO ISO 150
10K 1500 10 lc 1500 4700 z 1500 11500
BIAS WAS BIAS
Courtesy RCA
Fic. 6 -17. -The video i -f circuit for the RCA 630TS receiver. Note that the secondary circuit of the converter transformer
T2 is a parallel resonant circuit acting as a sound absorption trap similar to the LeC22 trap circuit of Fig. 6-16.
The secondary of transformer T104 is tuned to 19.75 10) five individual traps are employed, with two tuned.
mc and acts as a trap to the adjacent video i -f signal. to the same frequency, and the other three to different
The fóurth trap in this receiver is located in the frequencies. This means that we have to deal with
cathode circuit of the fourth video i -f amplifier. This four different trap frequencies. In the previous cir-
trap is the secondary circuit of transformer T105 and cuits, the maximum number of trap frequencies em-
is parallel resonant to 21.25 mc. the accompanying ployed was three, that of the accompanying sound i -f
sound carrier i -f signal. carrier, the adjacent sound i -f carrier, and the adja-
The primary of T105 is in series with C131 and cent video i -f carrier. In this receiver all these three
forms a series resonant circuit at 23.4 mc, the fre- types are employed, plus an additional one.
quency to which coil L185 is tuned. Since a series Let us examine the video i -f circuit of this receiver
resonant circuit offers a minimum amount of im- which is illustrated in Fig. 6-18. All the video i -f
pedance, the cathode of tube V113 is effectively coupling networks essentially consist of a tuned cir-
grounded at the frequency to which the primary circuit in the plate of one tube and another tuned circuit
cuit of T105 is resonant. This permits the tube to in the control grid of the next. These networks are
operate at a gain at this frequency (23.4 mc). The coupled together by small value capacitors. Across
parallel resonant frequency of the 21.9 -mc trap has a the grid of each i -f amplifier and the plate of the video
high impedance and thus reflects a high impedance detector appears a series resonant trap. Each trap
into the primary of T105 at this frequency. This essentially consists of a capacitor in series with a
means that the cáthode circuit has appreciable im- parallel combination of a coil and capacitor. Although
pedance, and an r -f voltage drop appears across this not readily evident from the circuit, the components
circuit at the frequency of 21.9 mc. This r -f voltage of these traps are so chosen that they are in series
drop causes degeneration in the tube to the extent resonance at the frequency indicated next to each
where- the tube does not contribute any appreciable trap in Fig. 6-18. (These traps are similar to the
gain at this 21.9 -mc frequency. series resonant trap consisting of C36 -L36 -C34 in
In the Stromberg -Carlson Model TV -10L (series the cathode circuit of the video detector of the GE
EE.{ SC TOM, .,.. ML ,,..
M PLATE
Or 6AL5 6AG5
COMVERTER 64G5 n ]AOl
tJiwu N f.)
b.
Courtesy StrombergCarlson Co.

FIG. 6-18.-The Stromberg -Carlson model TV1OL video
i -f circuit illustrated employs four different trap frequen-
cies for its five individual traps.
TO PLATE
Of VIOCO
OETEC.TOR
model 802 in Fig. 6-16.) Since they are in series at the frequency to which the trap is tuned. The nec-
resonance, they offer a minimum impedance to ground essary rejection of interference frequencies is thus
to the frequency for which each one is tuned. Traps produced.
A and C in Fig. 6-18 are resonant to 27.6 mc, which
is the adjacent sound i -f carrier. Even though the ad- Contrast Control
jacent video i -f signal has a carrier frequency of 20.1
mc, the 20.8-inc trap (D) probably has a response The contrast control, sometimes called the picture
broad enough to attenuate the adjacent video i -f control, of a television receiver determines the strength
carrier also. Trap E is resonant to the accompanying of the video signal input to the picture tube. The mag-
sound i -f carrier of 21.6 mc. Trap B is tuned to 29.6 nitude of the video signal in conjunction with the
mc, the 20.8 -mc trap (D) probably has a response picture tube and potentials thereon controls the con-
in the channel of the adjacent sound i -f signal. This trast between the light and dark areas in the repro-
trap will attenuate possible interference from this duced picture. The contrast control, usually a poten-
video i -f signal. tiometer, is manually operated. This control is usually
In the Consolidated Television model 2315, all the inserted either in the video amplifier section preceding
traps employed are situated in the cathode circuits of the video detector or in the video i -f section of the
the video i -f amplifiers. They function differently from receiver. Its use in the video amplifier section is dis-
any of the traps discussed so far. A circuit diagram cussed in the following chapter. In this section we
of the video i -f section of this receiver appears in will study some of the applications of the contrast
Fig. 6-19. The traps consist of a coil and capacitor in control in the video i -f section.
parallel and the frequency to which it is tuned The contrast control when used in the video i-f
is indicated next to each trap. Although no trap is section (as well as the video amplifier section) con-
tuned exactly to the adjacent or accompanying sound trols the amount of bias on most of the video i -f am-
i -f carrier or to the adjacent video i -f carrier, they are plifiers. By controlling the bias, the amount of i-f
tuned to frequencies that will help produce the neces- gain can be varied to whatever degree is desired,
sary over-all video i -f response. They also provide for likewise varying the magnitude of the video signal
the attenuation of signals that might cause interfer- input to the grid of the picture tube. There are many
ence. They all work on the following principle. contrast or gain -control arrangements used in tele-
The traps, since they are parallel tuned, offer a vision receivers. To discuss all the different methods
maximum impedance to those frequencies for which of controlling the gain of the video signal is beyond
they are resonant. This means that when the plate r -f the scope of this book ; however, we will analyze a
current flows through the cathode circuit, a maximum few of the types used in the video i -f section of pres-
r -f voltage drop will exist across these traps at their ent-day television receivers.
resonant frequency. This causes degeneration in each In many television receivers, a d -c potential exists
tube at the resonant frequency of the trap connected across the contrast control, with the positive side of
to it. This means that the tube offers negligible gain the control grounded. The simple circuit of Fig. 6-20
V-3 V-4 V-5 V-6

6AC7 6AC7 6ÁC7 6A C7
G14 -1000m1 CIS-IOOmmf
.:L
O
ep
_
R 31 R 32
M/.Aw Mnv4
200 K 200K 390K
VIO(O T -F SIGNAL
INPUT FROM MIXER) TO VIOEO
B1
OETECTOR
ALC
Courtesy Consolidated Television Corp.

Fm. 6-19.-In this Consolidated Television model 2315, all the traps employed are situated in the cathode circuits of the
video i -f amplifiers.
illustrates this type of connection. The number of tion of the Du Mont RA -102 receiver is illustrated
video i -f tubes to which the negative potential is ap- in Fig. 6-21.
plied is determined by the design of the individual From this diagram we see that the variable arm
receiver. If the contrast control is advanced to the of the 25,000 -ohm contrast control R6 is connected to
point where too much bias is applied to the video i -f the control grid circuit of the first and second 6AU6
tubes, their gain will decrease and the video signal video i -f amplifiers. Variation of this control changes
the bias on these tubes and hence their gain. Thus the
FIG. 6-20.-This type of signal input to the following video i -f amplifiers, video
TO NEGATIVE contrast control, in which detector, and video amplifiers is effectively controlled
D -C SOURCE TO GRID a d -c potential exists
CONTRAST
OF VIDEO across the contrast control by R6.
CONTROL
I -F AMPL.
and its positive side is
grounded, is common in
In the GE model 802, the contrast control is in the
television receivers. grid circuit of the first and second video i -f amplifier.
This control, unlike the previous one, receives its
input to the picture tube will decrease. This will cause negative voltage from the grid circuit of another tube
the contrast between the light and dark areas to be in the receiver. the 6AS7G clamping tube. The circuit
less distinguishable by effectively decreasing the in- arrangement of the contrast control in conjunction
tensity of the lighter areas on the screen of the pic- with the damping tube is shown ill Fig. 6-22. Accord-
ture. 1f the contrast control decreases the bias on the ing to the manufacturer's specifications the voltage
video, i -f tubes too much, their gain will increase, measured between grid to ground of the 6AS7 tube
thereby causing the video signal input to the picture is -6.5 volts.
tube to increase. Too great an increase in the magni- Since the values of R120 and R121 are very small
tude of the video signal will produce a greater con- compared to the addition of resistance R90, R115,
trast between the light and Clark areas. This effec- and R122, they cause only a negligible drop in the
tively increases the intensity of the lighter areas on -6.5 voltage. Therefore we can consider the voltage
the screen of the picture tube to a point where the from point A, the junction of R 120 and R90 to
intermediate grey tones of the picture are lost. The ground, as being approximately -6.5 volts.
resulting picture will be glaring and incoherent. Ill parallel with this circuit from point A to ground
The primary requisite for the operation of the con- is a resistance network consisting of the 10-megolun
trast control is a cl -c potential across it. This bias is resistor R134 and the 0.5-megohm contrast control
often obtained from a negative d -c source in the power R108A. (The impedance of this network is too high
supply. This is clone in the Du Mont model RA -102 to affect appreciably the operation of the damping
where the contrast control is part of a voltage dividing tube). Resistor R134 and potentiometer R108A act
network within the 40 -volt d -c supply and receives its as a voltage divider network to the -6.5 volts. The
necessary voltage from this supply. The contrast con- voltage appearing across the 0.5-megohm resistance
trol is connected to the control grids of the first two of R 108.1 is computed as follows :
video i -f amplifier tubes and varies the gain of these

two stages this in turn varies the video signal input X 6.5 volts = 0.31 volt approximately
;
10 + .5
to the picture tube. A schematic diagram of this sec - This 0.31 volt is the complete voltage available
4^ VIDEO LE. 99
1 TO GRID
CI4 CIRCUIT OF
001"1 3EdVI1:1EOI.F.
C 0 C16
FROM Dó d LS
(- OOIµf
PLATE WHITE
CIRCUITF) C tO 1.10
MISER
OF
T"' 5%
150
uE r
5%
C6 )1
Oa "1 C IS
0 .01"1
CS
.001"0 Bf
0
1'
-COMTRAST CONTROL
TO NEGATIVE
1
=
;, 2.1
W
Ms
E
1/xr
711
D -C SUPPLY
Courtesy Allen B. Du Mont Labs., Inc.

FIG. 6-21.-The potential across the contrast control is often obtained from a negative d -c source in the power supply as
d -c
in this Du Mont RA -102 receiver in which the contrast control is within the 40 -volt d -c supply.
across the contrast control, and this is the maximum put volume of the receiver. To compensate for these
voltage that can be applied to the first and second changes in output, most radio receivers employ auto-
video i -f amplifier grids. This means that the addi- matic volume control (avc ) circuits. In the usual type
tional bias applied to these grids will be anywhere of avc circuit in radio receivers the received signal is
from 0 to 0.31 negative volt. This may seem like a rectified at a certain point in the receiver and applied
very small amount of bias to control the gain of these as a negative cl -c signal (avc signal) to the grids of
stages. However if we know the characteristics of the certain tubes, thereby changing their bias and varying
tubes employed as video i -f amplifiers, we will be able their gain. If the input signal increases, the bias on
vz3 the tube will automatically increase in proportion to
R194 6Á57G TO HOR.
DEFL.
the increase in input signal, thus keeping the output
DAMPING TUBE COILS
of the receiver at a constant level. In like manner, if
R121
220c the signal input decreases, the isias will decrease and
CONTRAST
CONTROL the gain of the tubes will increase and thus tend to
R10CA
500K keep the output of the set level.
C50C
In television reception the strength of the signal
TO GRID 4C input to the television receiver can likewise change,
CIRCUITS OF 6AC7
1ST. E 2ND causing undesired variations in the video signal input
VIDEO I.F'S.
(SEE FIG. G-16) to the picture tube. In many television receivers the
After GE video i -f signal is rectified and applied as a negative
Foc. 6-22.-In the GE model 802 the contrast control is in cl -c signal, usually to the grid circuits of two or more
the grid circuit of the first and second video i -f amplifier and of the stages preceding the detector. This signal varies
receives its negative voltage from the grid circuit of the
6AS7G damping tube. the bias and therefore the gain of the stages in much
the same way avc does in radio receivers. Although
to understand how the small contrast voltage can vary at times this variation is referred to as avc in the video
their gain. The 6AC7 tubes that are employed as channel of television receivers, it is usually known as
video i -f amplifiers are high transconductance. tubes automatic gain control (agc), or sometimes as auto-
in which the transconductance, or g,5, of the tube matic level control (alc). There are various different
varies appreciably with a small change in bias. The means of obtaining this automatic control voltage, but
fixed bias (cathode bias) on the first 6AC7 video i -f diode rectification of the video i -f signal is most often
tube is -2.1 volts and on the second video i -f tube it used.
is -2.0 volts. The plate and screen potentials on A simple arrangement of an agc circuit is shown
these 6AC7 tubes are such that at -2
volts the g,,, in Fig. 6-23. The video i -f signal is transformer
is about 9000 micromhos and at -2.5 volts the g,,, is coupled to the plate of the diode. Since the cathode
about 7000 micromhos. This range of gm over the -2 is grounded, current will flow through the tube only
to -2.5 volts small bias change can be considered as
being linear. Since the negative voltage across the
contrast control is 0.31 volt, the bias on the first i -f VIDEO a
tube can be macle to vary from -2.1 volts to -2.4 I -F
SIGNAL
volts and on the second i -f tube from -2.0 volts to INPUT
-2.31 volts. Thus, even though the -0.31 volt

across the contrast control is small, it does cause
variations in the bias on the tubes that in turn cause Foc. 6-23.-Due to the direction of the current flow in this
appreciable changes in the g,,, of the tubes. simple agc circuit (indicated by arrow), the potential on R2
will he negative with respect to ground, presenting a ready
Since the gain of a pentode amplifier is directly point for tapping of agc.
proportional to the transconductance of the tube, it is
readily seen how this small voltage across the con- on the positive half cycle of the input signal. The
trast control, R108A in Fig. 6-22, can vary the gain electron current flow will be from cathode to plate,
of the video i -f section and hence the magnitude of the through the plate -load resistor R2 to ground, then
video signal input to the picture tube. back to the cathode of the tube to complete the d -c
path. Due to the direction of this current flow, indi-
Automatic Gain Control (agc) cated by the arrow in the drawing, the potential on
If the strength of the signal input to a radio re- resistor R2 will be negative at its plate connection
ceiver changes, this will produce a change in the out- with respect to its ground connection. This negative
108 RIDER'S -"HOW IT WORKS"
part of the load resistor presents a ready point for tial on its plate is more positive than that on the
tapping off agc. However, the current flowing through cathode. Thus for the circuit of Fig. 6-24, the tube
the tube varies in accordance with the envelope of the will conduct only when the potential on the plate ex-
modulated input signal ; to convert this output signal ceeds 1.36 volts. This means that not until the video
as closely as possible to a d -c signal and to prevent i -f signal input at the plate of the diode (which is
any i -f voltage from getting to the automatic -gain- coupled through capacitor C62) has a peak signal
controlled tubes, a filter network is employed. This strength equal to 1.36 volts will the tube conduct.
filter is usually a resistance -capacitance network as (Remember that the value of 1.36 volts is dependent
shown by the R,C, circuit in Fig. 6-23 and tends to upon the values of R115 and R116 being exactly as
level out the varying output signal. listed, but since each has a 10 -percent tolerance,
In many cases no agc signal is desired on the grids the potential on the cathode may be somewhat differ-
of the tubes to be controlled until the input signal to ent from that calculated. However, the difference will
the receiver reaches a certain value. This is to make not be great.)
sure that onweak signal inputs there will be no agc The tube will conduct only when the positive half
signal available and the tubes will thus offer maxi- cycle of video i -f signal input exceeds a peak value of
mum gain at low signal inputs. Such a circuit when 1.36 volts. When the tube starts conducting, a d -c
used in radio receivers is called delayed ave. When voltage drop will exist across the plate -load resistor
used in the video section of television receivers, it is R46 with the plate side of the resistor becoming nega-
commonly known as delayed agc because there is a tive with respect to its grounded side, as shown in
delay before an agc signal is available. Fig. 6-24. The signal across R46, after passing
In the Belmont model 22A21 a 6AL5 duo -diode through a resistance -capacitance filter network con-
tube is used, one diode section of which is used for sisting of R45, C60, and C59, is applied as a negative
video detection and the other diode for delayed agc. d -c signal to the grid circuits of the first video i -f
The circuit for this agc arrangement is shown in Fig. amplifier and the r -f amplifier of the receiver.
6-24. To make sure that the diode will not conduct The stronger the video i -f signal input to the diode,
on weak signal inputs, a positive potential of approxi- the greater will be the current flow and, hence, the
mately 1.36 volts is put on its cathode. This potential greater the voltage drop across R46. This means that
is obtained from the 300 -volt B supply through the the bias on the first video i -f tube and the r -f amplifier
voltage divider network consisting of the 220,000 - tube will increase, thus reducing their gain. With a
ohm resistor R115 and the 1000-ohm resistor R116 strong signal input, a high (negative) agc signal re-
C62 TO GRID
sults, which reduces the gain of the controlled tubes
T13 6AL5 R45 CIRCUITS CF in such a manner that the input video signal to the
VIDEO C59 1ST VIDEO
O. I Ff,R-F
AMPLIFIER
picture tube is effectively kept at a constant level.
S16NAL
INPUT 1500
A1MF TC60
TO CATHODE R115
1_ Video I -F Amplifier Tubes
OF VIDEO 300 VOLTS
DETECTOR 220,000 The video i -f amplifier section contributes most of
1.36 VOLTS
1RÓÓÓ the gain of the video channel of the television receiver.
Most of the video i -f amplifier tubes employed have a
Courtesy Belmont Radio Corp. high transconductance in order to contribute the nec-
FIG. 6 -24. ---This arrangement for a delayed agc is used in
the Belmont model 22A21. One section of the 6AL5 duo -diode essary gain. A large g,,, is a fundamental requirement
tube is used for video detection and the other for the delayed for high gain from pentode amplifiers because the gain
agc.
is directly proportional to the g,,, of the tube. Some of
Since the complete 300 volts exists across the series the different tubes employed as i -f amplifiers are the
combination of R115 and R116, then that portion of 6AC7, 6AU6, 6AH6, and the 6AG5. The transcon-
voltage existing across R116 alone is as follows :
cluctances of these tubes is quite high, their values
R116 ranging from. about 5000 to 10,000 micromhos.
x 300 volts
R115 + R116 Besides having high transconductance values, those
tubes that are generally used as video i -f amplifiers
300 = = 1.36 volts
221000 x 221,000 have sharp cutoff characteristics. The use of sharp
Thus we see that a positive potential of 1.36 volts cutoff tubes decreases the distortion of the system
exists on the cathode of the tube. A diode functions in because their plate current-grid voltage characteristic
such a manner that it conducts only when the poten- curves are more nearly linear at the bottom of the
curve than those of remote cutoff tubes. If the i -f am- this type of circuit is essentially the same as that em-
plifier is not a sharp cutoff tube, the plate current- ployed in a -m receivers, the design of the circuit is
grid voltage characteristic curve will have appreciable very different. In the design of video detector circuits
curvature at its bottom. If the input signal is such that such factors as the required polarity of the video sig-
it operates over the lower portion of this curve, dis- nal on the picture tube, the loading on the detector,
tortion will be present in the output of the amplifier. and the frequencies involved must be taken into ac-
When high transconductance tubes with sharp cutoff count. All these factors will be considered in the fol-
characteristics are used, a small change in bias pro- lowing sections.
duces a large change in plate current.
In some cases when agc or contrast control volt- Phasing of the Picture Signal
ages are used, it is desired that the tubes have a re- It was mentioned that the required polarity of the
mote cutoff characteristic. This gives the control volt- video signal input to the picture tube is a determining
age a wide latitude of variation in changing the plate factor in the design of the detector circuit. In other
current of the tube. Some sharp cutoff tubes exhibit words, the video signal input to the picture tube has
remote cutoff characteristics when a screen-grid re- to be correctly phased for the tube to reproduce the
sistor is used. By the use of the proper resistor, the picture as taken at the studio. If we were to examine
d -c potential on the screen grid can be made to vary the schematics of a number of television receivers, we
with the bias on the control grid in such a manner would note some very interesting details in the video
that the tube will exhibit remote cutoff characteristics. channel following the last video i -f amplifier section.
Many tube manuals list a typical value of screen re- Following this last video i -f amplifier is the usual
sistor to use with sharp cutoff tubes to obtain remote video diode detector, and following that the video am-
cutoff characteristics. plifiers. In most cases the video signal is fed into the
control grid of the picture tube, although in some
VIDEO DETECTOR instances the signal may be fed to the cathode of the
The tube following the last video i -f amplifier is picture tube. In either case the effective signal on the
usually the video detector. The video detector circuits grid of the picture tube should be a negative video
of television receivers are very similar in 'operation to signal.
those employed in a -nt radio receivers, because the In those receivers where the video signal is fed
video i -f signal is an a -m wave and essentially re- directly into the control grid of the picture tube from
quires the same basic type of detector circuit as any the plate circuit of the last video amplifier and where
a -m wave. (This detector circuit is often called the an odd number of video amplifiers is employed, the
second detector when the mixer or converter tube video i -f signal input to the diode detector will be
used is called the first detector. In this chapter, how- coupled to the plate of the diode. When an even num-
ever, only the term detector will be used and not sec- ber of video amplifiers is employed, the video i -f signal
ond detector.) input to the detector will be coupled to the cathode
In an a -nt receiver the detector rectifies the incom- of the diode. When the video signal is fed to the
ing i -f signal, and after passing through special i -f cathode of the picture tube, the diode detector con-
filters, the modulating audio signal corresponding to nections just discussed are reversed when the same
the amplitude variations of the i -f carrier is recov- number of video amplifiers are used. These different
ered. This audio signal is then applied to the necessary connections are necessary because a negative video
audio amplifiers for reproduction in the output of the signal is essential at the grid of the picture tube to
speaker. In a similar manner, the video detector in a insure proper reproduction of the picture.
television receiver rectifies the video i -f signal and, In the reproduction of an audio signal, the phase
after proper filtering, the modulating video signal (in- of the signal input to the speaker is not important
cluding the sync and blanking signals) corresponding because faithful reproduction bf the audio signal does
to the amplitude variations in the video i -f carrier is not depend upon the polarity of the audio signal input
recovered. After the proper separation is made, the to the speaker. This means that the ear is not sensi-
picture (true video) signals and sync signals are ap- tive to phase changes in the audio signal. However,
plied to their respective circuits for correct picture a different situation exists in television regarding the
reproduction. polarity of the video signal input to the picture tube,
The video detector generally employed is of the because the human eye is more sensitive than the ear.
diode rectifier type. Although the basic operation of If the polarity is reversed, the reproduced picture will
be a negative of that at the studio. In other words, the picture tube. When the polarity on the control
the black parts of the picture will be reproduced white grid of the picture tube is reversed, the positive peaks
and the white parts reproduced black. of the video signal, which represent the black part of
In chapter we stated that negative transmission
1 the transmitted picture, would make the grid less
is used for television. This means that in the video negative and the reproduced image would be brighter.
a -m signal the positive peaks of the video modulating That is, the black part of the transmitted picture
signal will represent the black portions of the signal would effectively be reproduced white. In like manner
and the negative peaks the white parts of the picture.
A video modulated signal is shown in Fig. 6-25 to-
gether with the modulating signal. For the sake of
discussion the video signal is shown as a sine wave.
In the video modulating signal, Fig. 6-25 (A), the
positive peaks are the black regions of the signal and
(A) GRID FED
the negative peaks the white regions. In the modu- PICTURE TUBE
COMPOSITE BLACK
lated signal, Fig. 6-25 (B) the peak represents the
, VIDEO WHITE
black regions and the troughs represent the white SIGNAL
INPUT
region. The sync pulse and blanking signals are also
shown superimposed upon the carrier signal. When
the modulated signal is applied to the video detector,
(B) CATHODE FED
it will be rectified and the undesired r -f (carrier) will
be properly filtered out. The signal that remains after
this process is the modulating component, which con- 6-26.-For correct operation of the picture tube the
black parts of the video signal should drive the grid more
sists of the video signal and the superimposed sync negative and the white parts, less negative. Depending upon
and blanking signals. This composite signal is then the polarity of the signal, it may be fed to either the grid (A)
or the cathode (B) of the tube.
feel through the video amplifier system. After ampli-
fication and then d -c reinsertion, the composite video the negative peaks of the same signal, which represent
signal is ready to be feci into the picture tube. the white parts of the transmitted signal, would be
When the control grid of the picture tube has its reproduced black.
bias effectively decreased (that is, made less nega- Besicles the composite video signal reproducing a
tive), a stronger beans of electrons will flow ill the negative picture as just described, the sync and blank-
tube, causing a brighter image upon the screen of ing pulses of the signal would result in bright retraces
across the screen. In some cases, perhaps, the sync
POSITIVE PEAKS
(BLACK) and blanking signals might be strong enough to drive
the grid positive and possibly cause the cathode of the
picture tube to burn out.
NEGATIVE PEAKS For correct operation of the picture tube, the black
(wH ITE)
parts of the video signal should drive the grid more
VIDEO MODULATING SIGNAL negative and the white parts less negative, as illus-
SYNC.
(A) trated in Fig. 6-26. In this manner the brighter parts
PULSE will cause more electrons to flow from the cathode of
BLANKING PEAKS TROUGHS
SIGNAL (BLACK) (WHITE) the picture tube and the blacker parts less electrons.
I
IIIIJIlI I hIg
I I1'tf1
II1'', , III't1III1' In Fig. 6-26 (A) the polarity of the video signal is
such that it lies below the reference line. Thus the
IIIIII
11 I '11
III VIII III VIII black parts of the signal are more negative and will
increase the bias on the grid and therefore reproduce
CARRIER I PEAKS TROUGHS
(BLACK) (WHITE) black images on the screen. The white parts of the
signal are less negative and will reduce the bias on
MODULATED SIGNAL the grid and, therefore, cause more electrons to flow,
(8) producing white images on the screen.
Ftc. 6-25.-How negative transmission is used for television
is here illustrated by representing the video signal as a sine If the video signal is fed into the cathode of the
wave. In (A) the positive peaks are the black regions of the picture tube, the phasing of the signal must be oppo-
picture and the negative peaks are the white regions. In (B)
the peaks represent the black regions and the troughs the site to that when feci into the grid. This is illustrated
white regions. in Fig. 6-26 (B) where the video signal is seen to lie
THE VIDEO i -F AND DETECTOR SECTION 111
above the reference level. The black parts of the signal as indicated in Fig. 6-27 (A). With this detector
make the cathode more positive with respect to the arrangement, an odd number of video amplifiers is
grid, which means that the grid becomes more nega- needed to produce the correct polarity of signal input
tive with respect to the cathode, thus decreasing the to the grid of the picture tube ; and an even number
number of electrons hitting the screen. The white of video amplifiers is required when the video signal
parts of the signal make the cathode less positive with is fed to the cathode of the picture tube.
respect to the grid, thus effectively decreasing the bias On the other hand when the video i -f signal input
on the grid and increasing the number of electrons to the diode detector is to the cathode of the diode,
hitting the screen. Consequently, feeding the video as shown in Fig. 6-27 (B) , the polarity of the output
signal to the cathode with the proper degree of polar- signal is reversed from that in Fig. 6-27 (A). The
ity, as shown in Fig. 6-26 (B), will result in a correct use of even and odd numbers of video amplifiers
picture on the screen. From this analysis we can should be reversed from that of part (A) to give the
readily see the necessity for the correct polarity of the correct polarity of signal input to the cathode or grid
video signal when fed either to the grid or cathode of of the picture tube. This is indicated in Fig. 6-27 (B).
the picture tube. In parts (A) and (B) of Fig. 6-26, Usually either one or two video amplifiers are em-
the level of the blanking signal is such that it will ployed.
drive the picture tube beyond cutoff.
It was stated previously that the video i -f signal Detector and Amplifier Action
input to the detector may be connected to either the It may not be readily evident how and why the
plate or cathode of the diode detector, depending upon phasings of the video signal, as indicated in Fig.. 6-27
the number of video amplifiers employed and the type (A) and (B) , come about. The phasing of the video
of connection to the picture tube. In order to visualize signal depends upon the action of the diode rectifier
the various possible types of circuit arrangements in and upon the inherent phase inversion qualities of
the video section when thé video signal output is fed amplifying vacuum tubes. The video signal input to
to the picture tube from the plate circuit of the last the diode detector is a modulated wave as shown in
video amplifier, let us refer to the simplified circuits Fig. 6-25 (B), assuming a sine wave video modulat-
of Fig. 6-27. If the video i -f signal is fed into the plate ing signal. This input signal is also indicated in Fig.
of the diode detector, then after rectification and 6-28 and is fed into the plate of a diode rectifier in
proper filtering, the modulating component that is fed (A) and into the cathode in (B). That part of the
into the following video amplifiers will lie above the video modulated signal that lies above the reference
zero reference line, or in the so-called positive region, line is considered to be in the positive or plus (+)
ODD NO. TO GRID CIRCUIT

OF VIDEO OF PICTURE TUBE
AMPLIFIERS
VIDEO I -F
SIGNAL INPUT
(A) EVEN NO. 7417\../-N./`lN ro CATHODE CIRCUIT

OF VIDEO
OF PICTURE TUBE
AMPLIFIERS
VIDEO
DETECTOR
EVEN NO. N-1,1 To GRID CIRCUIT

OFVIDEO
OF PLCTURE TUBE
AMPLIFIERS
VIDEO I -F
SIGNAL INPUT
(B) ODD NO.
OF VIDEO
ro CATHODE CIRCUIT
BOF PICTURE TUBE
AMPLIFIERS
FIG. 6-27.-These various circuit arrangements are possible in the video section when the video signal output is fed to the
picture tube from the plate circuit of the last video amplifier. The video modulating signal is assumed to he a sine wave.
112 RIDER'S --- "HOW IT WORKS"
region and that below the reference line in the nega- but it is of opposite phase to that of Fig. 6-28 (A).
tive or minus (-)
region. The two different types of detector arrangements of
The diode will conduct only when the plate of the Fig. 6-28 provide either a positive or negative video
tube is effectively more positive than its cathode. signal output.
Thus when the modulated i -f signal is fed to the plate Amplifying vacuum tubes have the inherent quality
of the diode Fig. 6-28 (A), only that part of the sig- of phase inversion between the input and output sig-
nal above the reference line will cause the diode to nals. If a signal voltage of zero degree phase is the
conduct. The negative part of the signal will make the
plate negative with respect to the cathode and thus the
diode will not conduct for this part of the input signal.
The output signal from the diode detector in Fig.
VIDEO
6-28 (A) is taken from the cathode circuit of the SIGNAL
diode and is the positive half of the input signal. This OUTPUT
output signal contains both the video signal and the
(A)
VIDEO
high -frequency i -f component, and the latter signal t
VIDEO SIGNAL AMPLIFIER
is then removed by a filter network. The output of INPUT
this filter network is the video modulating signal and
it is then fed to- the grid circuit of the first video
amplifier. 1
When the video modulated i -f signal is fed to the
cathode of the diode detector, only the negative part (B)
of. signal will cause the diode to conduct. Without any
signal applied to the diode, the plate ánd cathode are
essentially at ground potential. When the potential on
the cathode is made negative, the diode will conduct Fio. 6-29.-Because signals undergo a phase reversal of
180° between the input and output circuits of an amplifier
because the plate is effectively at a higher potential tube, the video signals shown in Fig. 6-28 (A) and (B), when
than the cathode. The output signal from the diode applied to the video amplifier grid, are reversed in phase in
the amplifier output, as shown in (A) and (B) above.
arrangement of Fig. 6-28 (B) is taken from the plate
circuit of the tube and contains only the negative part input signal to the grid of a vacuum tube, the output
of the input signal. This output signal contains the signal voltage will be shifted in phase by 180°. Thus
video signal and the high -frequency i -f component, if the video signal output from the detector circuit of
and then the latter signal is removed by a filter net- Fig. 6-28 (A) is fed to the grid circuit of a video
work in similar fashion to that of Fig. 6-28 (A) . The amplifier, the output signal from this amplifier will
waveshape of the output signal from the filter net- be reversed in phase from that at its input. This is
work of Fig. 6-28 (B) is the video modulating signal, illustrated in the circuit of Fig. 6-29 (A). When the
Ilh IIII,..IIh.,,nIII,.,.IIII.
t
11i1 I,111rN,r1t,f llt I -F
FILTER
(A)
INPUT V DEO -F I DIODE TO GRID CIRCUIT
MODULATED OF FIRST VIDEO
SIGNAL DETECTOR AMPLIFIER
I- F
III LILI,..dtla...ntll...,11.Lt, FILTER
( g)
Ftc. 6-28.-How the phasing of the video modulated signal is controlled by the diode detector is shown here. In (A) the
signal is fed to the plate of the diode rectifier which will conduct only when the plate is more positive than the cathode. That is,
only that part of the signal above the reference line will cause the diode to conduct. When, as in (B), the signal is fed to the
cathode of the detector, the reverse is true.
signal input to the amplifier is the output signal from The resistance of R should be made appreciably
the video detector circuit of Fig. 6-28 (B), the phase lower than the reactance of C (1/5 is considered a
good proportion) in order that the high video fre-
relationship between the input and output video sig-
quencies not be attenuated. For a capacitive reactance
nals of the amplifier is as shown in Fig. 6-29 (B).
of 4000 ohms, resistor R should be about 800 ohms.
If another amplifier were connected to the output The over-all diode load (R and C in parallel) should
of either circuit of Fig. 6-29 (where the coupling offer a fairly constant impedance at the video fre-
between these circuits does not introduce any appre- quencies so that approximately the same output volt-
ciable phase shift), the added amplifier would merely age is produced at these frequencies. Since the capaci-
reverse the polarity of the output signals illustrated tance is the only factor in the parallel RC combina-
in this drawing. The output signal from each second tion of Fig. 6-30 that varies with frequency, the effect
video amplifier would have the same waveshape and of this capacitance must be diminished. To do this R
phase as the video signal input to their respective first should be decreased below the value of capacitive
video amplifiers reactance.
Let us study two examples to see how the imped-
Detector Loading ance of the RC combination can be kept fairly con-
stant. The magnitude of the impedance in ohms (des-
Whatever the type of video detector used, the out-
ignated as Z) of a resistance and capacitance in par-
put loading circuit presents quite a design problem
allel is given by :
because of the frequencies involved. In a -m radio re-

ceivers the loading on the diode detector is usually Z in ohms
quite high because the shunting capacities present in
where X, equals the capacitive reactance in ohms of
the circuit do not affect the audio-frequency response.
C at the frequency in question and R the resistance
In video detection the diode load has to be much lower
in ohms.
because the shunting capacities present in the circuit
affect the video -frequency response. The detected volt-
Assuming that the capacitance is still about 10 µµf,
age is low because of this low diode load and this low then from the lowest video frequency of 30 cycles to
voltage must be amplified to the level necessary for the highest of 4 mc the reactance X, will vary from
proper operation of the picture tube. Although the about 530 megohms to approximately 4000 ohms. If
low impedance of the diode load causes the distortion
the resistance of R is about 10,000 ohms, then the
introduced by this diode to increase somewhat, the impedance Z for the highest video frequency will be :
added distortion introduced does not cause any bad Z
4000 X 10000
V (4000)-2 -1- (10,000)2
-
3710 ohms
visual defects in the reproduced picture since we are
dealing with a video signal. At the lowest video frequency X, is very large com-
The video frequencies are as high as 4 mc. The total pared to R, and the impedance Z is considered to be
shunting capacities in the load circuit are high enough equal to 10,000 ohms, or the resistance of R.
to offer a low impedance path to these high frequen- From this analysis, the impedance Z is seen to vary
cies. Let us refer to the simplified diode detector cir-
cuit of Fig. 6-30. Resistor R represents the load re- -
FIG. 6-30. Simplified
detector circuit wherein
VIDEO DETECTOR
sistance across the cathode of the diode. Capacitor C R is the load across the VIDEO -i I
represents the total shunting capacitance in the circuit. cathode and C represents 51611AL INPUT
the total shunting capac-
This value of C is a controlling factor in the value of itance, the value of C
the load resistance. For example if C is 10 µµf, then being the controlling factor in the value of R.
at the 4 -mc limit of the video signal the capacitive re-
actance will be approximately 4000 ohms. Thus if the from 3710 ohms to 10,000 ohms. This means that the
video frequencies up to 4 mc must be passed with suffi- voltage output from the diode will vary appreciably
cient amplitude, the load resistance has to be much over the complete range of video frequencies and the
lower than 4000 ohms. If R is a higher value, the response level for the different video frequency will
be anything but constant.
capacitive reactance of C will offer a lower impedance
path to the high frequencies arid will shunt them to If R is made smaller, the impedance will vary less
ground. If R is lower, C will offer the higher im- over the range of video frequencies. If R is made
pedance to the video frequencies and most of the sig- equal to about 1000 ohms, then for the highest video
nal voltage will appear across the resistor. frequency Z will be :
X 4000 ture. In order to make the level of all the video fre-
Z= - 970 ohms
000) + (4000) 2
(11000 quencies as constant as possible, many diode detector
At the lowest video frequency X is very large com- circuits use some form of peaking circuit in the output
pared to the 1000 ohms of R so that it has very little of the detector. These peaking circuits essentially con-
effect upon the magnitude of Z, and thus Z can be sist of coils placed in the output of the detector circuit
considered to be equal to 1000 ohms at the low video (part of the coupling system to the first video ampli-
frequencies. Thus when the resistance R is lowered fier) and their values are so chosen that they will reso-
to 1000 ohms, the impedance of the load will vary by nate with the shunting capacitances. (In some cases a
only 30 ohms (from 1000 to 970) instead of the shunt capacitor is added to develop the desired reso-
former impedance variation of about 6300 ohms. The nance.) Resonance occurs at the high video frequency
variarión of 6300 ohms is about 60 percent of the end and tends to peak the response curve there. Re-
maximum impedance of 10,000 ohms, and while the sistors are usually shunted across these peaking coils
variation of 30 ohms is only 3 percent of the maxi- to reduce the height of the peak to the average level
mum impedance of 1000 ohms. From this comparison of the video response curve. A complete detailed
we can readily see that the lower the load resistor R mathematical analysis of peaking circuits is included
(of Fig. 6-30), the more constant the over-all load in the next chapter.
impedance over the complete range of video frequen-
In the following sections we will analyze a number
cies and, hence, the more level the response.
of different video detector circuits used in today's
The capacitance C in the previous analysis repre- receivers with regard to the type of diode connections,
sents the total shunting capacitances. These shunting
number of video amplifiers, grid or cathode signal
capacitances include the interelectrode and output injection to the picture tube, loading on the diode, and
capacitances of the diode detector, those due to the
the coupling circuit from the detector to the first video
stray wiring of the circuit, and those of any circuit amplifier. In all of the circuits to be discussed, the
that is coupled to the output of the diode. The total video signal fed to the picture tube is secured from
value of these capacitances is quite high for the upper the plate circuit of the last video amplifier.
end of the video frequencies. Since they are inherent
parts of the circuit, they are difficult to reduce the
;
diode load resistance, therefore, must be reduced to

Garod Model 3912 - TVFMP
keep the output voltage from the detector constant
over the complete range of video frequencies.
The amount the diode load resistance can be re-
receiver model 3912 -
In the video detector circuit of the Garod television
TVFMP, one half of a 6AL5
duo -diode tube is used as the detector and the input
duced is critical because the value of the total shunting video i -f signal is fed to the cathode (pin 1) of the
capacitances C may run as high as 25 µµf. At 4 me tube. This may be seen from the schematic of Fig.
the capacitive reactance of 25 µµf is approximately 6-31. The loading on the plate of this tube consists of
1600 ohms, and the detector load resistance would a 4.7 µµf capacitor, two peaking coils L1 and L2, a
have to be quite low for proper operation of the cir- damping resistor of 39,000 ohms across L, and a
cuit. If this load resistance is too low, not enough 3900 -ohm plate load resistor. The remainder of the
signal voltage will exist in the output. Therefore we coupling network consists of a 0.1-µµf coupling
are faced with the problem of using the lowest pos- capacitor and the 470K grid -leak resistor of the fol-
sible value of load resistance to provide adequate sig- lowing video amplifier stage.
nal output and yet to keep this output voltage as The two coils in conjunction with the fixed 4.7-µ4
nearly constant as possible over the complete range capacitor and the inherent slmnting capacitances of
of video frequencies. the circuit are so chosen that they will resonate at the
Since insufficiem signal output exists when the high -frequency end of the video signal, thereby help-
value of the detector load resistor is too low, this re- ing produce a level video response curve. The 39,000-
sistor may have to be somewhat higher than antici- ohm resistor across the series peaking coil L1 is em-
pated. In practice this load resistor generally varies ployed to reduce or clamp the peak caused by the reso-
from about 1500 ohms to 5000 ohms. There is appre- nance of coil L1 and its associated capacitance to fur-
ciable attenuation of the higher video frequencies with ther level the response curve.
these values of resistances. This is undesirable, espe- The output from the detector system is fed into two
cially in those television receivers that employ large video amplifiers and then into the grid circuit of the
screens because discrimination against the high video picture tube. By comparing the video detector and
frequencies will be noticeable on the reproduced pic- amplifier system of Fig. 6-31 with that of Fig. 6-27.
it will be seen that the correct polarity of the signal employed. This system is representative of the block
input to the grid of the picture tube is maintained. diagram circuit at the top portion of Fig. 6-27 (A).
This polarity, as established, should be negative. If The interesting part of this circuit is the loading
an even number of video amplifiers is employed, then network in the cathode output of the tube. Coil L; is
the input video signal to the control grid circuit of the a series peaking coil and in conjunction with the 10-
µµf fixed capacitor C'67 and the other shunting capaci-
p6AL5 39K tances in the circuit it is used to peak the high -fre-
2ND PIX DETECTOR
.1µf TO TWO VIDEO
AMPLIFIERS AND quency end of the curve. The loading resistor in this
THEN TO
GRID
CONTROL
OF PICTURE circuit is a 2000 -ohm potentiometer R4. across which
TUBE
470K
most of the output voltage appears. The variable arm
VIDEO I -F LZ
SIGNAL of this potentiometer is connected to the grid circuit
INPUT
3.9 K
of the following video amplifier tube through a 0.1-µ/.4
BIAS
coupling capacitor C66. Thus the variable arm con-
trols the amount of voltage input to the video ampli-
After Garod Radio Corp. fier and, hence, the video signal input to the picture
6-31.-The video detector circuit
of the Garod model tube and thus functions as a contrast control.
3912-TVFM P wherein the video i -f signal is feci to the cathode
of the 6ALS detector.
General Electric Model 802
first amplifier should likewise be negative. This nega-
tive polarity is obtained by feeding the video i -f signal In the GE model 802, the video signal input is fed
input to the cathode of the diode detector. This was to the cathode circuit of the picture tube. An odd
analyzed in connection with Fig. 6-28 (B) but it can number (one) of video amplifiers is employed so that
also be seen from Fig. 6-31. In this latter circuit, the the video i -f signal input to the diode detector must
diode current flow is from cathode to plate and there- be fed to the cathode of the circuit in order to have
fore takes the path indicated by the arrow. The the proper polarity of video signal to the picture tube.
polarities that appear across coil L2 and the 3900 -ohm This was shown in the lower block diagram circuit of
load resistor are as indicated in Fig. 6-31, and thus a Fig. 6-27 (B).
negative polarity will exist at the junction of L, and The video detector circuit for this receiver is illus-
L2 with respect to ground. trated in Fig. 6-33. The detector utilizes one half of a
6H6 duo -diode tube. The load resistor R18 in the
plate circuit of this detector is quite low, being 1500
Belmont Model 21A21 ohms. This resistor is shunted with a fixed capacitor,
The video detector circuit for the Belmont model C'33, of 10 µµf in conjunction with the inherent shunt-
21A21 is illustrated in Fig. 6-32. The video i -f signal ing capacitances that already exist across the resistor.
Series coil L10 is employed to peak the video response
/6AL5 curve at the necessary video high -frequency end. The
VIDEO DETECTOR
VIDEO I-F interesting thing about this circuit is that this peaking
SIGNAL
INPUT
L4 C66 TO GRID OF (ONE)
coil, or coupling coil as it is sometimes called, is in-
C67 R4 H
VIDEO AMPLIFIER
AND THEN TO
serted after the load resistor.
,oµµí
-
2000
CONTRAST
CONTROL GRID OF
PICTURE TUBE
Peaking coils are placed in different parts of the
coupling circuit between the output of the video de-
After Belmont Radio Co. tector and the video amplifier (as well as between
FIG. 6-32.-The video detector circuit of the Belmont model video amplifiers) to separate the inherent shunting
21A21, wherein the video i -f signal is feci to a plate of the
6A L5 duo -diode. capacitances that exist in the circuit. The reason for
VISA
is fed to the plate of the diode detector which uses one VIDEO DETECTOR
V2 - 6H 6
half of a 6AL5 duo -diode, the other half being used L TO GRID OF (ONE)
VIDEO -F 8 VIDEO AMPLIFIER
for automatic gain control (agc). The final video am- I
I
SIGNAL INPU O \-11-AND THEN TO

CATHODE OF
plified signal is fed to the control grid of the picture RI8
1500 To33
PICTURE TUBE.
tube. Due to this arrangement an odd number of video

amplifiers has to be employed to Obtain the necessary
polarity of the video signal input to the grid of the 6-33.-The video detector circuit of the GE model 802
feeds the video i- f signal through one video amplifier to the
picture tube. In this receiver one video amplifier is cathode of the picture tube.
the variation in placement is that the shunting capaci- There is no heater to cathode, nor cathode to plate
tances in the circuit are divided by the series peaking interelectrode capacitance as when a regular diode is
coil, and the amount of shunting capacitance on either used. There are some receivers on the market that
side of the coil varies in different circuits. The place- employ a 1N34 crystal as a video detector and do not
ment of the coil depends upon obtaining the correct use any peaking circuits in the output because of the
capacitance needed to form the resonant peaking reduced shunting capacitances in the circuit. A higher
circuit. value load resistor can then be used in the detector
In all of the detector circuits discussed so far, the stage.
fixed value of capacitance inserted in the diode load
serves as a filter for the video i.f. The output signal Television Assembly Model F1-101
from the detector is a video -modulated i -f signal that
The video detector circuits discussed so far have
has gone through the process of rectification. It con-
tains the video modulating frequencies -
up to 4 nc
(if the over-all video i -f response is designed to pass
employed one half of a duo -diode 6H6 or 6AL5 tube
or a crystal rectifier as the detector. The video de-
tector circuit of the Television Assembly model
the high frequencies) and also the carrier frequency F1-101 uses a detector system different from any-
which has been converted to an i.f. The video i -f car-
thing yet discussed. A 6J6 duo -triode is employed in
rier can be considered ill round numbers to be about
a video detector system which utilizes both triode
25 mc. Only the video modulating frequencies are to
sections of the tube.
be passed and not the i.f. The fixed capacitance in the
diode load is of a high enough value to present a low The circuit for this unique arrangement appears in
impedance path to ground at the video i.f. and there- Fig. 6-35. Although not shown, the receiver has two
fore filter it to ground.
R58
6J6 22K C46 TO TWO VIDEO
VIDEO DETECTOR .05 AMPLIFIERS AND
Consolidated Television Model 2315 VIDEO
---THEN TO CONTROL
GRID OF PICTURE
I -F L16 TUBE
In the Consolidated Television Corporation model SIGNAL
L17
INPUT
2315 a 1 N34 germanium crystal is employed as a
video diode detector. The diagram for this circuit R32
3.3K
appears in Fig. 6-34. The terminal of the crystal R60 .T L 2
marked with a negative (-)

sign represents the
100
R31
cathode and that with the positive (+) sign repre- 47
sents the plate of the detector. Two video amplifiers
are employed in this receiver and the video signal Fm. 6-35.-Unlike any of the models previously illustrated,
from these amplifiers is fed to the grid of the picture the Television Assembly model F1-101 employs a duo -triode
6J6 tube as a video detector. The circuit is a diode rectifier
tube. With this arrangement, for the proper polarity driven by a cathode -follower stage, with the first triode sec-
of video signal input to the grid of the picture tube, tion acting as the cathode -follower driver and the second as
the ,diode detector.
the video i -f signal must be fed to the cathode, or
negative terminal, of the crystal as indicated in Fig.
video amplifiers, and the output from the plate circuit
6-34.
of the last amplifier is feci to the grid circuit of the
A shunt peaking coil PL -1 and a series peaking picture tube. In order to obtain the correct polarity
coil PL -2 are used in the output of this circuit to of the video signal input to the grid of the picture
VIDEO DETECTOR
TO TWO VIDEO
tube with this system. the video i -f signal should be
CRYSTAL PL -2
AMPLIFIERS AND fed to the cathode circuit of a diode detector. This was
THEN TO CONTROL
GRID OF PICTURE
FIG. 6-34.-In the Con- illustrated in the block diagram of the top half of
1N34
TUBE solidated Television
VIDEO
PL -1 model 2315 a 1N34 ger- Fig. 6-27 (B). At a first glance at the detector circuit
I -F manium crystal is used
SIGNAL as a video diode detector. of Fig. 6-35, it may be difficult to see what type of
INPUT R14
3300 detection is employed. However, if this circuit is
studied carefully, it will be seen to be a diode rectifier
increase the response at the high frequencies. The which is driven by a cathode -follower stage. Let us
load resistor is R14, 3300 ohms. The use of a 1N34 now analyze this circuit in detail.
germanium crystal as a diode detector reduces the The first triode section of this tube is used as the
over-all inherent shunting capacitances in the circuit. cathode -follower driver. The video i -f signal output
from the last video i -f stage is feel to the grid, pin 6, Hallicrafters Model T-54
of the 6J6 detector tube. The series circuit in the
cathode consists of a 47 -ohm resistor, R31, and coil In the video channel section of the Hallicrafters
L32. This network is unbypassed and the video i -f model T-54 is employed a unique video i -f, detector,
signal appears across this cathode load. The other and amplifier system. Because the video diode de-
triode section of the tube has its plate and grid tied tector is tied up with the operation of the entire sys-
together and in conjunction with its cathode will act tem, the discussion of this receiver is included in this
as a diode. This section of the tube acts as the video chapter. One half of a 6H6 duo -diode is used as a
diode detector. diode detector and also as a diode mixer and source
Since the cathode of the 6J6 tube is common to both for automatic gain control (agc). The output from the
triode sections, it represents a ready source from last video amplifier stage of this unit is not taken from
which the diode section can obtain its video i -f signal. the plate of the tube as was the case in all the other
The cathode-follower system is one method that re- systems discussed but rather froni the cathode. All
sults in a fairly high video i -f signal input to the de- these features will now be discussed in detail.
tector so that, with the usual low load impedance on A schematic diagram of the video detector and am-
the detector, an adequate amount of rectified voltage plifier unit is illustrated in. Fig. 6-36. Although not
output is still obtained. shown here, all the i -f amplifiers employed in the
The video i -f voltage drop across the cathode cir- video section can be considered as composite ampli-
cuit is the input signal to the detector. The output im- fiers. Ordinarily, separate i -f amplifiers are employed
pedance of the cathode follower is low compared to for video and sound to prevent interaction between
the series impedance of the detector section and the the two signals. In this receiver, however, the inter-
load on the plate of the detector. This output im- action is put to use; the two i -f carriers are fed into
pedance of the cathode follower (in this case consid- the video diode detector circuit and beat together to
ered as the source impedance, since the input or source produce a new i -f signal.
voltage appears across this circuit) and that of the
The composite sound and video i -f signal is fed to
detector plus the plate load present a voltage dividing
the cathode of the diode as seen in Fig. 6-36. The
network to the detected signal. This means.that after
bandwidth of the video i-f system is narrow and some
detection most of the detected video signal appears
across the load. high-frequency video signals are attenuated. How-
The load of the detector essentially consists of the ever, the seven-inch picture tube that is employed
series peaking circuit of L16 in parallel with R58, the still has adequate definition of the picture even with
shunt peaking coil L17, and the plate load resistor the loss of some of these high video frequencies. The
R32. The output video detected signal is coupled to accompanying sound i.f., being on the high -frequency
the following video amplifiers through the 0.05-µf end of the video i -f response curve, is also attenuated
capacitor C46. but enough accompanying sound i -f signal does pass
4.5MC SERIES
TO GRID CIRCUIT TUNED CIRCUIT 65H7
OF F -M SOUND VIDEO
I -F AMPL. C58
lµµf AMPL.
L20 C51
SN7
VIDEO
6H6 .25 OUTPUT
AGC B+
DIODE MIXER E L18
VIDEO DETECTOR
COMPOSITE
SOUND AND 8
L21 R64
VIDEO I-F 530K 1
PICTURE
SIGNAL INPUT R58 R69 TUBE
C59 1 MEG. .25
R55 1MEG
1MEG L19
C45
AGC R59 R65
C54 R55 R57 1000 6.8K
.05
T 560K 5.6K CONTRAST
B+
FIG. 6-36.-The video detector and amplifier unit of the Hallicrafters model T-54 makes use of the interaction of the video
and sound signals, and the composite signal is fed to a 6H6 duo -diode tube which functions as diode detector, diode mixer, and
source for agc.
through these circuits and enters the video detector signal is a 4.5 -mc beat signal which varies in fre-
with the video i -f signal. quency and amplitude. Although the beat signal volt-
In this receiver the video i -f carrier is 26.25 mc age is small compared to the video signal, it will have
and the sound i -f carrier is 21.75 mc, the difference sufficient strength after amplification by the 6SH7
being 4.5 mc, the standard separation between the video amplifier to be used as the sound i -f signal.
video and sound carriers. As mentioned, these two Directly at the plate circuit of this video amplifier is
i -f signals are mixed together in the diode detector a series tuned LC circuit, consisting of capacitor C58
and the resulting beat signal is 4.5 mc. Besides the and variable inductance L29, which is tuned to 4.5
diode acting as a mixer, it also acts as a rectifier to mc. This series tuned circuit. therefore, acts as a low
the a -ni video i -f signal. Thus detection of the video impedance path to ground for the 4.5 -mc beat signal.
signal also occurs in this tube. The 4.5 -mc beat signal Since this is a series tuned circuit, maximum current
together with the video detected signal is coupled to will flow at resonance and a maximum voltage drop
the following 6SH7 video amplifier stage, as seen in will appear across each element in the circuit. The
Fig. 6-36. 4.5 -mc signal is tapped off across L29 and fed to the
The coupling circuit is an RC stage to which series grid circuit of the only sound i -f amplifier in the cir-
and shunt peaking circuits have been added to peak cuit. From this sound i -f amplifier, the signal is fed
the high frequencies. The series peaking circuit con- to a ratio detector circuit. This circuit responds only
sists of coil L18 in parallel with R54 and the shunt to f -m and not to a -m variations in the input signal.
peaking circuit consists of L19 in parallel with R55. The f -m variations of this signal are then detected and
The 5600 -ohm diode load resistor, R57, is in series applied to the audio amplifiers of the receiver. A tele-
with the shunt peaking circuit, and both together offer vision sound system of the type just discussed, where
a fairly high impedance load to the diode so that both the sound and video i -f signal output of the fre-
appreciable voltage appears between the junction quency conversion system beat together to form a new
point of L18, L19, and ground. The composite signal, sound i -f signal, is commonly known as an intercarrier
the 4.5 -mc beat signal and the detected video, is fed sound system, since it depends for its action upon
to the grid circuit of the 6SH7 video amplifier intermodulation of the sound and video (i -f) carriers.
through the 0.1-4 coupling capacitor C52. The C58 -L29 series -tuned 4.5 -mc circuit prevents
Attached across R57, the diode plate -load resistor, the beat signal from interfering with the video signal
is a resistance capacitance network, R56-054, which reproduction. Thus only the video signal output from
acts as a filter to the alternating video signal. This the plate circuit of the first video amplifier is coupled
filter is similar to the avc filters used in a -m radio to the video output tube. As seen in Fig. 6-36, the
receivers in which the audio is filtered out and a nega- series and parallel peaking networks. L20-R62 and
tive d -c signal remains which is used for avc. Thus in L21 -R61 respectively, also appear in the plate circuit
Fig. 6-36, R56 and C54 filter out the alternating video of this tube. The video signal is finally coupled to the
variations and a negative d -c signal remains which is video output tube, which is one triode section of a
used for agc on the first and second video i -f ampli- 6SN7. This output tube functions as a cathode fol-
fiers. (It should be remembered that the composite lower where the cathode load is just the 6800 -ohm
input signal to the cathode of the detector tube has a resistor R65. The output from this cathode -follower
d -c component and therefore agc is possible.) circuit is applied to the cathode of the picture tube
Let us now return to the 6SH7 first video ampli- through the 0.25-µf coupling capacitor C45.
fier. The input signal, the 4.5 -mc beat signal and the Compensating networks are not needed because the
video modulating signal, is amplified by the tube. The video output signal is taken from the cathode and not
cathode resistor, R59, of this tube is an unbypassed the plate circuit of the video output stage. The output
1000-ohm potentiometer, and thus degeneration oc- tube does not contribute any gain when acting as a
curs in this tube. By varying the value of this resist- cathode follower, but the preceding video amplifier
ance, the amount of degeneration and bias of the tube provides enough gain for proper video signal repro-
is changed and the gain of the tube can be controlled. duction.
This potentiometer therefore acts as the contrast con- The type of coupling system to the picture tube is
trol of the receiver. The approximate gain of the tube completely different from any of the circuits pre-
is about 20 to 25 times. viously analyzed, so it does not fall into any of the
The signal resulting from mixing the 26.25 -mc a -m categories of Fig. 6-27. However, for proper opera-
signal (the video i -f carrier) and the 21.75 -mc f -m tion of the picture tube, the polarity of the video sig-
nal input tò its cathode should be positive with respect This negative video signal is reversed in phase by
to its grid. The proper polarity is obtained from this 180° due to the inherent phase reversing qualities of
circuit but this may not be readily evident. Let us the tube and appears in the plate circuit as an ampli-
analyze the complete circuit of Fig. 6-36 to see how fied, but positive polarity, video signal. This signal is
this occurs. next coupled to the grid circuit of the cathode -follower
The input video modulated i -f signal to the diode video output tube. Negligible phase change occurs
detector is fed to the cathode of the tube. Only the between the grid and plate circuit of this tube, so that
negative part of this signal will cause diode current to the positive polarity that exists for the video signal at
flow because it effectively makes the plate more posi- its grid also exists at its cathode. Since the cathode
tive than the cathode. The positive part of the signal of the picture tube is connected to the cathode of the
is thus clipped, and the negative video signal is then video output tube, it receives a positive video signal
coupled from the diode plate circuit to the grid of the which is necessary for the correct operation of the
6SH7 video amplifier with negligible phase change. picture tube.
CHAPTER 7
VIDEO AMPLIFIERS AND D -C RESTORERS

BY HENRY CHANES
The signal at the output of the video second de- solid line represents the response of a good audio
tector contains all the information necessary to amplifier.
form the picture on the face of the picture ttlbe. In an ordinary audio amplifier, the amount of
The sync signals will be removed from this signal phase shift of a signal going through the ampli-
and used to sync the sweep circuits this is dis- ; fier is not an important consideration, as the ear is
cussed in chapter 8. We are now mainly interested not capable of detecting phase distortion. In a video
in the blanking signals and the picture components amplifier this is a very important consideration.
of the signal at the output of the video second de- Let us consider phase shift in terms of time delay.
tector. The blanking signals are used to drive the The time that the signal is delayed in going through
cathode -ray -tube grid to cutoff during both hori- the amplifier is very small, being of the order of
zontal and vertical retrace. The spot on the screen microseconds (millionths of a second), when com-
is thus extinguished during the retrace time to pre- pared to our usual concept of time. However small
vent interference with the picture. The picture com- this time delay may seem, it is very important. Con-
ponents of the signal determine the brightness of sider a 12 -inch picture tube with a picture width
each small picture element. These elements, in turn, of approximately 9 inches. The spot moves across
make up the complete picture. The signal available the screen 15,750 times a second, or at the rate of
at the output of the video second detector is usually 142,000 inches per second. If a particular fre-
only about 1 volt, peak -to -peak. This is insufficient quency in the video signal that represented a light
to drive the picture tube properly, and therefore and dark pattern were delayed by only 1 micro-
this signal has to be amplified. The function of the second, the pattern would be displaced on the screen
video amplifier is similar to that of the audio ampli- by about 0.14 inch. If every frequency were de-
fier in a receiver, but its construction is complica- layed the same amount of time, the delay would
ted by stricter requirements. Let us examine these not be noticeable, as the entire picture would be
requirements. moved over a little. However, if different fre-
quencies were delayed by different times, the dis-
Requirements of a Video Amplifier tortion in the picture would be immediately appar-
ent. The relation between the phase shift and the
The video amplifier must amplify equally well time delay is given by
all frequencies from as low as 30 cycles to ashigh
as 4 mc. A good audio amplifier has to cover a
range of only about 70 to 10,000 cycles. Fig. 7-1
T= 360 f
Eq. 7-1
illustrates the different bandwidths required by the
two amplifiers. The clashed line .is the frequency where T = time delay in seconds
O = phase shift in degrees
response required of the video amplifier, and the
f = frequency in cycles
From this formula it can be seen that to keep the
2
a time delay constant for all frequencies, the phase
W
shift must increase in proportion to the frequency.
FACTUAL DESIRED
CHARACTERISTIC CHARACTERISTI C il
This .is known as linear phase shift and is illustra-
ted in Fig. 7-2.
10 CPS 100 CPS 1000CPS 10KC 100KC 1MC 10 MC The gain necessary in a video amplifier depends
FREQUENCY on the output of the video second detector and the
amount of signal necessary to drive the grid of the
FIG. 7-1.-The solid line represents the bandwidth of an picture tube properly. This gain varies from about
audio amplifier. The dotted line represents the desired band-
width of a video amplifier. 25 for a 7 -inch tube to about 50 for a 10- or 12-
120
VIDEO AMPLIFIERS AND D -C RESTORERS 121
inch picture tube. The number of stages necessary the next tube. There is also a phase shift due to
to achieve this gain is usually one or two. this resistor -capacitor combination.
The other requirements for the video amplifiers At the high frequencies the effect of Ct becomes
are those associated with any amplifier. They include appreciable. Ct represents the total shunt capacity
due to the output capacity of the first tube, the in-
put capacity of the second tube, and all the stray
capacity due to the wiring. At the high frequencies
the reactance of this capacitance becomes small
enough to lower the impedance of the plate load on
the tube. The plate load, instead of being a pure
resistance, will then be a resistance in parallel with
a capacitive reactance. This smaller load impedance
will cause the output voltage to drop and the capaci-
tive component of the load will cause a phase shift
o 1MC 2MC 3MC 4-MC in the output signal.
FREQUENCY Because of the deficiencies mentioned above, it
is evident that the ordinary RC coupled amplifier
FIG. 7-2.-The dotted line represents the desired linear phase
shift and the solid line represents the actual phase shift as it is not good enough for use as a video amplifier
varies with frequency. without some form of compensation to offset the
falling off in response at the low and high f re-
quencies and to give a linear phase shift to pre-
minimum harmonic distortion, a satisfactory sig- vent phase distortion. We will now consider the
nal-to-noise ratio, and decoupling when necessary different methods used for low- and high -frequency
to prevent oscillation. These offer no special prob- compensation.
lem, except where complicated by the necessity of
a large bandwidth and linear phase shift.
Low -Frequency Compensation
Basic Circuit for Video Amplifier We have seen how the combination of coupling
capacitor Cb and grid -leak resistor R0 reduced the
The basic circuit for a video amplifier is an RC low -frequency response of the video amplifier. These
coupled amplifier as shown in Fig. 7-3. This cir- effects of attenuating the low frequencies must be
cuit has no compensation for either the high- or low - compensated for in video amplifiers. Video ampli-
f requency response. The gain will fall off at both fiers employ pentode tubes of such design that they
the high- and low -frequency ends, and the phase have relatively high values of mutual conductance,
shift of the signal will not be proportional to the g,,,, and exceptionally small values of interelectrode
frequency. The amplitude and phase response char- capacitances. The reasons for these design features
acteristics of a typical amplifier of this type are will soon be evident. The background of the repro-
shown in Figs. 7-1 and 7-2. The solid line repre- duced picture seen on the screen of a television re-
sents the actual response, and the dashed line repre- ceiver is due to the very low frequencies of the
sents the response desired in the video amplifier. video signal received. If these low frequencies were
Examining the amplitude response, it can be seen attenuated, the picture would not be in good con-
that the response will have to be increased or trast to its background and this is undesirable. Be-
"peaked" for both the low- and the high-frequency fore going into a circuit analysis of low -frequency
ends of the frequency range. Let us briefly dis- compensation, it should be stated that it is difficult
cuss the reasons for the response dropping off in
this basic amplifier.
Referring to Fig. 7-3 and considering the low
frequencies, we can see that the coupling capacitor
Cb is in series with the grid resistor R0 of the fol-
lowing tube. The output signal from the first tube
is applied across this series combination. As the
frequency becomes lower, the reactance of the ca-
pacitor Cb becomes higher, and less of the output
signal appears across resistor R0 to be applied to FIG. 7-3.-Typical RC coupled amplifier.
to test video circuits for proper low -frequency re- phase shift between the two tubes. The coupling
sponse with the input and output measuring instru- capacitor Ch causes the shift. Why, then, is the
ments that are generally used for testing the re- value of Cb not made large enough so that it will
sponse of ordinary RC couplings. In checking the offer negligible resistance at low frequencies? The
degree of attenuation at the low frequencies, it is value of Cb could be increased and this would im-
well to use a low -frequency square -wave input and prove the low -frequency response, but since we are
to observe the distortion of the output signal wave- dealing with frequencies up to 4 mc, the larger the
form on an oscilloscope. Such square -wave analysis value of Cb, the greater will be the shunt capaci-
should cover the range from 20 cycles to 1000 tance to ground that it introduces. These shunt ca-
cycles. When the amplifier is properly compensated, pacitances, although extremely small, are very det-
the square -wave output resembles the square-wave rimental at the high frequencies. Low -frequency
input very closely. (On an oscilloscope distortions compensation is used to avoid the introduction of
as low as 5% can be recognized. This is considered these shunt capacitances. The insertion of R'bC'b
accurate for design purposes.) compensates for the loss in gain at the low fre-
When an RC coupled amplifier is compensated quencies and also introduces a phase shift to com-
at the low -frequency end, another resistance R'b is pensate for the undersired phase shift introduced
placed in series with the plate -load resistor Rb and by RcC6.
a capacitor is placed across the former resistor. Since the combination of ReCb at low frequencies
The reasons for the use of these extra elements reduces the gain, the combination of R'bC'b at these
and the typical values chosen will be discussed. In same frequencies is inserted to offer a higher load
Fig. 7-4 is shown a typical low -frequency com- impedance to tube V
and increase the gain. In-
pensated RC coupled amplifier in which R'bC'b is creasing the load impedance will increase the effec-
added to better the low -frequency response. This tive voltage across the load and bring the gain back
addition of R'bC'b appears to be nothing more than to normal. Again referring to Fig. 7-4, the opera-
a decoupling network. In reality it is a decoupling tion is as follows the voltage across the grid -leak
:
network with the screen B+ lead connected directly resistor decreases (thereby decreasing the input to
to B+ instead of between Rh and R'b. Besides its the grid of tube V2) because the reactance of Cb
immediate use of raising the low -frequency response, increases as the frequencies are decreased. The re-
it decouples any undesired feedback voltage clue to actance of capacitor C'b increases at the same time.
the common impedance of the plate supply. As a result, the parallel combination of R'bC'b is
Since the tube used is a pentode, we can make seen to be an impedance that is added to the plate-
some approximations for simplifying the analysis load resistor Rb and effectively increases the total
of this circuit. The grid -leak resistor Re is much load impedance on tube V, as the frequencies are
larger than the plate -load resistor Rh (RC>>Rb) decreased.
and the plate resistance of the tube Rp is m:.ch Since this RC coupled amplifier stage uses a pen-
greater than Rb (Rp»Rb) ; therefore all that is tode, then Rp»Rb and the gain at low frequencies
necessary is to take into consideration the essential in the case of this compensated stage is expressed
effect of the combination R,.Cb. At low frequencies as follows:
ReCh causes a drop in signal potential from the

1
plate of tube V, to the grid of tube V, and also
A iaw = A1ß., , I1 + ¡Xeb \2 Eq. 7-2
\Z load
The letter A is used to designate the gain of a
stage, and X is the expression for reactance and Z
that for impedance as usual. Therefore :
Elio, = the gain at low frequencies

= the gain at intermediate frequencies
Xeb = the reactance of the capacitor Ch
Z,0 , = the total effective load impedance,
8+ including Rb, R'b and C'b
From this equation it may be seen that the effect
FIG. 7-4.-Typical low -frequency compensated RC coupled of the original plate -load resistor Rb has been
amplifier. The decoupling network R'hC'h has been added to
the circuit of Fig. 7-3 to better the low -frequency response. changed to Zioae. Since the plate -load impedance has
been increased by the R'bC',, combination, the gain

= A t 1
Eq. 7-3
at low frequencies is also increased. As the fre- + fXcb\º
quency decreases, the ratio of XCh/Z,,,0,, also de-
creases, making Eq. 7-2 approach the gain at in-
1 R,/1
termediate frequencies, A;,,,. It would be ideal if re- If this equation is studied, it will be found that the
sistor R'b could be increased to an infinite arrount phase shift, due to the R,Cb combination, at the low
but the need for an appreciable amount of plate frequencies is:
voltage on the tube makes this impossible. If this
could be achieved, the gain would be equal to that B = tan -r X`'b Eq. 7-4
of the tube itself and no loss would be encountered R,.
;
the gain would be maximum and flat throughout where 0 is in degrees, and the expression tan--'
the response. From experimentation it has been means the angle whose tangent is the quantity fol-
found that for the proper response the time con- lowing the expression. It can be seen that the coup-
tants of R',,C'b and ReC,, should be equal to each ling capacitor Ch causes a lead in phase. Inserting
other. The voltage loss due to Cb will then be com- the C',R'h network into the load tends to cancel
pensated (in direct proportion) by the voltage gain this phase lead by introducing a phase shift in a
due to the total load impedance, in such a *way that lagging direction.
the voltage across the grid -leak resistor Re will Besides the grid -leak resistor and coupling ca-
remain constant. In many instances the combination pacitor combination, there are two other resistance -
of RbC'b is chosen first and the coupling capacitor capacitance combinations that cause a falling off in
Cb is made as large as is desired, then the grid - gain and a leading phase shift at the low frequen-
leak resistor Re is determined. For instance, assum- cies. First, there is the effect of the RhCh bias
ing that C',, is chosen to be 30µf, C,, chosen as 0.1 µf, impedance. The best way to eliminate this effect
and the time constant of ReCb to be 0.04 second would be to eliminate RhCh and apply bias another
(or 40,000 microseconds) it is desired to know way. This is usually not desired as cathode bias is
what the values of Rb, R'b and Re are, at the lowest preferable in many cases, so it has been determined
frequency of operation, which in this case is 30 experimentally that this undesired effect can be
cycles.Therefore: compensated for by making the time constant of
ReC, = R,C',, = 40.000 microseconds, RhCh equal to the time constant of R'hC'b. Mathe-
and since C'b = 304 matically we have:
RhCh = R',,C'b Eq. 7-3
40.000 40.000
Rb = 1333 ohms Secondly, there is the effect of the screen grid cir-
Cb 30
cuit. A pentode tube possesses a dynamic screen
and since Cb = 0.1 µf then : grid. resistance as well as a dynamic plate resistance.
This screen grid resistance R«; (so called to differ-
R,. =
40.000
0.1
- 400.000 ohms entiate it from the screen dropping resistor R8) in
combination with the bypass capacitor Cs also
R'b should be chosen to be about 20 times greater causes a falling off in response and a phase shift
that the reactance of C'b at the lowest frequency to (leading) at the low frequencies. The effect of this
be passed (30 cycles in this case) and yet not drop combination is similar to that of the R,.Cb combina-
too much of the supply voltage. Therefore : tion and experimentally it has been found that for
the phase shift not to exceed 2° at 60 cycles, the
R'b = 20 X e'r, = 20 X 27r time constant of R,56CR should be greater than
b 0.076 second. Since it is difficult to vary such an
element as R, 0, and because C, is the only variable
(7r) (30) (30) (10
10
6)
- 3540 ohms parameter, little compensation can be accomplished
by a single stage better compensation is obtained
;
We therefore find that : by the use of several stages of amplification.

Rb should be about 1300 ohms A very important thing to note is that in all these
Re should be about 400,000 ohms compensating problems R'bC'b plays the important
R'b should be about 3500 ohms role. Although R',C'b can be used to compensate
The equation for the gain at low frequencies of for any one of the above cases, only one case can
an uncompensated RC coupled amplifier is: be compensated for in a stage such as shown in
Fig. 7-4. A usual practice in cathode biased ampli- Considering a single stage of coupling, these capa-
fiers is to make Re very large (this is usually pos- cities include the output and input capacity of the
sible) and, therefore, the time constant of ReCb will first and second tubes respectively, the interelectrode
be large and the needed compensation is shifted to capacities of the tubes themselves, and any stray
the RKCK combination (instead of the ReCb=RbC'b wiring capacities, such as those between the coup-
combination) . ling capacitor Cb and ground. The equation for
In Fig. 7-5 are several curves showing the effect the gain at high frequencies of RC coupling, as-
of the insertion of the R'bC'b network on the low - suming Rr>>Rb and RC»Rb, becomes:
frequency response for different values of R'bC'b.
Eq. 7-6
gm Rb gm Rb
Ahica =
+
Ut7 1/1
2
Rb (2,fCß
J Rb)2
1
( Xcj)
hó
Let us consider equation 7 - 6 for a moment.
.2
o
74.2° CJ
10,000
What this relation states (every mathematical ex-
pression is a form of statement) is that the gain of
an RC coupled amplifier stage at high frequencies,
.1 1 0 10
FREQUENCY
(IN CYCLES
100
- 1000
Ah;5,b, is dependent upon four distinct factors, these
are the transconductance or g,,, of the tube used,
Fio. 7-5.-The effect of inserting different values of the the plate load, or Rb, on the same tube the fre- ;
R'hCh network is shown. As the time constant of the de - quency of operation f, and the total shunting capa-
coupling network is increased, the low -frequency response
increases. citances C,. The other quantities used in the ex-
pression are no more than actual numbers or nu-
From these curves it is seen that as the time con- merical constants. The expression further states
stant of the R'bC'b combination is increased, the that the true gain of the RC coupled stage involves
low -frequency response increases to frequencies as the four above mentioned quantities in such a way
low as a few cycles per second. However, there are that the gain is equal to the product of the trans -
practical limitations to the values of R'b and C'b conductance of, and the plate load on, the amplifier
that can be chosen. tube, all divided by the relation inside the square -
root sign. This square.-root relation is equal to one
(1) plus the square of the product of a numerical
High-Frequency Compensation constant 27r which equals 2 x 3.14 or 6.28. multi-
High -frequency compensation is even more implied by the frequency of operation f, the shunting
portant than low -frequency compensation because capacitances C,, and the plate load Rb.
of the higher band of frequencies to be passed. In Consequently from the above expression it is
video amplifiers, frequencies as high as 4 megacycles found that if the g,,, of the tube in question is high,
so will the gain, Aldo, be high but if the shunt ca-
have to be passed with a fairly flat response from ;
the minimum frequencies on upward. Therefore, pacitances, C,, of the tube and the frequency of
this type of compensation is considered as the more operation, f, are high. then the gain A,,;,,b will de-
important one. crease. It can be seen how much information this
There is really only one part of the RC coupling mathematical expression gives. Even an actual com-
circuit that reduces the response at high frequen- putation is not difficult, when all the values of g,,,,
cies, as compared with the three different networks Rb, f, and C, are known it can be easily and simply
;
affecting the low frequencies. Referring to Fig. 7-3, executed in a few steps. Let us assign values to the
the total shunt capacities, represented by the capaci- symbols and illustrate a typical calculation, which
tance Cr, existing in the RC circuit are the chief consists of a series of multiplications and a division.
reasons for the high -frequency response falling off. Assuming Rb equals 50,000 ohms, g,,, equals 1000
Since these shunting capacities are difficult to re- micromhos, f equals 1500 kc, and C, equals 30 µµf,
duce due to their inherent qualities, compensation then :
must be made to offset their effects. Since we are

Aeion =
dealing with very high frequencies, the reactance of .001 X 50,000
these shunting capacities are very low and they 1+(2X3.14X1500X103X30X10- 12X50,000)2
offer a low impedance path to these frequencies.
Starting with the denominator (the expression It is seen that with a tube of good mutual con-
under the square -root sign), we find that (2 X 3.14 ductance and a circuit with the total shunt capaci-
X 1500 X 103 X 30 X 10-12 X 50,000)2 is nothing tances Ct small, then the gain at high frequencies
more than a series of multiplications and is equal will be high. The gm of the tube is controlled by
to 14.1. Squaring this product gives 199. Adding this using special high gm pentodes designed especially
number to one (1), gives the total number under for video -band amplifiers. The undesired effect of
the square -root sign which is 1 + 199 or 200. C, is reduced by inserting inductances in the cir-
Evaluating the square root of 200, it is found that cuit at special places. These inductances effectively
14.14 is close enough for practical purposes. Up to increase the load impedance on the amplifier, there-
this point, only the denominator which is equal to by increasing the gain lost by the shunting capaci-
14.14 has been calculated. The numerator of the ex- ties. This is analogous to low -frequency compensa-
pression is equal to 0.001 X 50,000 or 50. Only one tion where the effective load impedance is also in-
more step is required to obtain the final value of creased. The simplest type of compensation at the
Ahigh, and that is to divide the numerator by the high frequencies is to insert an inductance L,
denominator. Therefore : (known as the "peaking coil") in series with the
plate -load resistor Rb. L, is known as a peaking
50 = 3.53
Ahagn coil because it effectively "peaks" the high-fre-
-14.14 quency end of the response curve. This type of
From the above mathematical computation the stage is sometimes referred to as a "shunt-peaked"
gain at the middle frequencies which is equal to stage because the coil is placed in shunt with the
g,,,Rb is 50, as compared to the gain at the high- tube (effectively from plate to ground) .
frequency end of the band, which is 3.53. This Fig. 7-6 shows a typical video amplifier compen-
drop in gain, as mentioned before, is due to Cr. sated at the high -frequency end only and its equiva-
Rp
(g) (C)
FIG. 7-6.-A typical video amplifier compensated only at thz. high -frequency end is shown in (A) and its equivalent circuit
in (B) and (C).
lent circuit at these frequencies. In Fig. 7-6(B) which is what we have at the 3 db point. Therefore
the coupling capacitor Cb is omitted because at these we get :
high frequencies its reactance is so low that it be- 1

comes effectively a short circuit. The total load im- Rb=Xe:-2,rfeCt Eq. 7-9
pedance on the first tube V, is the combination of
Rb in series with L9, both in parallel with the shunt- where fe = the frequency where correction starts.
ing capacities Ct as shown in the dotted portion of Under this criterion the radical of equation 7-8
Fig. 7-6(B). An analysis of this circuit is neces- becomes equal to the square root of two and the
sary in order to show how the effective load im- minimum load impedance that will be presented to
pedance is increased in order to increase the gain. the amplifier at the resonant frequency of L9 and
(The grid -leak resistor Re is neglected as part of Ct is:
the load impedance because it is usually very high Z1 = Xet = 1.41 Xet Eq. 7-10 (A)
and it has little effect on the parallel branch.)
The total load impedance represented by Z1 is given
or Z1 = X t, = 1.41 X,, Eq. 7-10 (B)
as follows :
We must decide what is the maximum frequency
that we desire to compensate for. Assuming this
rLt et VRb2 )(iv' Eq. 7-7 frequency to be the 4 megacycles mentioned, then
/X
V Rb2 + (X19 - Xe8)2 this 4 megacycle frequency is known as the correc-
tion frequency for which the wide band video
It is known that a parallel resonant circuit has a amplifier must be designed. When the parallel im-
very high impedance at resonance and, therefore, a pedance becomes effectively a pure resistive load to
large voltage will be developed across it. Conse- the tube (at resonance), then the total reactance of
quently, since the effective load impédance has to the circuit is zero and the reactance of one branch
be increased to offset the drop in response, the is equal to that of the other. At this point the in-
parallel circuit of Rb, L, and Ct is made to resonate ductance L,, becomes :
at the highest frequency that we desire to amplify.

The proper values needed for this resonance are
L,, = CtR2b Eq. 7-11
chosen from Rb and L9, because Ct is fixed. Assum- Eq. 7-11 states that if we know the load resist-
ing that the resonant frequency has already been ance Rb used and have figured the shunting capa-
established in the circuit, the capacitive reactance is cities, we can calculate the value of shunt peaking
equal to the inductive reactance (XL, = Xet), pro- inductance (L,,) needed. It has been found from
vided that the resistance Rb has a minor effect. The practice that when a response curve is plotted for
load impedance (Eq. 7-7) then becomes: all values of frequency versus gain, using the com-
Z1 = Xet V1 + (Xt,/Rb)t Eq. 7-8 (A) pensated circuit of Fig. 7-6(C) and Eq. 7-11, the
high-frequency end of the curve is improved con-
or Z1 = Xet V1 + (Xet/Rb)"- Eq. 7-8 (B) siderably over that of an ordinary RC coupled cir-
or Z1 = X1, V1 + (Xet/R02 Eq. 7-8 (C) cuit. The amplitude or the gain of the response
curve receives the improvement, but the over-all
or Z1 = X19 V1 + (X1,/14)2 Eq. 7-8 (D) curve is not as flat as we would desire because the
Eq. 7-8 (A), (B), (C) and (D) are all equal value of inductance used (as calculated by Eq. 7-11)
because they are evaluated at the resonant frequency seems to be too large and has caused too high a rise
where XL, = Xct under the assumption above. in gain at the critical frequency. The rise or peak
From our previous knowledge of resonance of the curve due to coil L,, is the reason that L,, is
curves, it is known that the critical frequency or the termed a "peaking" coil. From experimentation it
minimum frequency (for either the high or low has been found that for over-all flat response, down
end) acceptable for design purposes is that point on to the frequency where the load resistance Rb is
the curve where there is a 3 db drop in gain. At equal to the capacitive reactance of Ct, the peaking
the high end, it is this frequency that we start with coil should have a value equal to half of that repre-
in correcting the gain because all the frequencies sented by Eq. 7-11 or mathematically :
between the intermediate ones and this high one L,,= 1/2 Ct(Rb)2 Eq. 7-12 (A)
fall into the acceptable area. For good response in
RC coupled amplifiers the load resistance is made and since we know that at the frequency of correc-
equal to the reactance of its shunting capacities, tion
0.5Rb 2210
Rb =
1
where Ct =
1 Lp = 27r 4ir (3 X 106)
= 58.6 microhenrys
2vrf,C1 2orfoRb
and substituting this in Eq. 7-12 (A) we get Therefore we find that the approximate value
0.5Rb needed for the plate -load resistance Rb is about
L, = 1 / 2X 1 X (Rb) 2= 2200 ohms and that for the shunt peaking induc-
2rfcR° 2nic
tance L, is about 60 microhenrys.
and since As an illustration of how the peaking inductance
wc = 27rfc
L, affects the high frequencies let us examine some
then L, can be written as typical response curves. Fig. 7-7 shows a number
of different curves, where each one utilizes a dif-
0.5 Rb
L, = 0,0
Eq. 7-12 (B) LP:q/gCT Rb2
LR.,/2CTRb2
Eq. 7-12 (A) and (B) above are design formu-
1
LP2/gCTRb2
las that have actually been derived from practice. NO Lp
The response curve will be very flat down to the fre-

quency of correction if the above formulas are used.
The actual resonant frequency of the parallel cir-
cuit (in which L, is derived by formula 7-12) is
approximately V2 or 1.41 times the frequency of
correction.
In summarizing the above equations we can say
CORRECTION FREQUENCY
that the shunt peaking inductance L, can be deter- FREQUENCY
mined in one of two ways. FIG. 7-7.-The manner in which the peaking inductance
First, all that would have to be known to find affects the high frequencies is shown in these typical response
L, would be load resistance Rb and the frequency curves, each utilizing a different value or L,.
of correction and not the shunting capacities. This
ferent value of L,. Note that the curve where the
is evident from Eq. 7-12 (B).
response is the flattest is when L, = 1/2 CtR2°.
Secondly, we do not have to know the frequency
of correction in order to find the shunt peaking in-
ductance L, but just the load resistance Rb and Series Peaking Compensation
shunting capacities Ct. This is evident from Eq.
Another type of high -frequency compensation is
7-12(A).
known as series peaking. This type of circuit is
As an example, suppose that the shunting capa-
shown in Fig. 7-8 (A) and (B). Here the peaking
cities of a stage such as is shown in Fig. 7-6(A)
coil is placed in series with the plate of the ampli-
are found to be equal to 24 µµf measured at a fre-
fier, instead of in parallel with it. This type of
quency of 3 megacycles. We must now determine
compensation is used when higher gain and more
the value of the shunt peaking coil L, and the value
linear phase shift are desired than a shunt com-
of the plate -load resistor Rb needed. We know that
pensated circuit provides. (In the previous case the
for flat response at the high frequencies, the plate -
phase shift is also compensated but not in a linear
load resistor Rb must equal the reactance of the
fashion.) Fig. 7-8(A) shows a typical "series peak-
shunting capacities at the maximum frequency to
ing" high -frequency compensated video amplifier
be amplified (which then becomes the correction
circuit. In this type circuit the peaking coil L8 (it
frequency). Therefore, with fc = 3 megacycles:
is still called a peaking coil because it still "peaks"
1 1 up the high -frequency end of the response curve)
Rb
274,Ct - 27r (3 X 108) (24 X 10-12) is inserted in such a way that it isolates the shunt-
= 2210 ohms ing capacities related to both tubes. C, represents
and for the peaking coil L,, from Eq. 7-12 (A) and those shunting capacities relative to tube V, and
(B): C2 represents those shunting capacities relative to
Ly = 72 CrR2° tube V2. The combination of C1, L8 and L2 repre-
= y, X 24 X 10-12 X 2210 sents a "filter" network which has the character-
= 58.6 microhenrys istics of a low-pass filter (wide -band). That is why
or series peaking compensation is sometimes known as
"filter" coupling. By a "filter" we mean a circuit peaking depended upon the total capacity Ct. There-
so designed that it will pass certain frequencies and fore, the lower the value of C1, the higher the value
attenuate others that are undesired. Rb can he, and the greater the gain, since the gain
Variations in the circuit of Fig. 7-8(A) are depends upon the size of Rb. It then follows, from the
sometimes made in order to isolate the shunting ca- above discussion, that both the values of C, and Cy
pacities at some specific ratio. For instance, in Fig. have to be estimated quite closely for the proper de-
7-8 (B) the series peaking coil Ls is shifted to the sign and use of the plate -load resistor Rb and the
other side of the load resistor because in some cir- series peaking coil Ls.
cuits the shunted capacities are more evenly dis- The addition of the series peaking coil helps the
tributed under these circumstances. While in Fig. high frequency response in the following way [re-
7-8(C), the coupling capacitor Cb is placed on the ferring to the typical circuit of Fig. 7-8 (A) ] The :
left hand side of the peaking coil L8, so that the voltage output from tube V, is impressed across
shunting capacity between the capacitor Cb and the plate -load resistor Rb. Assuming no compensa-
ground will be added to C, and not to C= as in Fig. tion (Ls removed), then this voltage across Rb
7-8 (A). In all of the above cases, the series peak- would be attenuated at the high frequencies due to
ing coil is inserted in such a way that the shunting the effect of the total shunting capacity. With peak-
capacities are divided on either side of the coil with ing coil Ls inserted, the total shunting capacities are
the ratio between C, and C, that is most beneficial divided, with the lesser amount going to C1. The
to the circuit. For most practical purposes and in previous loss which was due to the effect of the
our discussion C2 is equal to twice C,. In practice total shunting capacities is now greatly reduced by
the load resistor Rb is usually placed on the side of the effect of C, alone. This output voltage from tube
the lower shunting capacity, namely C,, however, V1, that is now developed across Rb and C1, in par-
[as seen in Fig. 7-8 (B) ] this is not always done allel, is impressed on the voltage dividing network
since the capacities may'be so nearly equal that it of Ls and C2. Cb presents a short circuit at high
makes little difference. Since the total shunting ca- frequencies and the value of the grid -leak resistor
pacities are isolated by the coil Ls, the value of the R, is so large compared to the reactances of L, and
load resistor Rb in most cases depends upon C1; C2 that they (Cb and R,) can be considered as
whereas the value of the load resistor Rb in shunt having negligible effect on the voltage dividing net -
(A)
(B)
(D)
FIG. 7-8.-A typical series peaking high -frequency compensated video amplifier circuit is shown in (A) and a simplification
of this circuit in (D). The circuits in (B) and (C) are variations of (A), these being sometimes used to isolate the shunting
capacities at some specific ratio.
work L, and C,, and are consequently neglected in horizontal dashed curve 3 is the 3 db curve where
the- circuit analysis at high frequencies. Even though the gain drops to 70.7% of its maximum value.
the shunting capacity across tube V, is reduced (Keep in mind that this 3 db value is the point of
from the total shunting capacity Ct to just the minimum gain allowable for most proper designs.)
shunting capacity of C,, there still remains some Assume that we want to have a flat frequency re-
loss at high frequencies due to this C1. The prob- sponse characteristic to a correction frequency fc
lem now is to offset the high -frequency attenuation of 3 mc. Then, from curve 1 we note that at this
caused by C,. This can be done by offering an im- frequency the gain is at point (a) and therefore
pedance to the output voltage of tube V, at high undesirable. We want to bring point (a) up to
frequencies that is considerably lower than that rep- somewhere in the shaded area. Ideally we want to
resented by the RbC, combination, so that most of bring point (a) up in such a manner that curve 1
this signal output voltage will follow the path of is raised at the high frequencies so it has a relative
low impedance and very little will be shunted to voltage gain of unity. In other words, even though
ground through C,. upon correction, point (a) may advance anywhere
The question that now arises is how this effect into the shaded area (the area of allowable gain)
can be attained. Referring to Fig. 7-8(D), which between points (b) and (c), it is preferable to ap-
is a simplified circuit of Fig. 7-8 (A) ; at some fre- proach point (c). By actual experimental determi-
quency, call it f°, we notice that L, and C2 are in nation, it has been found (when C, = 2C1) that
series resonance. Therefore, at this frequency the this correction is ideally approached when the ele-
impedance offered by the series resonant circuit of ments L, and C2 are made to resonate at the cor-
L,C, to the output of tube V, will be a minimum, rection frequency fO. Under such conditions curve
and most of the output signal voltage from tube 1 is shifted to curve 2. From this curve it is evident
V, will follow the path of minimum impedance of-

fered to the tube, namely L,C,, and very little will
that at the correction frequency f
there is a peak
rising slightly (which is due to series resonance be-
be shunted to ground by the higher impedance of tween L, and C2) above the level of unity gain; but
C1. The signal current flowing from tube V, will in general the shape of the curve approaches the ideal
then flow through the voltage dividing network of case very closely. We therefore have two ways to ex-
L, and C,, and the maximum amount of voltage plain series peaking compensation-one by the ex-
(at high frequencies) will be impressed on the grid planation of the actual response curve and the other
of tube VQ by the voltage drop across capacitor by the theoretical explanation of how the low im-
Cp. It is, therefore, seen how capacitor C, offsets pedance path clue to L,C, helps the gain at high fre-
the drop in gain at high frequencies by the charac- quencies.
teristics of its resonant effects, at these frequencies, The formulas for the design of the series peak-
with the series peaking inductance L,. In actual ing coil L, and the plate -load resistance Rb depend
practice, the ratio between C, and C, is set approxi- primarily on the value of the frequency desired for
mately equal to two, as stated before. The values correction. Calling this frequency, as we have be-
of C, can be measured pretty accurately and so can fore, the correction frequency f0, we know now
the total capacitance Ct (Ct = C1 + C2), and from that L,C, should be made to resonate at this fre-
knowing C, and Ct, Cs can be found. Circuit com- quency, whereupon we have :
ponents may have to be shifted around in order to

find the approximate ratio.
In working with design formulas for the series
peaking inductance and also for the load resistance,
the operation of the series peaking circuit of Fig.
7-8(A) may be better understood by discussing
typical response characteristics at the high end of
a curve-such as shown in Fig. 7-9.
In this figure the gain is only relative and its
FREQUENCY
maximum value is taken as unity at the intermediate (HIGH FREQUENCY ENO OF THE CURVE)
frequencies.
Curve 1 represents the actual curve of the ampli-
fier without any compensation. It is readily seen 7-9.-Typical response characteristic curves, wherein
the voltage gain is relative, its maximum value being taken
how this curve falls off at the high frequencies. The as unity at the intermediate frequencies.
Knowing that the total shunting capacity Ct is
1 h,g. 7-13 (A) equal to the sum of C, and C2 and that
f. r
2 VL. C2
C1 = s
where the shunting capacity C2 is known, by meas- then subst. Ming for C,, Ct becomes :
urement as discussed beforehand or by close ap- 3

Ct=
proximations. Squaring both sides of the equation 22+ C2= L2
above and solving for L, we get : and solving this equation for CQ we get :
C, = 2/3 Ct
1 Eq. 7-13 (B) and substituting this equation in Eq. 7-15(B) we
L,
(2 7 fc) 2 C2 get :
and also knowing that C2 = 2C, and substituting L,

3 Eq. 7-13 (G)
this in Eq. 7-15(B) we get :
u fo)2 2 et
rearranging we get :
1 Eq. 7-13 (C)
L, 3
ir fß)2 2C,
2 Ir fc L. = . 2 T1f8 et
From Eq. 7-15(B) and (C) we can readily solve
for the series peaking inductance in terms of either or Xt, = 2 X,t = 1.5 X,t Eq. 7-13 (H)
shunting capacity C, or C2. We can rearrange Eq.
which states that the inductive reactance of the
7-13(C) in the following manner :
series peaking coil at the frequency of correction is
equal to 3/2 the reactance of the total shunting ca-
(2 f) L, - 1
2 (2 r .4)
1
C,
Eq. 7-13 (D)
pacity Ct at the frequency of correction.
In determining the value for the inductance of
which states that the inductive reactance of L, at the peaking coil we can utilize any one of the above
the frequency of correction is equal to 1/2 the ca- eight equations [Eq. 7-15(A to H)] according to the
pacitive reactance of C, at the same frequency. parameters that are known. These equations make
If we rearrange Eq. 7-13 (C) in another way by use of the fact that in order to solve for L, all that
solving the equation for the value, V2
following:
we find the f has to be known are any two of the following ele-
ments that fit into the particular formula used :
Taking the square root of both sides of Eq. 7-13 Ct = total shunting capacity in the circuit
(C) we have : (C, Cs)
C, = shunting capacities around the output of the
2 lift ,/' first tube

C, = shunting capacities around the input of the
second tube
and solving for V2if, we get :
= frequency of correction (or that frequency
fe
1 Eq. 7-13 (E) where L, and C, will be in series resonance)
fr = that frequency which is V2 times the fre-
2 r 1/L, Cl
quency of correction (or that frequency
Referring to equation 7-13 (E). it will be noticed where L, and C, can be considered to be in
that the right-hand side of the expression is itself resonance)
representative of a resonant frequency. In other Now that we have decided on numerous ways
words, a frequency exists at which the series peak- for determining Lwe are ready for the proper de-
ing coil L, is in resonance with the lower valued termination of the plate -loading resistor Rb in these
shunting capacity C,. The resonant frequency of series peaking circuits.
L,C, is then equal to V2 times the frequency of From all other previous topics we have deter-
correction. That is to say, if we let fr equal the mined that the plate -load resistance Rb should be
resonant frequency of L, and C,, then equal to its shunting reactance for proper design
fr = V/2 fc Eq. 7-13(F) purposes. In shunt peaking, this reactance was the
Eq. 7-13(F) means that there exists for L, and total shunting reactance X, in the circuit.
C, a resonant frequency (f,.) that is equal to V2 Therefore, for design purposes Rb was set equal
times the frequency of correction. to 1/Xct at the frequency of correction, which was
the minimum acceptable 3 -db frequency. One should Some fundamental properties of the series peak-
understand that these assumptions are not just ing circuit are that it has 50 percent more gain
guesses but are based upon actual experimental than a shunt peaking circuit (compare Eq. 7-14
data taken on such amplifier circuits. When the re- with Eq. 7-9) where both circuits have the same
sponse curve of Fig. 7-9 was plotted, it was found values of Rb, fe and C1; and also that the phase
by experimentation that the plate -load resistor Rb shift response in series peaking is more linear than
was approximately equal to one-half the reactance in shunt peaking. Thus it is seen that series peak-
of the shunting capacitor C, at the frequency of ing betters the gain and phase characteristics of the
correction. Accordingly, the basic design formula amplifier at the high -frequency end of the curve as
for the load resistor Rb for series peaking is : compared to shunt peaking.
1 1 Eq. 7-14 (A) To further better the amplification and phase
Rb = 2XCl = 2 (2 a ff C,) shift characteristics, two such series peaking coils
are put into a circuit as is shown in Fig. 7-10. This
Since it is easier to measure the total shunting TAPPED PORTION
capacities Cg than C, alone, we shall rearrange Eq.
7-14(A) in terms of Cg instead of C, by substituting
-i_
C, = (since C2 = 2C1) Ì Ci
3t Rb
R
_1
C3
Hence
Eq. 7-14 (B)
Rb 22afCt 2X, = 1.5 Xe,

Eq. 7-14 (B) states that for proper series peaking
compensation, the plate -load resistor' Rb is made
equal to one and one-half times the total shunting
reactance Xe, at the frequency of correction (when 7-10.-In order to improve amplification and phase -
C2= 2C1). shift characteristics, two series peaking coils are employed,
thus bettering the response characteristics that one such coil
We also can combine the relation between the provides.
equations 7-13(D) and 7-14(A) in such a manner
that L, and Rb are in the same equations. Substitut-
ing Eq. 7-14 (A) into Eq. 7-13(D) we get: type of circuit improves the response characteristics
that one such series peaking network gives and also
X ta = Rb provides some other terminals from which extra
and substituting 2irfcL for XLs we get :
signal energy may be derived for uses such as
synchronization.
2irfcL, = Rb SD far we have studied two separate high -fre-
where quency compensating networks, namely shunt peak-
ing and series peaking. The latter type of peaking
L.=2 Rbfe Eq. 7-15 (A) was found to give a better response and phase shift
than the former, but a combination of these types
Solving Eq. 7-14(B) for 27rfc we get: gives a better response than either one alone. Fig.
2af = R 3 7-11 illustrates a typical circuit combining the effect

of series and shunt peaking with the shunting ca-
pacities C, and C,, shown. With this type of com-
and substituting this equation in Eq. 7-15(A) we pensation there is a sharper cutoff at the correction
get: frequency than in the series case alone. Much great-
Eq. 7-15(B) er gain is obtained than when shunt or series peak-
2 ing alone is used. It has been found that combina-
L,
3/(2Rb
C, Rb)
- 3
C, (Rb)2 = .67 Cr (Rb)2 tion peaking gives as much as 80 percent more gain
than a shunt peaking system alone. This means that
From the above equations it is seen that we can the total effective load impedance is also 80 percent
solve for the plate -load resistance Rb by knowing greater than that for shunt peaking.
the frequency of correction f, and any one of the So far we have discussed the separate types of
following quantities-C,, Cr, or L3. frequency compensation (high and low) which may
i

be applied separately to any system desired. But in low -frequency compensated case (see Fig. 7-4).
video amplifiers, as previously pointed out, the de- One of the fundamental actions of capacitors CA,
sire is to have as flat a response as possible from C8, and C'b is to bypass low or audio -frequency
the lowest audio f requency encountered to the high - components to ground and, therefore, prevent these
Ls c, frequencies from getting into certain parts of the
circuit where they are not desired. For instance,
feedback is prevented by such bypassing as was
previously discussed. The values of these capacitors
CA., C,,i and C'b are hence made large so that their
reactance to low frequencies will be small. That is
fine for low frequencies but when the low -frequen-
;
cy compensated network is combined with the high -

frequency compensated network, a separate path to
bypass the high frequencies must be provided to
prevent feedback. Two separate bypassing paths are
thus needed, one for low frequencies and the other
FIG. 7-11.-A typical circuit combining the effect of series
and shunt peaking with the shunting capacities Cl and CQ for the high. We already have the low -frequency
indicated across the coil L8. bypass capacitors in the circuit, and since bypassing
of the high frequencies is necessary, then all that
est video frequency needed. It is evident that to has to be done is to shunt capacitors CA., C8, and
obtain these qualities in a single circuit, the low- C'b with the other capacitors C'A, C'8, and C"b, re-
f requency and high -frequency compensated net- spectively. Since these latter three capacitors are
works have to be combined. Fig. 7-12 illustrates a needed to bypass the high frequencies they are much
number of different video amplifiers in which both smaller than the former ones. Some typical values
low- and high -frequency compensation are employ- employed in practice are as follows :
ed. An interesting feature of these circuits is the C,, = 100µf C, = 16µf C'b = 30µf
additional capacitors used. C'k= 0.1µf C',4=0.01µf C"b=0.1µf
In all the circuits it will be noted that the capaci- In visualizing these capacitors one must remem-
tors C'A, C'8, and C"b are shunted across the capaci- ber that their size depends not only on the capacity
tors CA. C8, and C'b that were originally used in the value but also on the voltage rating. For example,
.s
V2
LS Cb
FIG. 7-12.-Three different video amplifier circuits in

which both low- and high -frequency compensation are em-
ployed. It will be noted that two separate bypassing paths
are provided-one for the low frequencies and the second
for the high.
the 100 µf cathode bypass capacitor need only have in the proper ratio to compensate for the falling off
a small value of breakdown voltage depending on due to the coupling circuit to the next stage. This
the voltage drop across the cathode resistor Rk. compensation occurs because a higher load imped-
Assuming the bias on the tube to be -10 volts, ance in the plate circuit of the 6AU6 results in a
then the drop across the bias resistor Rk is 10 volts greater output signal.
and for a factor of safety the cathode bypass capa- The values of resistance and capacitance are also
citor Ck (and likewise C'k) usually has a voltage chosen to make the phase shift linear with respect
rating of 25 or 50 volts. Such a capacitor of 100 µf to frequency. The value of R141 is limited by the
has to be of the electrolytic type and it is not very drop in voltage across this resistor, which decreases
large, since the breakdown voltage is to be 25 or the plate voltage on the 6AU6. It is interesting to
50 volts. note the time constant of the R141-C223B combi-
nation. This time constant is equal to 6800 X 10-5 or
RCA Model 630TS 0.068 second. Referring to Fig. 7-5 which shows
the amplitude response for this type of low -fre-
The first video amplifier of the RCA model 630TS quency compensation and interpolating between the
receiver is shown in Fig. 7-13. The coupling net- curves shown, it can be seen that under ideal con-
work between the first video amplifier and the second ditions the response will he flat to about 10 cycles,
video amplifier employs both low- and high -fre- which is well below the lowest frequency (30
quency compensation to produce a flat response and cycles) that has to be passed.
linear phase shift over the entire range of video The high -frequency combination employs both
1sr VIDEO 1199 2.0. VIDEO series and shunt peaking. This is done so that the
6AU 6 6K6 -GT
V 11 5 1a0µ11
C140
V116 response at the high frequencies will be better than
.05µf if only shunt or series peaking were used. The
shunt peaking used in this circuit is similar to that
1.190 shown in Fig. 7-6(A). The peaking coil L190 is
93µN
equivalent to the peaking coil L in the previous
17141
discussion. This coil resonates with the shunt capa-
R140 R143
+135
R14I
6800
3300
I 1.1M
C223B
to µc
820K citance across the input circuit of the 6K6. This
includes the input capacitance of the 6K6, about
5.5 µµf, plus all the stray capacitance due to the
-18V
+175
wiring. and the parts used in this circuit. It is diffi-
cult to estimate this capacitance as it varies quite a
Ftc. 7-13.-The first video amplifier circuit of the RCA
model 630TS receiver. bit, depending on the method of wiring used, the
Courtesy RCA physical size of each component, its proximity to
ground, etc. In the actual design of a circuit like
frequencies, approximately 30 cycles to 4 mega- this, the shunt capacitances are actually measured
cycles, The low -frequency compensation is accom-
with an impedance bridge or similar instrument and
plished by resistor R141 and capacitor C223B in then the correct values of the peaking coils and the
the plate circuit. This network corresponds to R'b plate -load resistance are determined from the de-
and C'b in the previous discussion of low -frequency sign equations that have been discussed.
compensation. The plate -load resistor R140 cor- The series peaking coil L189 is equivalent to the
responds to Rb, the coupling capacitor corresponds peaking coil L, in our previous discussion of series
to Cb, and the grid -load resistor R0 would be the peaking. The output capacitance of the 6AU6,
equivalent resistance of R142 in parallel with R143, which is about 5 µµf, plus the stray capacitance in
which is 487,000 ohms. If the compensating circuit this circuit correspond to the shunting capacitance
were not used, the gain at low frequencies would C1, and the input capacitance of the 6K6 plus strays
fall off due to the effect of the C140-R142, 143 correspond to the shunting capacitance C, in Fig.
network. The reactance of capacitor C140 goes up 7-8 (B) The coil resonates with these shunting capa-
.
as the frequency goes down, causing less of the out- citances to produce a peak at the high end of the
put voltage of the first stage to appear at the grid frequency range. This peak, together with the peak
of the second stage. To correct for this, R141 and caused by the shunt -peaking coil, is sufficient to flat-
C223B were chosen .so that the impedance of this ten the response up to the highest video frequency,
combination would increase with lower frequencies about 4 megacycles. The 22,000 -ohm resistor, R139,
is placed across the peaking coil L189 in order to cal values on one side of the zero -voltage axis. The
reduce the effective Q of the coil. It was probably video signal must therefore contain both a d -c and
found that the resonant peak produced by this coil an a -c component in the same way that a rectified
in conjunction with the shunt capacitance was too a -c voltage representing the output of a power sup-
high. To get an over-all frequency response that is ply contains a d-c component-the d-c voltage of
fairly flat over the entire range, it was necessary to the power supply-and an a -c component-the hum
reduce this peak. This can be done by damping, or ripple of the power supply.
lowering the Q of the resonant circuit, and is ac- Physically what does it mean to say that a video
complished in this particular circuit by placing the signal contains a d -c as well as an a -c component ?
resistor R139 in shunt with the series -peaking coil, Actually it is only another way of looking at the
L189. In a circuit that uses both series and shunt signal, as Fig. 7-14(A) clearly shows. In this figure
peaking, the shunt -peaking and the series-peaking we describe the signal by saying that the light
coils may both be used to peak at the same frequen- values are represented by fluctuations in both di-
cy if the loss at some particular high frequency is rections from an average level, which we can call
quite high. Or, the resonant frequency between both the "average brightness," or the "picture back-
circuits may differ somewhat, so that their resonant ground." Thus instead of describing the light value
peaks may be slightly apart. This results in an in- at any point in the scene by stating how much
crease in voltage gain at a wider range of high brighter it is than black, as in (A), in (B) we ac-
frequencies than if both coils are peaked at the complish exactly the same thing by stating how
same frequency. much brighter or blacker is the particular light value
than the average brightness. In (A) we use the
black level as the reference level whereas in (B)
AVERAGE BRIGHTNESS AND we use the average brightness as the reference level.
D -C RESTORER CIRCUITS Electrically speaking, the average brightness level
In our previous discussion of the video signal, represents the average value of the video signal, or
we looked upon this signal as containing a series of in other words the d -c component of the signal.
voltage values each of which corresponded to a par- On the other hand, the fluctuations in the signal on
ticular value of light intensity in the televised either side of the average brightness level, which is
image. We considered that in the same way that electrically the a -c axis of the signal, represent the
light values can be reckoned from black as a ref- a -c component of the signal. An important charac-
erence level, so a particular voltage value can be teristic of the average brightness or a -c axis is
assigned to black and then each light value repre- that the area between the positive part of the cycle
sented electrically by assigning a higher (or lower) and the a -c axis is equal to the area included be-
voltage to the signal, depending upon the brightness tween the negative part of the cycle and the same
of the light value at the scanned area. This method a -c axis. This is shown in Fig. 7-14(B), where it
of looking at a video signal is shown in Fig. 7-14 can be seen that area 1 is equal to area 2 + area 3.
(A), in which it is clear that all light values are From the preceding it will be clear that proper
with reference to the black level-which is taken as operation of the picture tube requires that both the
the zero -voltage axis. d -c and a -c components of the video signal be passed
Insofar as video amplifier operation is concerned, by the video amplifier. For if only the a -c component
the important thing about Fig. 7-14(A) is that it reaches the grid of the picture tube, then the picture
shows that a video signal is inherently a pulsating tube will have information only on the fluctuations in
d -c voltage. In other words (as in the case of a light values with reference to the average brightness
rectified a -c voltage which is also a pulsating volt- level but it will have no information whatsoever on the
age) all the light values are represented by electri- value of this average brightness level. Thus the pic-
t
"
Wh/TE WN/ TE 'F VOL 7,5
YOL75
GRAY- GRAY
AVERAGE
BR/GHTNG55
AX/.5
L_ VOLTS
GLACK O v04715
\ AERA6L
s/r/ÇN7NGt5
t _VOLTS
BLACK
(A) ((3)
Fm. 7-14.-The light values in a video signal can be reckoned in two ways as here illustrated: in (A) the various light
values are referred to the black level; in (B) the same light va lues are referred to the average brightness of the signal.
ture cannot be reproduced accurately since the same In Fig. 7-15, the picture being scanned is a white
variations (represented by the a -c component) might triangle on a black background, with the vertex of
be superimposed on either a dark or a light back- the triangle near the top. Part (A) shows two lines
ground of any shade. The average brightness, or the scanned near the top of the picture; part (B), near
d -c component of the video signal, must be present the middle of the picture; and part (C), near the
before the picture can be reproduced. bottom. Thus (A), (B), and (C) represent three
sections of the picture where the average bright-
ness is low, medium, and high respectively. Fig.
D -C Restorer Circuits 7-15 can be said to represent the signal as it is at
Unfortunately, it is not possible to transmit the the output of the camera tube or as it appears at
d -c component of any signal without using a direct - the video second detector. Of special importance is
coupled amplifier. However, the use of direct -coup- the fact that all the blacks are lined up so that
led amplifiers is not practical in television receivers the d-c component of the signal is represented in
because of their comparative instability and high all cases. Note that the d -c component, like the
cost. For this reason, television engineers have de- average brightness which it represents, is succes-
veloped circuits which make possible the use of con- sively larger in (A), (B), and (C).
ventional a -c amplifiers with capacitor coupling and Now what happens to this signal when it is pass-
at the same time provide for the restoration or re- ed through an a -c amplifier which contains coupling
creation of the d -c component after the a -c com- capacitors and, therefore, will not pass the d -c com-
ponent has been amplified by itself. ponent? This is clearly illustrated in Fig. 7-16
It will be helpful at this time to consider several which shows that the blacks are no longer lined up,
video signals which have different values of average but instead the separate average -brightness or a -c
brightness and to describe the action which takes axes are all lined up. In other words, when the d -c
place when these signals pass through an amplifier component is lost, only the fluctuations on either
which employs a blocking capacitor as a coupling side of the average brightness are transmitted. and
element between stages. It is because of the pres- this axis must of necessity be the zero -voltage axis
ence of the blocking capacitor that only the a -c in all instances because no d -c component can get
component is passed and the d -c component is lost. through the amplifier.
s. A
8
C DC COMPONENT_
OCCOMPONENT
AVERAGE BRIGHTNESS.) (AVERAGE BR/GHTNESS)
A I C
f WHITE
AC
-AX/5
VOLTS
BLACK-u -- jv Ü Ü
WH/ re B C
X
i -o --- - ------ÿ
BLACK-u ---
Ü- -L1-7 Z
FIG. 7- /6
FIG. 7-15.-The video wave produced when two lines are scanned at A, B, and C of the white triangle on the
black back-
ground that is shown above the waveforms. Since the black level is the same for all three cases, the signal contains the d-c
component.
FIG. 7-16.-These are the waveforms of the three signals of Fig. 7-15 after passing through an a -c
amplifier. Note that
the d -c component has been lost so that black in the signal no longer corresponds to a fixed voltage level as it does in Fig. 7-15.
Having investigated the video signal both with ner by using a diode to rectify the "black" half of
and without the d -c component, let us now examine the a -c video signal (which contains the sync sig-
the manner in which the picture tube is affected by nal) and in this way the required voltage is pro-
the presence or absence of the d-c component. In duced. Thus in Fig. 7-16 at (A) this rectification
Fig. 7-15, we see that black in every case corre- produces the voltage x; at (B) it produces the volt-
sponds to the same definite voltage value and this age y; and at (C) it produces the voltage z. In gen-
is also true for white and every intermediate shade eral, the addition of this varying d -c voltage lines
between black and white. Thus when the voltage of up all the pedestals to produce the original signal
Fig. 7-15 is applied to the grid of the picture tube, (Fig. 7-15) with the d -c component restored.
the picture will be reproduced without any distor-
tion.
This, however, is not true of the signal in Fig. Basic D -C Restorer Circuit
7-16, from which the d -c component has been re- In Fig. 7-17 is shown a straightforward circuit
moved. Black no longer corresponds to the same which is used to restore the d -c component in the
value in all instances, but instead the signal voltage video signal before the signal is applied to the con-
associäted with black takes on a value which is en- trol grid of the picture tube. The video signal is
tirely dependent upon the average brightness of the developed across the 3000 -ohm plate resistor R1
strip being scanned. In the same way, this figure and fed to the control grid through a 0.1 µf coupling
shows that white, and every intermediate shade as capacitor Cl. As the sketch shows, the signal has
well, has a different voltage value which is also de- r n n
pendent upon the average brightness of the strip

being scanned. A little reflection will show that this (oc REIrORCO.)
gives rise to serious distortion because the same light
value in different parts of the picture does not corre-
spond to the same voltage value in the signal. Thus,
for example, the grid of the picture tube receives a
different voltage for the saute shade of gray in (A),
(B), and (C) of Fig. 7-16 and as a result this shade
is reproduced differently in the three instances.
In order to remove this source of distortion, it
is apparent that the average brightness or the d -c 7-17.-The heavy portion of the circuit shows a typical
d -c restorer circuit arranged to restore the d -c component to
component must be restored. At first glance this the video signal before it is applied to the control grid of the
seems impossible, for how can the d -c component picture tube.
be restored after it has once been lost in the video
amplifier? As a matter of fact, it is generally not the required positive polarity both at the plate of
possible to restore the d -c component of a pulsating the video amplifier and the grid of the picture tube.
wave after the d -c component has been lost in trans- However, the d -c component is not present at the
mission, because no information relative to the d -c plate side of Cl, whereas the diode circuit shown in
component is contained in the a -c component; the heavy outline has restored the d -c component on the
two are entirely independent of each other. Fortu- picture -tube grid side.
nately, however, the sync pulse is transmitted at Let us see just how the diode circuit restores the
the end of each line, and the pedestal on which the d -c component. Considering the series circuit com-
sync pulse stands corresponds to a definite black posed of the plate -load resistor R1, the 0.005-µf
reference level. These facts make it possible to re- capacitor C2, and the diode in shunt with a 1-meg
store the lost d -c component. Thus to restore the d -c resistor, we can see that the following takes place:
component to the signal of Fig. 7-16 it is only neces- On the positive half of the cycle, the cathode of the
sary to modify the signal so that all the synchroniz- diode is swung positive with respect to its plate so
ing pulses are lined up. When this is done, the sig- that no current flows in the diode circuit. On the
nal in Fig. 7-16 is exactly the same as that in Fig. negative half of the cycle, however, the cathode is
7-15 and the d -c component has been completely re- swung negative with respect to the diode plate, ánd
stored. current flows in the diode circuit through R3. It is
The solution of the problem depends upon find- this current flow which charges C2 and makes the
ing this varying d -c voltage and adding it to the cathode end of R3 positive, and as a result provides
signal. This is accomplished in a very simple man- the d -c restoring bias.
Note that the d -c voltage across R3 satisfies all shades of gray and white will be reproduced cor-
the conditions required to line up the sync pulses rectly.
so as to restore the d -c component. Essentially the Fig. 7-18 shows the illumination characteristic of
action in the diode circuit is such that the negative a picture tube and the manner in which the bright-
part of the video wave is rectified and the diode ness of the scanning spot depends upon the bias
current charges C2 positively to a value equal to
the amount by which the sync pulses in (A) are
depressed below the a -c axis. Since the grid of the
picture tube is returned to this d -c voltage through FIG. 7-18.-The effect of
bias voltage on the brightness
the 50,000-ohm resistor R2, the d -c voltage is added of the image. The brightness
to the video signal at (A) and "raises" the pedestal control should be adjusted so
that the black level on the sig- rocrwcc-
GCACK
so that it lies along the zero -voltage axis, as shown nal occurs approximately at 4-11,As
o
at (B). the cutoff of the picture tube
characteristic.
Now suppose, as in Fig. 7-16 (C), that the aver- .c..e.
GcA<w
age brightness is higher than that shown at (A) in

Fig. 7-15. This results in the pedestal being more
negative with respect to the a -c axis, so that the voltage on .the control grid. Referring again to
diode produces a more positive voltage which again Fig. 7-17, it will be observed that the cathode of
raises the pedestal to the same zero -voltage axis as the picture tube is returned to a potentiometer which
at (B) in Fig. 7-15. Thus the action is entirely makes it possible to vary the bias voltage from 0 to
automatic so that all the pedestals are lined up at 90 volts.
the grid of the picture tube, regardless of the value When the bias is highly negative, the tube is cut
of the average brightness. off and the spot intensity is zero. As the bias be-
It is important to understand that the voltage comes more positive, the intensity of the spot in-
produced at the cathode varies constantly through- creases. For correct operation, the brightness con-
out the scanning and that its value at any time de- trol should be adjusted manually so that the ped-
pends upon the average brightness of the portion estals occur at the black or cutoff points. Once the
of the picture being scanned at that particular time. brightness control has been set, the d -c restorer cir-
The time constant (RC) of the diode circuit is de- cuit automatically keeps the pedestals in alignment
signed so that C2 will not discharge appreciably so that no further adjustment is required.
during the interval between successive sync im-
pulses. At the same time, the time constant is suffi- Grid Leak Capacitor Restorer
ciently small so that when the average brightness
changes, the capacitor is able to change its charge Another type of d -c restorer circuit which is very
rapidly enough to respond to the new conditions. widely used is shown in Fig. 7-19. In this circuit
Insofar as the grid of the picture tube is con- the output video amplifier .tube is operated at zero
cerned, it receives the a -c component of the video bias and the grid -cathode elements are used as a
signal through Cl, and only the d -c component diode to insert the missing d -c component. To pre-
through R2. Although the a -c component is also vent the loss of the newly restored d -c component
present at the cathode of the diode, R2 acts as a fil- which would occur if capacitor coupling were used
ter resistor to prevent that portion of the video in coupling the plate to the picture tube, the plate
signal present at the cathode from reaching the pic- is coupled directly to the picture -tube grid.
ture -tube grid through R2.
--J---
P/NA L C°I
(A) V/Ceo AMPI.
(e)
Brightness Control vierURE TUGSC
Y/OEo 'I C/
J/GNAL
We have just seen how the d -c restorer circuit R2
automatically lines up all the sync pulses so they are 1 1
at the same voltage level. For correct operation of 300 BRIGHTNESS
12GO
the picture tube, the bias on the picture tube must CON TROG
be so adjusted that these aligned pedestals occur at FIG. 7-19.-A widely used type of d -c restorer circuit in
the cutoff or black level. The sync pulses will then which the video amplifier output tube is operated at zero bias
so that the grid -cathode elements f unction as a diode to re-
lie in the blacker -than -black region and the various insert the d -c component.
From the explanation already given of the proc- tween the plate and the grid of the picture tube,
ess of d -c restoration, it is easy to understand how the sync pulses remain in alignment.
circuits of this type operate. The waveform of the As in the previous circuit, provision must be
video signal on the left side of CI is shown at (A) ; made for initial adjustment of the bias of the pic-
at this point the signal has a negative polarity and ture tube so that the pedestals of the signal will
of course the d -c component is missing. When this occur at cutoff (black level) on the picture -tube
a -c signal is impressed on the grid of the output characteristic. To accomplish this, the cathode of
tube through Cl, the positive parts of the video the picture tube is returned to a bleeder -potentio-
cycle (including the sync pulses) place the grid meter which makes it possible to place up to a maxi-
positive with respect to its cathode so that grid cur- mum of +300 volts on the cathode. Since the grid
rent flows through Rl. The direction of this grid - can at most have a potential of +250 volts, this
current flow is such that the grid becomes negative makes a bias of -50 volts available on the grid of
with respect to ground. the picture tube. Under actual operating conditions,
When the average brightness of the signal is the voltage drop across R2 makes the grid more
small, the a -c axis of the signal will be near the negative; this is compensated for by making the
black level, the positive peaks will be small in am- cathode of the picture tube less positive by adjust-
plitude, only a small value of grid current will flow, ing the brightness control R3.
and consequently the grid will have only a slight When the receiver is first turned on, there is no
negative potential on it. On the other hand, when voltage drop across the plate -load resistor R2 until
the average brightness is great, the a -c axis is away the output video tube warms up and draws plate
from the black side, the positive peaks will be large current. The final adjustment of the brightness con-
in amplitude, a relatively large flow of grid cur- trol therefore should not be attempted until the re-
rent will take place, and the grid will be made high- ceiver has been turned on for a few minutes. In
ly negative. In each instance, the grid will be made particular the brightness control should be all the
more negative by an amount equal to the height of way to the left (with the rotor at +300) until the
the sync pulses above the a -c axis of the wave. As receiver has warmed up; this is done to avoid pos-
a result of this action, all of the pedestals in the sible damage to the picture tube because of exces-
signal are depressed by an amount sufficient to align sive beam current.
them to the same zero -voltage level at the grid. This
is shown by the signal waveform at (B). Essen-
Video Output of Belmont Model 21A21
tially the action here is the same as that previously
described for the circuit using a separate diode. The video output circuit of the Belmont 21A21
The capacitor C1 performs the same function of television receiver, as shown in Fig. 7-20, is an
storing the grid charge as in the previous method. example of the grid leak -capacitor type of d -c re-
At the plate of the output tube, the polarity of storer. The d -c restorer in this circuit is the same
the video signal is reversed and as (C) shows, the as the basic circuit shown in Fig. 7-19. The nega-
signal at this point has the required positive polar- tive bias on the grid of the 6AI-16 will be deter-
ity. Since no blocking capacitors are interposed be- mined by the height of the sync pulses above the
6ANe R51
TO VIDEO
X OUTPUT 33K
PI
DETECTOR R55
T13 4300 16P4 7-20.-The video out-
C66 gRIûHTNE55
PICTURE put circuit of the Belmont
TUBE
model 21A21 is an example
R52 R57 of the grid leak -capacitor type
1M
3300 250K of d -c restorer, which is the
R53 same as the basic circuit
6800 shown in Fig. 7-19.
C44 -C
CONTRAST R49
R45
2000
470K C71
470
R56
1500 R58
I00K
11
C72
.1µ;
g+
a -c axis. This will line up all the sync pulses at R152 is adjusted for the correct operating bias of
zero level and restore the d.c. exactly as explained the 10BP4. This is about -50 volts. This voltage is
with regards to the basic circuit in Fig. 7-19. The applied to the plate of the diode (1/2 of a 6AL5)
time constant of the C66 -R49 combination is equal through resistors R151 and R150. The cathode of the
to 470,000 X 10 -' or 47,000 microseconds. The time diode is connected to the top of the plate-load re-
between successive horizontal sync pulses is equal sistor of the 6K6, R147, through the series combi-
to the reciprocal of the horizontal sweep frequency nation of R146 and C142, thereby applying the
or 1/15,750 which is equal to 63.5 microseconds. It video signal output of the 6K6 to the cathode of
can be seen that the time constant used in this cir- the diode. The only time the diode can conduct is
cuit is very much greater than the time between when the signal on the cathode goes more negative
sync pulses. Capactor C66 will not be able to dis- than the -50 volts or so that is applied to the plate
charge between sync pulses and, therefore, the cir- of the diode; this is the same as saying that the
cuit will not respond to changes in brightness of plate goes positive with respect to the cathode. If
each individual line. Let us now consider the effect the circuit is adjusted properly this will happen
of a change in average brightness of the entire pic- only at the bottom of the sync pulses. The sync
ture. This will not take place more often than, let pulses at the output of the 6K6 are in the negative
us say, twice a second. The time between changes direction in order to get the correct polarity on
will therefore be 0.5 second or 500,000 microsec- the grid of the Kinescope. When the diode con-
onds. This is more than 10 times the time constant ducts, the capacitor C142 will charge and produce
of the RC circuit used for d -c restoration. Capaci- a d -c voltage on the cathode of the diode. This volt-
tor C66 will be able to discharge appreciably in this age is applied to the grid of the Kinescope through
time and therefore the circuit will respond to the resistor R148. This voltage is added to the video
changes in average brightness of the entire pic- signal and will raise the pedestal above a reference
ture. level which is equal to the negative voltage applied
The 2000 -ohm potentiometer in the input circuit 120µH
to the 6AH6 varies the signal applied to the grid
of this tube. This enables the range of brightness
C14I
or the contrast of the picture to be adjusted. The
10BP4
potentiometer R57 varies the brightness of the pic- 22K
KINESCOPE
R148
ture by changing the bias on the picture tube. When 6K6 L190 100K
2No VIDEO
properly adjusted, the scanning spot will be cut off 93
by the blanking signal. Notice that only one video

amplifier is employed in this model, as the signal C142 R149
foAl5
D.C.RES70RER
output of this amplifier is sufficient to properly R147 .05 1M
3300
drive the 7GP4. The compensating circuits used
in this set are similar to those that have been dis- R150
47K
+135
cussed. Combination shunt and series peaking is BRIGHTNESS
used. The coil assembly T13 contains both the shunt
and series peaking coils.
TO SYNCH.
AMPL. 1 ISO K
0143
zsµf -loo R152
50k
Video Output of RCA Model 630TS F1c. 7-21.-The video output stage of the RCA model
630TS is an example of a diode d -c restorer. It will be noticed
that several changes have been made when it is compared with
The video output stage of the RCA model 630TS the basic circuit of Fig. 7-17.
is shown in Fig. 7-21 as an example of a diode cl -c
restorer. Comparing this circuit to the basic circuit
as shown in Fig. 7-17, we see that it is basically the to the plate of the diode by the brightness control.
same circuit but that several new features have been Thus all the pedestals will be lined up at the grid
added. First, notice the brightness control. Instead of the picture tube at the correct operating point of
of supplying a d -c bias to the cathode of the picture the tube.
tube as in the basic circuit, this brightness control The time constant of the d -c restorer circuit de-
supplies a negative bias to the grid of the picture pends mainly on the combination of C142 and R149
tube. The object is the same in both cases, that is, and is equal to 0.05 second. This time constant is
to set the operating point of the Kinescope at the sufficiently larger than the time between horizontal
correct value. In Fig. 7-21 the brightness control sync pulses so that the circuit will respond only to
1

changes in the average brightness of the entire pic- determines the magnitude of the video signal ap-
ture. The action is similar to that discussed in the plied to the grid of the picture tube. Its counter-
grid leak -capacitor type of d -c restorer. Resistor part in a sound receiver is the volume control, and
R146 in series with capacitor C142 isolates the stray in the same way that the sound volume control de-
capacitance of the diode circuit from the output cir- termines the range of intensities between the loud-
cuit of the 6K6. Resistor R150 serves as a load est sound and the softest sound, so the video con-
resistor for the sync output circuit. Since current trast control determines the range of light intensi-
flows through this resistor only when the sync ties between the highlights and the shadows of the
pulse occurs and the diode conducts, the sync pulses picture. Referring to Fig. 7-18, the setting of the
alone will appear across R150 and can be used to contrast control determines how much of the pic-
supply the input of the sync amplifier. Therefore ture -tube characteristic is used and how bright will
this circuit performs another function, that of re- be the brightest element of the scene. When the
moving the sync signals from the video signal. contrast control is not advanced far enough, the
picture lacks brilliance and the highlights are com-
paratively dark ; when the contrast control is ad-
The Contrast Control vanced too far, the picture becomes blurred, there
is a loss of detail in the highlights, and in general
The "contrast control" is the gain control which the intermediate shades are lost.
CHAPTER 8
SYNCHRONIZING CIRCUITS
BY RICHARD F. KOCH
The circuits devoted to synchronization in a vision receiver, the picture on the cathode-ray tube
television receiver are those which remove the sync may be intolerably distorted.
information contained in the complete video signal One of the weakest points in a television receiver,
and utilize it to control the timing of the horizontal as regards its sensitivity to static, is the sync sys-
and vertical deflection oscillators. The need for this tem. This is particularly true of the horizontal
has been discussed in connection with the subject synchronization, because of the high frequency of
of the composite transmitted signal. There it was the horizontal sweep and the narrow sync pulses
shown that an important part of the transmitted employed. (Horizontal sync pulses arrive at the
signal is the sync signals, which lie in the "blacker - rate of 15,750 per second and are only 4/ to 5/
than -black" region. Without these sync signals, the microseconds wide ; vertical sync pulses, on the
picture on the cathode-ray tube of the receiver other hand, arrive in groups occurring only 60
would not stay put ; at the very least its motion times per second, and these groups are some 190
would be annoying, and in most cases the motion microseconds in width.) A burst of static may,
would be so great as to make the presentation use- therefore, mask a substantial number of successive
less as entertainment. horizontal sync pulses; or, if static occurs in pulse -
like form, it may cause the horizontal sweep to sync
We have also seen that the television transmitter with it instead of with the broadcast sync pulses. The
broadcasts two different kinds of sÿnc signals- same burst of static would have much less effect on
horizontal and vertical. The sync circuits are re- the vertical sync ; in fact, it might easily come be-
quired to separate these two signals, in addition to tween two consecutive groups of vertical sync
the other functions they must perform. Having pulses and have no effect whatsoever on the vertical
separated these signals, the sync circuits must route deflection circuit. If there is a considerable amount
them to the appropriate deflection circuits. We shall of static, some of it will probably arrive at the
discuss the manner in which this is accomplished receiver at the same time as a group of vertical
shortly, but first let us consider some generalities sync pulses, and so affect the operation of the ver-
in the characteristics of sync circuits and the re- tical deflection circuit. However, for a given in-
strictions that are imposed upon them by outside tensity of static, the horizontal deflection circuit is
influences such as static. more likely to be affected than the vertical sweep.
General Requirements Although this difference in susceptibility to inter-

It must be remembered that of all the broadcasting ference must be borne in mind, it must also be re-
services, television is the most sensitive to static inter- membered that it makes little difference to the aud-
ference. It has been stated (although this state- ience whether the picture is given to jumping verti-
ment has been disputed) that a 500 -microvolt sig- cally or horizontally. In either case they will be
nal is required to get a good picture on the cathode- annoyed, and so both sync circuits must be protected
ray tube of a television receiver. There are several from interference. The best way of doing this is to
reasons why such a strong signal is needed. Most keep the signal-to-noise ratio as high as possible. This
important is the difference between the human eye is accomplished partly by getting the strongest pos-
and the human ear, for the former is far more sible signal to the input terminals of the receiver and
sensitive to visible distortion than the latter is to partly by using circuits that will respond to sync
audible distortion. As a result, flaws in the sound pulses in the desired manner, while effectively reject-
coming from a loudspeaker due to the presence of ing interference such as static. This rejection may be
a certain amount of static may be negligible, but accomplished in one of two ways or by combining the
when the same amount of static acts upon a tele- two methods. One way is to use limiters or clippers
141
that prevent any signals corresponding to substantially ceding pulse than it was when it was transmitted.
more than 100 -percent modulation from passing Similarly, the apparent spacing between the begin-
through the sync circuits to affect the sweep circuits. nings of the second and third pulses seems to be
Since the sync pulses correspond to the top 20 to 25 greater than the true spacing. If there is apprecia-
percent of the video modulation imposed upon the ble static, this effect may be repeated frequently,
video carrier at the television transmitter, this limiting causing considerable jitter in the picture seen on
action will pass the sync pulses, while cutting off inter- the picture tube.
ference riding above them. As was pointed out previously, the horizontal
sweep circuit is more susceptible to synchronizing
Sync Methods: Instantaneous Locking vs. AFC difficulties of the sort illustrated by Fig. 8-1 than
is the vertical sweep circuit. For this reason it has
As stated above, there are two methods of re-
seemed desirable in some television receivers to use
ducing the effects of interference on the operation
an afc (automatic frequency control) circuit to pro-
of the sync circuits. One of these has already been
vide horizontal sync. In this method of controlling
mentioned ; namely, clipping or limiting. The other
the horizontal sweep, the individual sync pulses do
involves an averaging action, causing the sweep to
not exert a direct effect on the sweep oscillator. In-
be synchronized at the average frequency of the
stead, the sweep frequency is locked to a value cor-
sync pulses. The average frequency of the sync
responding to the average spacing of perhaps one
pulses generated at the transmitter is of course
hundred or so sync pulses. (How this is accom-
constant, and their spaci1 g is quite regular. How- plished will be seen later, when we consider the
ever, extraneous effects, such as static, may modify
horizontal sync circuit of the RCA television receiver
the apparent spacing of these pulses.
model 630T S. )
A way in which this may occur is shown in Fig. Now let us re-examine Fig. 8-1 in order to see
8-1. The three pulses are shown here in a very simple the benefit of this averaging process. In this simpli-
manner. Simultaneously with one of these pulses a fied example, the apparent spacing between the first
burst of static occurs. As a result, the apparent am- and second pulses is less than the true spacing with
plitude of the concurrent pulse, as seen by the receiver which these pulses left the transmitter, and the
sync circuit, is much greater than its fellows, and, apparent spacing between the second and third
what is more important, it seems to start sooner. pulses is greater than the true spacing. However,
Looking at the portion of the figure illustrating the the spacing between the first and the third pulses
as they are received is the same as when they were
TRANSMUTED
transmitted, because the static has had very little
PULSES effect upon them. Since the first and third pulses
have not experienced an apparent shift in position,
STATIC
FIG. 8-1. -
parent change in
Ap- as the second has, the average spacing of the three
pulses is equal to the true spacing. This is so be-
sync pulse spacing
caused by static. cause the spacing from the first pulse to the second
REQUIRED SYNC.
PULSE HEIGHT
has apparently decreased, while the spacing from
the second to the third has apparently increased by
L_- x.
I -
-J RECEIVED PULSES
APPARENTSPACING
TRUE SPACING
received pulses, you will observe a line labelled "re-

the same amount as the spacing from the first to
the second decreased. The decrease and the increase
balance each other, and thus the average spacing
quired sync pulse height." This is the height to which of the three pulses is equal to their true spacing.
the sync pulses must rise in order to perform normal We pointed out before that when the individual
synchronizing action. Since simple synchronizing is sync pulses are used to synchronize the sweep, static
something like the action of an on -off switch, a pulse may cause the picture on the cathode-ray tube to
failing to reach the required level will have the effect jitter. This occurs because the apparent spacing of
of no pulse at all, while a very high pulse will affect the pulses is affected by the static. In the example
the circuit pretty much like one of the proper (lower) of Fig. 8-1 we have seen that the static does not
height. Since the middle pulse, riding on top of the affect the average spacing of the three pulses shown.
static burst, reaches the height required for syn- Therefore, if the average value of the sync pulse
chronizing sooner than it would in the absence of spacing is used to determine the sweep frequency,
the static, it seems to be spaced nearer to the pre- static will not cause the picture to jitter. (The
SYNCHRONIZING CIRCUITS 143
example of Fig. 8-1 is simplified to the point where nally the sync signal is fed to a selecting circuit
it is not entirely realistic, for under actual operating which separates the horizontal sync pulses from the
conditions static would probably affect a series of vertical sync pulses. This separation is accom-
pulses, rather than just one out of three. However, plished by circuits which depend for their action
if we considered a much larger number of pulses upon the difference in the time duration of the
that three, we would find that in actual practice two types of pulses.
the averaging effect holds, just as in the simplified The circuits which discriminate between long
example.) (vertical) and short (horizontal) sync pulses are
simple combinations of resistors and capacitors.
Fig. 8-2 shows a resistor and capacitor in series
Sync Separation and the voltages that appear across each when the
The circuits used to separate the sync pulses from capacitor is charged from a battery and then dis-
the picture portion of the signal are called "sync charged, in both cases through the resistor. Two
separators" or "clippers." These circuits are de- sets of curves have been drawn, corresponding to
signed so that they are responsive only to the sync two combinations of resistance and capacitance. It
pulses, which lie above the pedestals (that is, in the may be seen from this illustration that when the
"blacker -than -black" region) ; thus they reject the pic- product of resistance and capacitance is large, a
ture portion of the signal. This rejection is necessary relatively long time is required to charge the capacitor
to prevent the video signal from traveling through to full voltage or to discharge it to zero. However,
the sync circuits to the sweep oscillators. The video when this product is small, a relatively short time is
signal might affect the operation of the sweep oscil- required. Because of the relationship between the time
lators, causing them to fall out of their required required for charge or discharge and the product of
synchronism. The clipper or sync separator is gen- resistance and capacitance, this product is called the
erally followed by additional amplification, and fi - "time constant" of the circuit.
Now let us look at Fig. 8-3, in which is shown.
VOLTAGE ACROSS the behavior of series resistance -capacitance circuits
BATTERY RESESTOR
VOLTE such as those shown in Fig. 8-2, when series of
pulses are impressed across them. In this figure two
special cases are illustrated; these cases are very
important in television receivers, as will be seen
-REPRESENTS CASE VOLTAGE'ACROSS later. Part (A) shows the signal observed across
OF SMALL PRODUCT CAPACITOR
OF RESISTANCE AND BATTERY the resistance when narrow pulses are impressed
CAPACITANCE. VOLTAGE
across a series combination having a short time
- -REPRESENTS CASE constant. (In this discussion it must be remembered
OF LARGE PRODUCT
OF RESISTANCE AND
CAPACITANCE.
that the terms "short" and "long" are relative.)
SwIT SWH
CLOSES OS
TO A B
CTO
FIG. 8-2, above.-Sync separator circuit

consisting of a resistor and capacitor in
series. The voltages that appear across
each when the capacitor is charged from
a battery and then discharged, in both
cases through the resistor, are shown at
the right. INPUT OUTPUT
(A)
FIG. 8-3, right.-Effect of series resist-

ance -capacitance circuit on input pulses,
showing (A) the signal observed when
like narrow pulses are applied and (B)
when the input consists of two different INPUT
kinds of pulses.
Comparing this with the resistor voltage curve for the space that preceded it, the rise that it produces
the short time constant case in Fig. 8-2, we see that is less than the drop which it follows. Therefore,
in both cases the voltage rises immediately to the at the end of this narrow pulse the output voltage
peak value of the applied voltage, and then drops is lower than it was at the end of the third wide
quickly back to zero. When the applied voltage pulse. The interval before the occurrence of the
drops to zero and the capacitor discharges, the flow next narrow pulse is relatively long, and so the out-
of current is opposite to that which existed during put voltage falls appreciably during this time. Then,
the charging pulse, and a pulse similar to the charg- during the narrow pulse it rises slightly again, and
ing pulse, but of opposite polarity is produced. The finally falls to zero in the interval succeeding the
output then remains at zero until the next input last narrow pulse.
pulse arrives. It may also be observed that there Let us review briefly the actions illustrated in
is a very short interval between the pulses of oppo- Fig. 8-3(B). The narrow pulses produce small rises
site polarity when the output is at zero. Each pair of in voltage, small because the capacitor is able to
output pulses corresponds to one input pulse. acquire only a small charge during the short inter-
In Fig. 8-3(B) a somewhat more complicated val of each pulse in this long time constant circuit.
case is illustrated. Here the input consists of two In the long intervals following the narrow pulses
different kinds of pulses. The output in this case all the charge acquired by the capacitor during the
is obtained across the capacitor, and the time con- preceding pulse leaks off. When the pulse -and -space
stant of the circuit is long. Because the time con- situation is reversed, however, the output voltage
stant is long, the first narrow pulse produces a small changes materially. Here, the pulses are wide and
rise in voltage in the output this rise continues for
;
the intervals between them are narrow, so that the
the duration of the pulse, and at its conclusion the capacitor has an opportunity to acquire a rather
output voltage returns to zero. Compare this with high charge when a pulse is applied, and only a
the capacitor charge and discharge curves in Fig. small part of this charge leaks off in the interval
8-2, for the long time constant case. Returning to between pulses. Therefore, when the wide, closely -
Fig. 8-3(B), we see that the second and third nar- spaced pulses are applied, the output voltage rises
row input pulses produce the game kind of low, to a point considerably above zero ; while when nar-
triangular pulses in the output as that produced by row, widely -spaced pulses are the input, the output
the first pulse. However, the fourth pulse is of rel- voltage is very low. After the three wide pulses,
atively long duration and the charge on the capaci- narrow pulses are applied again, and the output
tor-and, therefore, the output voltage-rises to a voltage drops back to zero. Of course, if narrow
much higher level than during the short pulses. pulses continue to be applied, they will produce low
When the pulse ends, the charge starts to leak off triangular pulses in the output, just as the first
at a rate determined by the resistance ; but the next three did, but for most of the time the output will
pulse, also a wide one, comes along very soon, and be zero.
the drop in output voltage is only a small part of the Looking once more at Fig. 8-3(B), we observe
rise that occurred during the first wide pulse. The a very important interrelation between the input
new rise in charge-and output voltage-that takes and output waves when the input consists of nar-
:
place during the application of the second wide row pulses, spaced far apart, the output consists
pulse to the circuit therefore starts from a level of low triangular pulses, and the general level of the
higher than that at which the rise produced by the output is practically zero. However, when wide,
first wide pulse started. As a result, the output closely spaced pulses are applied to the input, the
voltage rises to a new high value at the end of the output level rises considerably, without any fall to
second wide pulse. During the interval between the zero between input pulses. Finally, when narrow,
second and third wide pulses there is again a drop far -spaced pulses are reapplied, the output falls
in voltage; nevertheless, the point from which the nearly to zero, and conditions return to those that
rise due to the third wide pulse starts is higher existed when the narrow pulses were applied origi-
than the starting points of any of the previous nally. Thus the circuit discriminates between nar-
rises. The end of this third pulse therefore sees the row and wide pulses by producing a small, almost
output voltage at the highest value yet. A drop, of insignificant output for the former and a large out-
course, follows and then another rise due to the put for the latter.
succeeding narrow pulse. However, because the To illustrate the principles involved in sync cir-
duration of this narrow pulse is less than that of cuits, we shall consider some specific examples from
television receivers on the market today. As an ex- the receiver. However, synchronization is so im-
ample of sync clipping and separation we shall con- portant in the operation of the television system
sider the circuits used in the Belmont models 22A21- that great care must be taken to make synchroniza-
22AX21 and 22AX22. RCA model 630TS has been tion positive in action,
chosen as an example of the afc type of horizontal Referring to the partial schematic of Fig. 8-5,
sync control. we see that the video signal is fed to the grid of the
sync clipper tube, one half of a duo -triode type
6SL7-GT. The distinguishing features of this stage
Belmont Models 22A21, 22AX21 and 22AX22 are that the tube is operated at the very low plate
As the block diagram in Fig. 8-4 shows, the video voltage of 5.6 volts, and at a bias close to cutoff,
signal is taken from the first video amplifier and provided by the flow of grid current through the
passes through the sync clipper, which removes large grid resistor R60. (The very low plate volt-
most of the video signal from the sync signal. The age is attained by using a very low effective supply
sync signal is then fed to a sync limiter stage which voltage-about 8 volts obtained from a bypassed
bleeder across the 300 -volt plate supply.) As the
1HOR.PULSEH TO HOR. waveform at the grid of this tube shows (Fig. 8-5),
SELECTORM U LT. the sync pulses are the most positive part of the
IST VIDEO
AMPL.
SYNC.
CLIPPER - SYNC.
LIMITER ""--
VERT. PULSE
input signal. Since the plate voltage of this tube is
very low, only a very small negative signal is requir-
SELECTOR
TO VERT.
MULT. ed to drive it into cutoff, particularly since there is
already a relatively high (considering the low plate
FIG. 8-4.-Block diagrams of the sync circuits used in the
Belmont models 22A21, 22AX21, and 22AX22.
voltage) negative bias due to R60. Therefore the
tube is cut off by the picture components of the
video signal, and passes plate current only during
fixes the height of the sync signals, .removing small the most positive part of the signal, that is, the
variations in amplitude that may persist after the part devoted to the sync pulses. As a result, the
action of the avc in the r -f and i -f amplifiers. The picture components of the signal have no effect on
output of the limiter is applied to two pulse selec- the plate current (except to keep it at zero), but
tors-one for horizontal and one for vertical sync the sync pulses cause it to flow in corresponding
pulses. These selectors pass the sync pulses re- pulses. These pulses of current produce voltage
quired by the multivibrator sweep oscillators to pulses across R62 (and R64), as shown in Fig. 8-5.
which they are connected, while virtually elimina- The signal has thus been transformed in a desirable
ting the undesired pulses. manner by being passed through the sync clipper;
On the whole, the block diagram presents quite the picture components have been eliminated, so
an imposing array of circuits, considering that it that they cannot cause synchronization at the wrong
represents but a small part of all the circuits in time, while the sync pulses have been retained.
L
i65L7-GT
C)t1
0.1
265L7 -GT
R65
.25
MF
C72 1500
MMF
I HORIZ. SYNC
PULSES TO
VIDEO SIG. 470K
MF C73 HOR!Z. MULT.
FROM 1" R67
VIDEO AMPL. +300V 4700
R64
2.2
R60 MEG
8.2 MEG R62 R66
10K 1000
RB4
.005 l C88
+8v. MF T 4700
-
V V
VERT. SYNC
.01 I PULSES TO
C91
MF VERT. MULT.
FIG. 8-5.-Partial schematic of the sync circuits of

Belmont models 22A21, 22AX21, and 22AX22.
The clipped sync signal is next fed to the grid frequency. However, pulses whose timing is far
of the second half of the 6SL7-GT, which is a off will have negligible effect unless their amplitude
cathode follower providing limiting action. The is very large. In this circuit the sync limiter keeps
combination of R65 and C72 has a time constant of the pulse level at a point where the extra pulses
nearly one -eighth of a second, which is sufficiently due to the vertical sync pulses will not interfere
long compared to the intervals between sync pulses with the proper operation of the horizontal multivi-
to keep the plate of the tube at a virtually constant brator.
voltage. This plate voltage is 63 volts, and the Let us refer once more to Fig. 1-21. Here we see
cathode voltage between input pulses is about 0.5 that each vertical sync pulse group is both preceded
volt. Since the input pulses are negative and of an and followed by equalizing pulses. These pulses are
amplitude almost equal to the plate voltage of the half the width of the horizontal sync pulses and
sync clipper stage (in other words, about 5 volts), every other one lies where a horizontal sync pulse
they will cut off the limiter tube. Thus, the limiter ordinarily lies. Because they are so narrow they
acts as a switch, passing plate current in the absence are passed by the horizontal pulse selector just as
of pulses and stopping it in their presence. As a the horizontal pulses themselves are. Their spacing,
result, the amplitude of the pulses appearing at the and therefore their effect on the horizontal multivi-
cathode of the sync limiter stage is unaffected by brator, is just the same as that of the negative
the amplitude of input pulses, but the timing of the pulses generated by C73 and R67 from the vertical
input and output pulses is exactly the same. sync pulses. Thus the synchronization of the hori-
Connected to the cathode of the sync limiter are zontal multivibrator is maintained during the verti-
the two pulse selector networks, which separate the cal synchronization period, even though horizontal
horizontal and vertical sync pulses. The combina- sync pulses as such are not present in this interval.
tion of C73 and R67 has a time constant (about 7 In other words, the vertical sync and equalizing
microseconds) which is approximately equal to the pulses are made to do double duty, for they work
width of a horizontal sync pulse. These pulses are, in the horizontal sweep circuit as well as the verti-
therefore, passed through it with little change, as cal sweep circuit. (This principle of double duty
shown by the final output to the horizontal multivi- applies to television receivers generally; it depends
brator in Fig. 8-5. When individual vertical sync upon a fundamental characteristic of the standard
pulses reach this circuit they are converted to sharp television signal, and all receivers intended for re-
double pulses, somewhat as shown in Fig. 8-3(A), ception of this signal take advantage of this.)
for their duration is approximately three times the The complete sync signal is also fed into the
time constant of the combination of C73 and R67. vertical pulse selector consisting of R83 and C88.
Because the input pulses are negative, the first pulse An additional resistor -capacitor combination (R84
of each pair will also be negative, and the second and C91) is used in order to increase the effective-
positive. Referring to the standard television signal ness of the vertical pulse selection. The action of
shown in Fig. 1-21, we see that the beginning of R83 and C88 in differentiating between the vertical
every other vertical sync pulse lies exactly where a and horizontal sync pulses is a direct result of the
horizontal sync pulse ordinarily lies. Therefore dur- longer duration of the vertical pulses. Fig. 8-3(B)
ing each vertical sync pulse group there will be shows how a circuit of this sort distinguishes be-
generated a series of negative pulses, equally spaced, tween narrow and wide pulses. The output builds
and every other one will correspond exactly in po- up to a maximum at the end of the last wide pulse
sition to the point where a horizontal sync pulse (in the case of R83 and C88 the last pulse in each
might be found. (There .are also generated positive vertical sync pulse group), and then steadies down
pulses, as previously mentioned, but these have negli- almost to zero, remaining at this value until the
gible effect on the horizontal multivibrator.) beginning of the next vertical sync pulse group.
One might well ask, "What effect will the extra To increase the circuit effectiveness, another sim-
negative pulses in this interval have on the hori- ilar resistor -capacitor combination (R84 -C91) is
zontal multivibrator ?" The answer is, "None that used in series with the first as shown in Fig. 8-5.
is significant." The reason for this is that the mul- This circuit accomplishes a further discrimination
tivibrator's frequency is only partly dependent on between the vertical and horizontal sync pulses so
the sync pulses; sync pulses that arrive at or very that a sharper output pulse is produced once during
near the proper point in the multivibrator's oper- each field. The peak of this pulse is used to syn-
ating cycle will affect it, holding it at the desired chronize the vertical deflection oscillator.
For exact interlacing it is important that the ver- acting like a variable reactance, with the reactance
tical pulses produced at the output of the vertical dependent upon the control voltage applied to the
selector circuit be the same on alternate fields. The tube. Since this variable reactance (which may be
equalizing pulses, described in chapter 1, accom- either capacitive or inductive) is part of the oscil-
plish this by making the conditions exactly the lator's tank, it affects the frequency of the oscilla-
same before and after the transmission of the actual tor. Therefore, the error signal from the discrimi-
broad vertical sync pulses. As a result capacitors nator can be used to change the frequency of the
C88 and C91 charge up to the same values of peak local oscillator, and if the circuit is connected prop-
voltage on both the odd and even fields, so that the erly, this change will be such as to reduce the error.
vertical oscillator is maintained in the correct In this manner a self-correcting system is set up,
timing. which tends to correct for errors that may exist in
its initial adjustment.
Fig. 8-7 shows a block diagram of an afc system
AFC Circuits of the type just described. Let us contrast it with
The block diagram in Fig. 8-6 shows the general that shown in Fig. 8-6. There are two differences
arrangement of the sweep afc circuit used in the in the inputs to the discriminators shown. One of
RCA model 630TS television receiver. This may these is minor. It is this in Fig. 8-6 the oscillator
:
be recognized as bearing a similarity in principle to feeds the discriminator directly, whereas in Fig.
the afc schemes used for correcting errors in tun- 8-7 it does not. However, in the latter case we are
ing superheterodynes. In the case of the superhet- not concerned directly with the frequency of the
erodyne tuning, it is desired to overcome an error incoming (r -f) signal, and so we assume that any
in the frequency of the local oscillator frequency. error in the intermediate frequency is due to mis -
tuning of the local oscillator. Therefore, the dis-
FROM 2"D
SYNC. AMPL. criminator input depends directly upon the local
oscillator frequency, even though there is not a
direct signal path between them.
SYNC. TO NEXT The other difference between the two figures is
DISCRIM- 05C. STAGE a very important one. In Fig. 8-7 there is only one
INATOR
input to the discriminator, but in Fig. 8-6 there
are two. What is the significance of this difference ?
REACTANCE
In a circuit of the type illustrated in the latter fig-
TUBE
ure, the tuning of the discriminator exercises a direct
control over the frequency. In the circuit of the
FIG. 8-6.-Block diagram of sweep afc circuit as used in
RCA model 630TS. In this circuit the pulses from the second former figure, the sync pulses derived from the second
sync amplifier are of primary importance in determining the sync amplifier are of primary importance in deter-
output from the discriminator.
mining the output from the discriminator. In this
case the discriminator serves only to compare the
This error is such that the actual intermediate fre- frequency of the signal to be controlled with the
quency generated is not that for which the set was
designed. In order to produce a correction, part of
the i-f signal is fed into a discriminator, whose out-
put depends upon the error in the i.f. This output
is a d -c voltage, whose amplitude depends upon the TO
DETECTOR
magnitude of the error, and whose polarity depends
upon the direction of the error. That is, if the error REACTANCE
is due to the oscillator being tuned too high, the out- TUBE
put of the discriminator will have, let us say, posi- FIG. 8-7.-Blcck diagram of type of afc scheme used for
tive polarity, and then if the oscillator frequency correcting errors in superheterodyne tuning. In contrast with
were too low, the discriminator output would be the circuit of Fig. 8-6, the oscillator does not feed the dis-
criminator directly, and, more important, there is only one
negative. This control voltage (the output of the input to the discriminator.
discriminator) is applied to a reactance (or oscilla-
tor control) tube, which is part of the tank circuit
of the oscillator. This tube (or, rather, the circuit sync pulses, and its tuning is of secondary impor-
of which this tube is a part) has the property of tance.
A complete schematic diagram of the horizontal the effect of R194 on the capacitive circuit is so
sweep oscillator and its associated afc circuit used small that the oscillator -frequency current passing
in the RCA model 630TS receiver is shown in Fig. through it is almost purely capacitive. To under -
8-8. The oscillator tube is V125 ; it operates in a PRIMARY
Hartley type circuit. The oscillator coil serves a OFT104
r--1
dual purpose-its basic function and as the pri-
mary of the discriminator transformer T108. The
oscillator coil is tuned by the parallel combination FIG. 8-9. - Schematic
diagram of the horizontal
of capacitors C169 and C164 and the reactance tube sweep oscillator circuit
V124, as shown in Fig. 8-9. In addition to the re- of RCA model 6311TS.
actance tube there is another element in the tank
circuit that is not a fundamental part of the oscilla-
tor. This is the 10 -ohm resistor R194. Ordinarily R 194
the oscillator tank is tuned by a pure capacitance stand the desirability of this capacitive component
çonnected from one end of the inductance to the of current in R194, let us briefly consider the man-
other, but in this case there is a small resistance in ner in which a reactance tube operates.
series with the capacitance. The total capacitance
(C169 + C164) is 16,200 µµf. At the horizontal
sweep frequency of 15,750 cycles the corresponding Reactance Tube Operation
reactance is approximately 620 ohms. Therefore the In Fig. 8-10 we see a closed box containing an
10-ohm resistance will have very little effect upon unknown circuit, and having two terminals. We
the current through the capacitive branch of the cannot open the box, but we can measure the cur-
tank, and the series RC combination will behave as rent flowing from the box through whatever ex-
though it were almost pure C. ternal circuit may be connected, and we can meas-
What, then, is the function of R194? Look again ure the voltage between the terminals. In addition,
at Figs. 8-8 and 8-9. Reactance tube V124 is con- we find that the current and voltage are 90° out of
nected directly across C169 and C164. (Capacitor phase ; this means that the box contains a pure re-
C173 is simply a d -c blocking capacitor.) In addi- actance of some sort. Knowing this, we can apply
tion, its cathode is connected to the junction of Ohm's law in the form it takes for reactances, namely,
these capacitors and resistor R194, and therefore namely,
the cathode voltage of the tube is the voltage ap- X_E
pearing across R194. There are two components of
current through R194; one is the usual d.c. found From this equation we can calculate the size of
in a cathode resistor, the other is the capacitive tank the reactance contained in the box, and thus we
current of the oscillator. As was pointed out above, have learned much of the box's contents without
FROM 2"
0166 SYNC AMPL. TO NEXT
82 GALS T1o6 C172 6K6 GT STAGE
.004 V125
MMF 4123 MF
R191
470 K
8192
470K R199
10K
-ZV J
.004 R200
S 5K
---MF t169_
C167.
8195 .015
560 MF
R193 39K
470K C170
os
MF 1 R194
10
R188
27K
1 0174
«275V
6AC7 T05MF
V124
F1c. 8-8.-Complete schematic diagram of the horizontal sweep oscillator and its associated afc circuit used in the RCA
model 630TS.
opening it. Now suppose the box contains a device high. Thus the reactance tube we have constructed
which can generate a current 90° out of phase with can be controlled by varying its grid bias.
the voltage applied. across the terminals of the box.
Then, to us outside the box, the device appears to Sweep AFC Circuit in RCA Model 630TS
be some sort of reactance. If, further, the amount of
current generated can be controlled, the apparent Having considered some of the basic principles
reactance of the device in the box can thereby be upon which reactance tube circuits are founded, let
controlled. This relationship stems from Ohm's us return to our specific example. Fig. 8-11 shows
law for reactances, as stated above. For example, it in simplified form. We have already seen that the
if the current is doubled( and the voltage is un- current through R194 is almost purely capacitive
changed), the reactance is halved. and, therefore, leads the voltage across the tank cir-
If we can build a device that will actually do what cuit by almost 90°. Since the current through a re-
the imaginary device in the box does, we will have sistor is in phase with the voltage across it, the
a controllable reactance. Such a device can be built sweep frequency voltage across R194 therefore
using a pentode tube. From an examination of the leads the voltage across the entire tank circuit by
characteristic curves of a pentode, such as the 90°. Since the full tank voltage is applied to the
6AC7, we see that the plate current is almost in - plate of the tube through C173 (whose reactance is
very low, and can therefore be neglected), the
I -9.
FIG. 8-10. - This
closed box having two
cathode and plate voltages are 90° out of phase,
terminals and contain- with the cathode voltage leading. The signal applied
ing an unknown circuit to the grid is a d -c control voltage from the dis-
illustrates in a simple
way how it is possible criminator; so far as sweep frequency signals are
to obtain a controllable concerned, the grid appears to be at a fixed poten-
reactance using a pen-
tode tube. tial. The grid bias is approximately two volts that ;
is, the grid voltage relative to the cathode is ap-

dependent of plate voltage over a wide range of proximately -2 volts. In other words, the average
values. However, the current does depend upon the cathode voltage is +2 relative to the grid. Then, if
control grid voltage. Thus it is possible to connect the oscillator signal voltage on the cathode at a
the tube with its plate and cathode across a source certain instant is driving the cathode so that it is
of alternating voltage, without having this voltage more positive with respect to ground than its average
affect the plate current. However, if an alternating voltage, this will also make the cathode more positive
voltage is applied between the grid and cathode, the with respect to the grid than its average value. This
plate current will change in phase with the alterna- is the same as saying that at this instant the grid is
ting grid voltage. Now, suppose that the alternating becoming more negative with respect to the cathode.
voltages applied in the grid and plate circuits are The opposite, of course, is also true. When the os-
of the same frequency, but 90° out of phase. Then cillator signal on the cathode is going negative, its
the plate current will have an alternating component effect is the same as the same signal applied to the
of the same frequency but 90° out of phase with grid, but going positive. This can be summed up by
the plate voltage (since it is in phase with the grid saying that a signal applied to the cathode, with the
voltage). Thus we have constructed a reactance HIGH END
of OSC TAN K
using a pentode, for the current flowing in the plate
1C173 R 198&
circuit is 90° out of phase with the voltage across R 231
the plate circuit. Fic. 8-11.-Sim- C169

plified circuit of re- &C164
V124
It still remains to be seen how the reactance we actance tube used in D.C.CONTROL
have constructed can be made controllable. Look RCA model 630TS. VOLTAGE
RI94
again at the pentode's characteristic curves. One of
these shows how the transconductance of the tube
varies with the bias on the control grid. The trans - grid voltage fixed, has the same effect as the identi-
conductance is a measure of the sensitivity of the cal signal, 180° out of phase, applied to the grid
plate current to changes in grid voltage it tells us
; with the cathode voltage fixed. Thus, the reactance
how much the plate current will change for a change tube in the RCA model 630TS operates as though
in grid voltage. From this we see that a given the sweep frequency signal were applied to its grid,
change in grid voltage will cause a larger change in lagging the voltage applied to the plate by 90° (It
plate current when the bias is low than when it is appears to be lagging, when considered in terms of
its effect at the grid, because of the 180° difference coil (the primary of T108). This coupling provides
between the grid and cathode as regards inputs to a sine -wave signal from the oscillator for the diode
one or the other.) V123. Since the coil is center -tapped, the signals
As was stated above, an alternating voltage ap- fed to the two diode plates are of opposite phase
plied between grid and cathode will produce an al- and equal amplitude. In addition to the 15,750 -cycle
ternating component of plate current in phase with sine -wave signal from the oscillator, sync pulses
the grid signal. In our example, therefore, the plate are applied to the diodes. These sync pulses come
current lags the plate voltage by 90°. As a result, from the second sync amplifier through the short -
this reactance tube behaves like an inductance, for in time -constant circuit C166 -R229. (As explained
an inductance the current lags the voltage across it. previously, a short -time -constant circuit is used to
We have seen that when the bias (d.c.) on a tube separate the horizontal sync pulses from the verti-
is increased, the variation (a.c.) in plate current cal sync pulses.)
due to a change (a.c.) in grid voltage decreases.
Since the sync pulses are applied to the center
Thus, if the bias on the reactance tube is increased,
tap of the secondary of T108, they appear at the
the sweep -frequency component of plate current
diode plates in phase, whereas the sine -wave sig-
decreases. Referring once again to Ohm's law for
nals are of opposite phase. As a result, the com-
reactances, we recall that a decrease in current is
bined signals at the diode plates will not be the
accompanied by an increase in reactance. In an in- same. These signals are shown in Fig. 8-12 for
ductive circuit an increase in reactance at constant
three cases. In part (A) we see the case where
frequency means an increase in inductance. Finally,
the natural frequency of the horizontal oscillator is
then, tube V124 acts as an inductance whose value
correct. By this is meant that frequency at which
varies with the grid bias, being relatively large for the oscillator would operate if no sync pulses were
a high bias and small for a low bias.
applied to the discriminator. It is called "natural"
Returning to Fig. 8-9, we see that the oscillator because it is the frequency that would exist in the
tank circuit consists of two capacitors (C169 and
absence of outside influences (sync pulses). In (B)
C164) and two inductors (the primary of T108 is shown the condition existing when the natural
and the effective inductance of the reactance tube,
frequency is low, while in (C) is represented the
V124). Since one of the components of the tank-
high natural -frequency condition. In the latter two
the reactance tube-is a variable controlled by the
cases it may be seen that the pulses are shifted in
discriminator, a means is provided for automatical- phase with respect to the sine wave. This shift has
ly controlling the sweep frequency. The next ques-
been exaggerated in the figure so as to be more
tion is how the discriminator functions to provide
apparent but although in actual practice only very
;
the control voltage necessary for the reactance tube.
small shifts occur, it is they that produce the re-
Fig. 8-8 shows that a coil (the secondary of
quired discriminator action.
T108) is coupled to the horizontal oscillator tank
In Fig. 8-12(A) the pulses lie on the sine waves
at the 0° point for the lower diode and at 180° for
the upper diode. Therefore under these conditions
the sine waves add nothing to the pulses (the sine
J
UPPER DIODE
of 0° and the sine of 180° are zero), and so the
f v
peak -to -peak amplitudes of the plate signals of the
two diodes are equal. The voltages produced by rec-
tification of these signals appear across R191 and
R192, which are equal. Since the plate signals are
equal, the rectified voltages are also equal. Now the
LOWER DIODE voltages across R191 and R192 are in opposition, so
far as the output to the reactance tube is concerned.
(B)
Jr \/
(C)
This is so because the currents through the indi-
vidual diodes flow in opposite directions from the
junction of the two resistors. Current in the upper
Ftc. 8-12.-Combined sync pulse and sine -wave signals at diode flows up through R191, making the upper.
the diode plates when (A) the natural frequency of the hori-
zontal oscillator is correct, (B) the natural frequency is low, end of the resistor positive with respect to the junc-
and (C) the natural frequency is high. In (B) and (C) the tion. Current in the lower diode flows down through
shift of the pulses with respect to the sine waves has been
exaggerated to make it more apparent. R192, making the junction negative with respect to
the lower end of R192. Thus the junction is nega- may take place in the relative phase of the sync
tive with respect to the -2v. bias source, but the top pulses and the sine wave generated by the horizon-
of R191 is positive with respect to the junction. In tal oscillator. Such phase changes may be due to
this way the voltage across R191 opposes that across temporary apparent displacements of the sync pul-
R192. If the voltages are exactly equal, as is the ses, as are caused by bursts of static. We have al-
case when the oscillator's natural frequency is cor- ready seen; however, that these apparent displace-
rect, they will just balance out, and the control volt- ments tend to cancel themselves out, if a large
age applied to the reactance tube will be the -2v. enough sequence of pulses is considered. Therefore,
applied to the lower end of R192 from the bias sup- if the high -frequency components of the control sig-
ply. nal are removed, the effects of short -time apparent
In the case illustrated by Fig. 8-12(B), the peak - displacements of the sync pulses will thereby be
to -peak plate voltage of the upper diode is greater removed. At the same time, if the low -frequency
than that of the lower. Therefore, the rectified volt- and d -c components are passed through, slow varia-
age across R191 is greater than that across R192. tions in the frequency of the horizontal oscillator
The net effect of this is to make the upper end- of will be properly compensated. Thus, this afc circuit
R191 positive with respect to the lower end of synchronizes the horizontal oscillator to the average
R192. As a result, the -2v. bias is partially can- frequency of the sync pulses, but eliminates jitter
celed, and the bias (control voltage) on the react- in the picture on the cathode-ray tube screen due to
ance tube is lowered. We saw before that a low bias erratic short-time changes in the apparent spacing
on the reactance tube will make it act like a relative- of the sync pulses.
ly small inductance. Decreasing the inductance in The variable resistor R168 is the Horizontal Hold
the tuned circuit of the oscillator raises its frequency. Control. It has a very slight effect on frequency,
This is just what is desired if the natural frequency and allows the operator of the set to adjust the
is low. horizontal oscillator so that the afc circuit can per-
Turning to Fig. 8-12(C), we ,find the opposite form its functions properly. Coarse tuning of the
case to that of ( B). The relative values of the diode oscillator, required in alignment, is accomplished by
plate signals and the rectified voltages across R191 means of the movable core in the primary of T108.
and R192 are reversed. The upper end of R191 is The movable core in the secondary is used to bal-
thus made negative with respect to the lower end ance the discriminator action. The discriminator
of R192, and the bias applied to the reactance tube must be balanced so that it neither adds to nor sub-
is increased. This causes the reactance tube to in- tracts from the -2v. bias when the natural frequency
crease the inductance of the tank circuit, bringing of the oscillator is correct.
the frequency down. This, of course, is the neces- The cathode resistor, R194, of the reactance tube
sary direction of compensation when the natural has an unusually low value because it is in the tank
frequency is high. circuit of the oscillator, and a large resistance would
impair the efficiency of that circuit. In addition, a
voltage shifted 90° relative to the plate signal of the
Additional Circuit Features tube is required. As we have already seen, R194
must be small in order to approach this value as
We have now covered the major points to be closely as possible. By injecting the phase -shifted
considered in a discussion of this circuit. However, signal into the cathode of the reactance tube instead
there remain some details of interest, which we of, into the grid along with the control voltage, sep-
shall go over briefly. aration between these two signals is accomplished.
Capacitors C167 and C170, together with resistor This separation prevents undesired interaction be-
R193, form a filter to reduce the high -frequency tween the oscillator and control circuits. Because
components of the control signal coming from the R194 is so small, the d -c voltage across it is negligi-
sync discriminator. The discriminator circuit it- ble. Hence the necessity for the -2v. fixed bias ap-
self will respond very rapidly to any changes that plied through the discriminator.
CHAPTER 9
SWEEP CIRCUITS
BY RICHARD F. KOCH
In chapter 1 we explained briefly how the elec- "trace" is that which carries the spot from the left
tron beam in a picture tube can be deflected by saw - of the screen to the right, or from the top to the
tooth waves applied to the deflecting elements of bottom, as the case may be. During the portion
the tube. (More will be said on this subject in chap- marked "retrace," the spot returns to the left or
ter 10.) The necessity for synchronizing these top of the screen. While the trace voltage or cur-
waves was also explained in chapter 1 and in chap- rent is being applied to the deflecting system, the
ter 8 we showed how the sync signals are selected screen is illuminated, but during retrace it is blanked
at the receiver. The next problem is to generate the out (see chapter 1). In other words, the picture is
saw -tooth waves, control them by the sync pulses, actually being created only while the spot is moving
and apply them to the sweep (or deflection) sys- in the trace direction. Therefore it is very important
tem of the picture tube. We have already seen, in that the motion of the spot during trace be the same
chapter 8, one way in which the sync pulses may as the motion of the scanning spot in the camera tube.
be used to synchronize the horizontal sweep oscil- Otherwise, the relative positions of the various ele-
lator. ments in the received picture would not be the sanie
The block diagram of Fig. 9-1 shows the essen- as in the original, and thus the received picture would
tial parts of the sweep circuits used in television be distorted. However, since the spot does not illu-
receivers. From the sync separator described in minate the screen during the retrace period, its exact
chapter 8, horizontal sync pulses are applied to the motion at this time is not important ; it is only neces-
horizontal sweep circuit (through the afc circuit,
sary that the beginning and end of the retrace be
if one is used) and vertical sync pulses to the ver-
the same as that of the spot in the camera tube.
tical sweep circuit. Both sets of sweep circuits are
This property of the retrace is very important in
basically similar ; the differences which exist are
magnetic scanning systems, as we shall see later.
due to the difference in operating frequencies. (You
will recall that the horizontal sweep frequency is
15,750 cycles, and the vertical sweep frequency is Electrostatic vs. Magnetic Sweep
60 cycles.) The most pronounced difference ap- We have seen in chapter 1 that an electron beam
pears in the deflection coils used in magnetic sweeps, may be deflected either by a suitable electric field
and in the circuits immediately associated with set up between a pair of plates within the tube, or
them. by the required magnetic field which is produced by
The saw -tooth waveforms shownin Fig. 9-2 were a pair of coils (actually a single coil divided into
discussed in chapter 1. There it was pointed out balanced halves) around the neck of the tube. Both
that the portion of the saw -tooth wave marked of these methods have their advantages, and both
OSCILLATOR
AND DISCHARGE SWEEP
CIRCUIT AMPL.
FIG. 9-1.-Block diagram which
VERTICAL DEFLECTION CIRCUITS shows the est hl p are o the
sweep circuits which are used in
DEFLECTION television receivers. Horizontal
YOKE OR
SYNC. PLATES
and vertical sync pulses go from
SEPARATOR the sync separator to the respec-
tive deflection circuits as shown.
Both sweep circuits are basically
HORIZONTAL DEFLECTION CIRCUITS
similar, the differences being due
to the operating frequencies.
r
-. AFC
(IF USED)
1
.-_ OSCILLATOR
AND DISCHARGE
CIRCUIT
SWEEP
AM PL.
L J
1 52
SWEEP CIRCUIT'S 153
their disadvantages. The great advantage of the This effect is not so noticeable at the comparatively
electrostatic sweep is its simplicity, which makes low anode voltages used in the smaller direct -view-
for compactness. (We shall see later in this chapter ing tubes; but in the large tubes and the projection
why the electrostatic sweep is the simpler of the tubes, both of which use high voltages, it is im-
two types.) Next in line is the low power require- portant. Another advantage that makes magnetic
ments, particularly when it is used for the smaller sweep desirable when high -voltage picture tubes
picture tubes. A further advantage is that the "ion are used is the fact that the sensitivity of a picture
tube to magnetic deflection does not decrease nearly
so rapidly with increasing voltages as when elec-
trostatic sweep is employed. For this reason, the
Wú
sweep voltages required for electrostatic deflection
of a high -voltage picture tube are impractically high,
N
T/ME -
/5750 .SECOND V z 60
/
.52 -CONO
9-2.-At the left is shown the saw -tooth waveform

while the magnetic fields needed for sweeping such
a tube are within reasonable limits. Stating this a
bit differently, we may say that in the larger pic-
ture tubes and in projection tubes very high volt-
ages are used to accelerate the electrons in the beam,
much higher than in the small cathode-ray tubes.
Because of this, the large tubes require higher de-
needed for horizontal scanning and on the right, the waveform
for vertical scanning. While the "trace" voltage is applied to flection voltages or fields, depending upon whether
the deflecting system, the screen is illuminated, but during
"retrace," it is blanked out. electrostatic or magnetic sweep is used. However,
the increase in deflection voltage required by higher
spot" difficulties associated with magnetically de- anode voltages is so great as to make electrostatic
flected tubes are not found when electrostatic de- sweep in the high -voltage tubes impractical, while
flection is used. the necessary increase in magnetic field is not so
An ion spot is a mark on the face of the picture great.
tube produced by localized bombardment of the
marked spot by negative ions. In the case of mag-
netic deflection, the ions are negligibly affected by Sweep Oscillators
the deflecting fields, but are focused in the electron
gun (when electrostatic focusing is used, as is There are two main types of oscillators in use
generally the case) in a manner similar to that today fo produce the saw -tooth waves required in
producing the electron beam. Therefore, they pro- deflection circuits. Both of these have nonsinusoi-
ceed straight down the tube, striking the screen in dal outputs ; that is, they do not produce sine waves.
the center. After prolonged bombardment by these The Potter sweep circuit (named for its inventor)
ions, the screen is damaged, producing a defect in produces saw -tooth waves directly. The blocking
the picture at that point. This is called an ion spot. oscillator either can produce saw -tooth waves di-
In an electrostatically deflected tube, the ions are rectly or can be used in conjunction with another
deflected with the electrons, and so their effect on circuit, employing what is called a discharge tube,
the screen material is spread out over the entire to perform this function.
picture. In a magnetic -deflection system an ion trap (Other circuits are used, of course, besides these
may be used to prevent the formation of an ion two types. For example, you will recall that in
spot (how this is accomplished is described in chap- chapter 8 we considered a horizontal sweep afc cir-
ter 10) but of course the addition of another de-
; cuit, in which the oscillator was of the Hartley type,
vice to a television receiver further complicates it. and produced sine waves. These waves were then
So far we have considered the advantages of passed through circuits which modified them in such
electrostatic deflection. They are of sufficient weight a way as to produce the desired saw -tooth volt-
so that this method is used almost exclusively for ages.)
the smaller direct -viewing picture tubes. Compared The Potter sweep circuit is a special type of mul-
to these advantages, the following are the advan- tivibrator. The multivibrator is an old circuit (con-
tages of magnetic sweeps. Distortion of the picture sidering the recent birth of the electronic art), con-
shape can be kept lower with magnetic sweeps that ; sisting basically of two amplifiers, interconnected
is, with a magnetic sweep, the sides of the picture so that the output of one feeds the input of the
will be straighter than with electrostatic deflection. second, and the output of the second feeds the in-
put of the first. This is shown in block diagram The output of the left-hand amplifier in Fig.
form in Fig. 9-3. No frequency -selective (tuned, or 9-4 is taken from a resistive plate load in the usual
resonant) circuits are included; if they were, this fashion and applied to the grid of the amplifier on
arrangement would give us a sine-wave oscillator. the right. This amplifier functions as a cathode -
In the absence of tuned circuits, the two amplifiers follower, insofar as its output which is fed to the
left-hand amplifier is concerned. We thus find in
OUTPUTS
Fig. 9-4 an arrangement such as that shown in the
AMPLIFIER
INPUTS
J
AMPLIFIEa I
FiG. 9-3.-Block diagram
of a multivibrator. block diagram, Fig. 9-3.
Let us now turn to Fig. 9-5 and consider the
do not produce a smoothly changing output, such as manner in which this circuit operates. To begin
a sine wave. Instead, the output exhibits abrupt with, let us assume that the heaters have been turn-
changes, as though it were produced by the action ed on, but not the plate supply. This being the case,
of a two -position switch. Actually, a multivibrator neither C, nor C, is charged. When the plate volt-
is just that-an electronic switch. age is turned on, several actions occur at the same
Figs. 9-4 and 9-5 show the Potter sweep circuit time. These actions are all dependent upon each
in its simplest form. In Fig. 9-4 this circuit has other, but the one that may be considered basic is
been drawn so that an easy comparison with Fig. the charging of C,. This charges through R,, the
9-3 can be made, and in Fig. 9-5 is a diagram suited cathode -grid circuit of V,, and R4. (An apprecia-
to a discussion of the operation of the circuit. Re- ble charging current can flow from the cathode of V,
ferring to Fig. 9-4, we see on the left an amplifier to its grid because the grid is.made positive at this
of the grounded -grid type, ordinarily found in v -h -f time with respect to the cathode. The grid, there-
or u -h -f amplifiers only. We saw in chapter 8, under fore, acts like a diode plate for the cathode, per-
the discussion of the reactance tube used in hori- mitting the flow of current necessary to charge C, )
zontal afc, that such an arrangement is fundamen- Some charging current will also flow through R,,
tally no different from the conventional grid -input but this contribution is negligible, since the resist-
type. If the potential of the grid is fixed (by ground- ance of R, is much higher than the resistance of
ing it), and the potential of the cathode is varied R4 and the grid -cathode resistance of V, (when the
in accordance with some signal, the grid -to -cathode grid is positive). The charging current through R,
voltage is thereby likewise varied. Ordinarily when lowers the plate voltage of V, below the supply
a signal is applied to the grid of a tube and this voltage, and the flow of this current through R4
signal goes negative, the flow of plate current is raises the cathode voltage of this tube to a value
reduced and when it goes positive, the plate cur-
; considerably above ground. The combination of
rent is increased. In this circuit, however, when the these two factors-low plate voltage and high cath-
ode bias-causes V, to be cut off. This condition ob-
AMPLIFIER i AMPLIFIER
tains immediately after the plate voltage is turned
9+
on, but it lasts for only a very short time, as we
shall see.
As soon as C, starts to charge, that is, as a volt-
FTG. 9-4.-Simple form
of the Potter sweep cir- age builds up across it, the plate voltage of V, rises.
o S cuit, which is arranged At the same time, the charging current falls off.
to correspond with the
gir. block diagram of Fig. (You will recall that when a capacitor is charged
9-3. through a resistor, maximum current flows when the
circuit is first completed. Thereafter, as the voltage
across the capacitor increases, the charging current
decreases.) Since this charging current passes
through R4, the voltage across this resistor de-
signal is applied to the cathode, a positive signal creases with the decreasing current through it.
makis the grid effectively go negative with respect Thus, after the first instant, in which V, is cut off
to the cathode, and so a positive signal decreases by the low plate voltage and high cathode bias volt-
the plate current. When a negative signal is applied, age existing on it, the plate voltage rises and the
this has the effect of making the grid -to -cathode cathode voltage falls.
voltage change in the positive direction ; in this case, It takes only a very short time for these poten-
then, the plate current increases. tials on tube V, to reach such values that V, will
SWEEP CIRCUITS 155
begin to conduct plate current. When this happens tive condition takes place almost instantaneously.
the voltage drop that occurs across R, is transfer- It was pointed out previously that the transfer of
red through C, to the grid of V,. Let us consider plate signal from V, to the grid of V, through C1
just how this transfer occurs. When V, begins to is due to the discharge of C, through R,. This dis-
conduct, this has the effect of suddenly applying a charge takes an appreciable length of time ; this
negative voltage (in series with whatever voltage time is determined by the "time constant" of the
already existed) to the upper plate of C,. When this circuit. The time constant of a series RC circuit is
happens, C, must begin to discharge, since the a measure of the length of time required for the
charge which it already had was such that its upper capacitor to lose a certain percentage of its charge;
plate was positive. In order to discharge C, elec- in such a circuit it is numerically equal to the prod-
trons must flow away from its lower plate. They uct of the resistance in ohms and the capacitance in
cannot flow from the grid of V, to the cathode farads. In this case, the resistance in series with the
(in any appreciable quantity), because of the diode capacitance is R,, R4, and the plate resistance of
action that occurs. there. However, they can flow V,. In practical design, R, is usually much the larg-
through R,. This flow makes the end of R, to which est of these three resistances, so the discharge time
C, and the grid of V, are attached negative with of C, is dependent upon the product R,C,. While
respect to ground. Thus we may say that the sud- this discharge takes place, a negative voltage exists
den voltage appearing at the plate of V, is trans- on the grid of V,. In this connection, another fact
ferred through C, to the grid of V,. should be remembered, namely, that as the dis-
While C, is charging, the grid of V, is positive charge proceeds, the amplitude of the discharge cur -
with respect to its cathode, as we mentioned before. re. -it decreases. As this current decreases, the volt-
Therefore, at this time V2 is conducting. This puts age across R, due to it also decreases. Since this
a partial short circuit across C,, and prevents it voltage is the bias on V,, a point in this decrease
from acquiring any appreciable charge. However, will be reached where V, is no longer cut off. At
when the grid of V, is driven negative by the con- this point, of course, V, will begin to conduct. An-
duction of V1, the plate current in V, is greatly de- other factor in determining when V, again becomes
creased. The voltage drop across R4 due to the plate conductive is the charging of C, through R,. As
current of V, likewise decreases. Since the cathode C2 charges, the plate of V,, which is connected to
of V, is connected to the tipper end of R4, this con- it, becomes more and more positive. Because of this,
dition applies a negative signal to this cathode. it conducts sooner than if the plate voltage were
(Since the upper end of R4 is less positive, it is constant, and only the grid voltage changed.
going negative.) As we saw previously, a negative When V2 starts to conduct, C, discharges through
signal at the cathode of a tube causes the plate cur-
rent to increase. Therefore, the drop across R, in-
creases, and this makes the upper plate of C, go
still further in the negative direction. This negative
signal is transferred to the grid of V, as before.
The cathode of V, then becomes still less positive.
This negative signal from the cathode of V2 causes
a further increase in the plate current of V,. Thus
we have a cumulative action ; the decrease in the
plate voltage of V, puts a negative signal on the
grid of V,, which transfers a negative signal to the SAW TOOT
OUTPUT
cathode of V1, causing an increase in plate current
in V, and a drop in plate voltage, which puts a
further negative signal on the grid of V,, etc., etc.
This cumulative action continues until V, is cut off;
after this the plate current of V2, being zero, can-

not decrease further; and the plate signal of V1 can
no longer be transferred back to its cathode. When
Vf is cut off, the near short circuit that existed
across C, when V, was conducting is changed to an FIG. 9-5.-The Potter sweep circuit. The output of the left-
hand amplifier is fed to the grid of the second amplifier V,,
open circuit. This being the case, C, charges which functions as a cathode follower, as far as its output
through R,. The switching of V, from the conduc- that is fed to V, is concerned.
it, causing a much larger current to flow through resistor. Another minor change sometimes found is
R4 than when V2 is cut off. The resultant large the use of a dual triode tube having a common
voltage drop applies a positive signal to the cathode cathode (such as the 6J6), instead of two separate
of V,. As we have seen, such a signal causes a de- triodes.
crease in plate current. A decrease in the plate cur-
rent of V, will reduce the drop across R,. This sud-
The Blocking Oscillator
den application of a positive potential to the upper
plate of C, causes it to charge through the grid - In Fig. 9-6 is shown a very simple form of the
cathode circuit of V2 as it did when the plate sup- blocking oscillator In addition to the saw -tooth
ply voltage was first turned on. In order for this wave obtainable in its plate circuit, very sharp
charge path to exist, the grid of V2 must, of course, pulses may be derived from the grid circuit. This
be positive with respect to the cathode. Under this property is of great importance, and considerably
circumstance, the tube conducts very heavily in its broadens the utility of this oscillator. For example,
plate circuit, discharging C2 rapidly. When C2 is it is widely used as a pulse generator in radar equip-
discharged, the current through R4 drops to a rela- ment. In addition, as we shall see later, these pulses
tively small value. (R2 is a relatively large resistor, can be very important in sweep circuits.
so only a small current can flow through it.) The In order to see how the blocking oscillator oper-
resultant drop in voltage across R4 applies a nega- ates a very important observation should be made,
tive signal to the cathode of V,. This makes the namely, that this circuit has the sanie form as the
plate of V, go negative. We have thus arrived at common tickler feedback oscillator, with the feed-
the same condition as existed a very short time after back transformer tuned by its own distributed ca-
the plate supply voltage was first turned on. That pacitance. The difference between the two oscilla-
is, C, has been charged and V2 is conducting. As be- tors lies in the choice of values for R, and C,. In
fore, the charging current C, decreases as its charge the usual tickler feedback oscillator, as used, for
increases, causing a smaller drop across R4 than example, in a superheterodyne receiver, the oscil-
existed when C, first began to charge. The sequence lator operates class C, and C, and R, are used to
continues with V, becoming conductive and I'2 be- generate grid bias. The oscillator generates a con-
ing cut off, allowing C2 to charge up once more. tinuous signal at the resonant frequency of the
Thus C2 is charged slowly through R2 and discharg- feedback circuit at the positive peak of each cycle
;
ed quickly through 172 and R4i giving the saw - a little grid current is drawn, putting a charge on
tooth waveform shown in Fig. 9-2. C,. This charge leaks off between positive peaks
When this circuit is used in a television receiver through R,, biasing the grid negative. The strength
or an oscilloscope, the grid of V, is grounded of this bias is dependent upon several factors,
through a resistor, instead of directly. Sync sig- among them the values of C, and R,.
nals can then be injected into the circuit across this In the blocking oscillator, C, and R1 are delib-
erately chosen so large as to interfere with the con-
B+ tinuous production of sine -wave oscillations at the
resonant frequency of the feedback transformer.
Instead of a continuous train of such oscillations,
a series of pulses, each corresponding to a positive
peak of the suppressed sine waves, is formed. The
frequency of these pulses is determined primarily
by R, and C,. How this is accomplished, and how
a saw -tooth wave is formed in the plate circuit, we
shall now see.
SAW TOOTH
OUTPUT Let us begin, as in the case of the Potter sweep
circuit, by assuming that the heater of the tube is
lit, but the plate supply is not yet turned on. This
being the case, C, and C2 will not be charged. When
the plate supply is turned on, plate current will
start to flow in the tube. (C2 will start to charge at
this time, but this effect is very slight, because the
FIG. 9-6.-A saw -tooth wave is obtained from the plate cir- resistance of the plate winding of the transformer
cuit of this blocking oscillator and very sharp pulses may be
derived from the grid circuit. in series with the plate resistance of the now heavi-
SWEEP CIRCUITS 157
ly conducting tube is much smaller than R2.) Since to them is through the high resistance R,. While
plate current is now flowing, where an instant pre- electrons flow from C, through R, to ground (and
viously it was not, there has been a change in the back to the transformer side of C,) to discharge
current through the plate winding of the trans- the capacitor, the grid is kept negative. Since R, is
former. This change in current is coupled into the a much higher resistance than the cathode-to -grid
grid winding, appearing there as a voltage across resistance of the tube when the grid is positive, a
the winding. The polarity of this voltage is such as much longer time is required to discharge C, than
to make the grid positive with respect to ground is required to charge it. This holds the grid nega-
(and the cathode). This causes the tube to conduct tive for a much longer time than it is positive.
still more heavily, causing a further increase or The negative voltage applied to the grid cuts off
change in the same direction in the plate winding, the tube, removing the near short circuit that was
causing the grid to become more positive, causing placed across C, when the tube was conducting. C,
the plate current to increase still more, etc., etc. At is thus allowed to charge through R,, and so a slow,
the same time grid current is being drawn, piling steady rise of voltage takes place across it. This of
up electrons on the plate of C, nearer the grid, course makes the plate voltage more and more posi-
thereby charging this capacitor. The cumulative active ; no plate current is flowing, so there is no volt-
tion which we have described-the interdependent age drop across the plate winding of the trans-
increase of plate current and grid voltage-takes former, and therefore the plate voltage is equal to
place very rapidly, at a rate determined primarily the voltage across C,. At the same time, the dis-
by the resonant frequency of the transformer. How- charge current from capacitor C, is steadily drop-
ever, it cannot go on indefinitely. As the plate cur- ping, since the charge on the capacitor is falling
rent increases, it reaches a value where the rate of off. In turn, the voltage drop across R, due to the
increase drops off; in fact, it tends to stop increas- discharge current also decreases, and thus the grid
ing altogether. This tendency is due to limitations bias becomes less and less. Finally, the grid bias
imposed by circuit impedances and tube character- becomes low enough, and the plate voltage high
istics. enough, so that plate current can flow.
When this happens, it produces a major reaction
in the grid winding of the transformer. You will
As soon as plate current begins flowing, a posi-
tive voltage is applied to the grid through the trans-
recall that the voltage across the grid winding is
former, increasing the flow of plate current-and
determined by the change of current in the plate
thus the chain of events that occurred after the
winding. Up to now, the current has been increas-
plate voltage was first applied is repeated. Once
ing, but now it is starting to decrease. Therefore,
more a near short circuit is placed across C,, dis-
the direction of change has reversed (although the
charging it through the plate circuit of the tube.
current itself is still flowing in the same direction).
This causes the voltage across the grid winding to At the same time, C, receives a large charge
reverse also. The grid is now suddenly driven neg- through the grid circuit. As before, this takes place
ative, and the flow of plate current is decreased
very quickly, and then the tube returns to cutoff,
in which condition it remains a comparatively long
still further by this action. Now another cumula-
time. Thus, C, is allowed to charge for a long time,
tive action is set up; this time the decrease in plate
current drives the grid negative, which further de- but slowly ; when it is discharged, this operation
creases the plate current, which drives the grid
takes but a fraction of the charge time, consequently
further negative, etc., etc. In this manner the tube the rate of discharge is much greater than the charge
changes very rapidly from a condition of heavy rate. This produces the desired saw -tooth wave across
plate -current conduction to one of. cutoff. C,.
When the grid is driven negative, no apprecia- In actual practice there are two modifications to
ble number of electrons can flow from it to the the circuit shown in Fig. 9-6. In this figure one end
cathode. However, a large number of electrons, of the grid winding of the transformer is shown
which flowed through the low -resistance circuit that grounded ; ordinarily, however, it is connected to a
existed from cathode to grid when the latter was source of sync signals. The other modification, al-
positive, are piled up on the grid side of the capaci- though frequently used, is not nearly so universal
tor C,. With the reversal of voltage across the grid as the first. This is the use of an additional triode,
winding of the transformer, these electrons are connected as in Fig. 9-7 (a sync input is also shown
driven away from that plate by the negative poten- here) R4 in Fig. 9-7 completes the d -c path in the
.
tial placed at the opposite plate. The only path open grid circuit of the blocking oscillator, this path hav-
ing been broken to permit injection of the sync sig- Required Sweep Waveforms
nals. C4 shunts R4, so that the a -c impedance of the
Fig. 9-2 shows the manner in which the position
grid circuit is not considerably increased by the
addition of the sync circuit. Without C4, R4 would of the scanned spot on the picture tube must change
increase the charging time of C,. Since the charging with time. If electrostatic deflection is used, the
final sweep amplifier can be connected to the deflec-
time of C, corresponds to the retrace part of the
saw -tooth wave shown in Fig. 9-2, this would have tion plates as shown in Fig. 9-8. The loads on the
final sweep amplifier tubes (their plates only are
an undersirable effect, for it is necessary that the
shown, at the left) are purely resistive; the effects
retrace be very short, as shown.
Looking at Fig. 9-6, we see that the discharge of the d -c blocking capacitors and the input capa-
citance of the picture tube are negligible. Since
path of C2 lies in part through the plate coil of the
the deflection of the spot in the picture tube is di-
transformer. This inductance has a very pro-
rectly dependent upon the voltages applied to the
nounced retarding effect upon the rate of discharge.
deflection plates, a saw -tooth voltage of the shape
Therefore, if the discharge is to take place quickly,
shown in Fig. 9-2 is required to obtain this same
only a partial discharge of C2 can be effected in the
short time allowed. The result of this will be a saw -
tooth wave with a very low amplitude. Conversely, CENTERING
if a large amplitude is required, that is, if the charge CONTROL
and discharge of C2 must cover a large range of

voltage, a relatively long time must be allowed for H V.
POWER
the retrace. SUPPLY
To get around this difficulty, a second tube may

be used as a switch to charge and discharge the
capacitor across which the saw -tooth waveform is
produced. Such an arrangement is shown in Fig.
9-7. Here the left triode is connected in the same FIG. 9-8.-When electrostatic deflec ion is used on the pic-
manner as the triode in Fig 9-6 (except for the ture tube, the final sweep amplifier can be connected to the
deflection plates as shown in this schematic.
addition of the sync input circuit). The grid of the
right triode is connected to the grid of the left
triode, and so they are cut off and conductive si- saw -tooth variation of position. Therefore, since the
multaneously. Since the saw -tooth output is no amplifier loads are resistive, their plate currents
longer obtained across C2, the saw -tooth amplitude and grid voltages must be of the same shape as
here is not important. Therefore the retrace time their plate voltages, namely, a simple saw -tooth
can be made short. On the other hand, since there wave. This permits the use of a simple sweep gener-
is no inductance in the discharge path of C,, a large ator such as the Potter sweep circuit or the block-
part of the charge on C, can be removed very ing oscillator, and the waveform needs no modifica-
quickly. Thus a saw -tooth wave having a high amp- tion after it is generated. The over-all sweep system
litude and rapid retrace is produced by the use of (sweep oscillator plus amplifiers) for use with an
the extra triode. electrostatic -deflection picture tube is, therefore,
e+
rather simple. Let us now compare this system with
those required for magnetic deflection. We shall
see that the vertical deflection waveforms are more
complicated than those used in an electrostatic -de-
flection system, and the horizontal even more so.
Typical Sweep Circuit for Magnetic Deflection

Fig. 9-9 shows the vertical deflection circuit of
the Stromberg-Carlson TV1OP Series 10 and 11
television receivers. In order to simplify the illus-
tration, connections to different points in the posi-
tive low-voltage supply are distinguished by mark-
Ftc. 9-7.-The addition of the second triode in this blocking ing one with a prime and the other with a double
oscillator enables the production of a saw -tooth wave having
a large amplitude and rapid retrace.
prime. The 6N7 tube operates in a blocking oscil-
SWEEP CIRCUITS 159
lator circuit of the type shown in Fig. 9-7. Al- as a type of negative feedback to improve the char-
though there are several differences between the cir- acteristics of the sweep signal. (We say "a type
cuits, there is only one that is more than a matter of" because the nonsinusoidal input to the 6SN7
of detail. Let us consider these differences, dis- is distorted by capacitor C-126-D. However, the
posing of the simple ones first. nature of this distortion is such as to give a signal
In the first place, there are the variable resistors. which helps to cancel some of the distortion which
In discussing Fig. 9-6 we saw that the sweep fre- exists in the input to the 6SN7 in the first place.)
quency is determined principally by C, and R1, and To determine the purpose of R-144 we must first
this is also true of the circuit of Fig. 9-7. In Fig. 9-9, consider the properties of the vertical sweep coil.
the corresponding resistance (R-136 and R-137 in The coil has both inductance and resistance. This
series) is variable. This permits the operator of is a very important point. Another important point
the receiver to adjust the natural frequency of the is that in a magnetic sweep system, saw -tooth de-
oscillator so that the sync pulses can hold it at flection of the electron beam is accomplished by
exactly the right frequency. Hence the name passing a saw -tooth current through the deflection
VERT. HOLD. Potentiometer R-140 controls the coil. This is different from an electrostatic -deflec-
voltage available to charge the capacitor C-128 tion system, in which a saw -tooth voltage is re-
across which the sweep saw -tooth wave is formed. quired across the deflection plates to give a saw -
It therefore functions as the VERT. AMPLI- tooth sweep. Now let us look at Fig. 9-10. At (A)
TUDE control. is shown the saw -tooth current wave required to
No d -c path to ground from the transformer give saw -tooth deflection on the beam. At (B) we
side of C-127 is shown. However, this path does see an inductance and resistance in series, repre-
exist, through the plate circuit and the plate voltage senting the actual inductance and resistance that
supply of the second sync amplifier. exist in a deflection coil. In an actual coil, of course,
We now come to the significant difference, the the two components are not separated as in Fig.
resistor (R-144) in series with the capacitor 9-10(B), but this representation is convenient for
(C-128) across which the sweep signal is generated. our discussion and does not introduce any errors.
Although the lower plate of this capacitor is not Using this convenient separation of the resistive
directly grounded, as in the circuits of Figs. 9-6 and inductive components of the coil, we can de-
and 9-7, it is almost at a fixed potential (the equiva- termine the voltage waveform across the coil when
lent of ground for the sweep signal frequency) be- a saw -tooth wave of current is passed through it.
cause of the large bypass capacitor C -126-D. Act- In Fig. 9-10 (C) are shown three waveforms.
ually, a very small signal exists here. This serves As implied by their captions, they are related to
R-138 PART OF
100K L.-105
R-147
20
VERT =
CENTERING
6N7 R14 4
4.7 K
C-128
0.1µr
-
R-136
1.5 MEG R-145
R- 37 C -126-D 2K
1MEG 5Oµ1 R-143 VERT.
VERT. 470K LIN.
NOLD R-146
560
FIG. 9-9.-The vertical deflection circuit of the Stromberg-Carlson model TV10P, Series 10 and 11.
the waveform at (A) and the coil at (B). We know figures are right angles, but in the square there is
from both theory and experimental observation that the additional specification that all sides are equal.
when a steadily changing current (that is, a current The voltage waveform across the resistive com-
having a linear rise or fall) is passed through an ponent is the same as the current waveform that
inductance, it produces a fixed steady voltage across produces it. This follows from Ohm's law : E = IR.
the inductance. This voltage is proportional to the Since the resistance is fixed, the voltage must vary
rate at which the current is changing and to the in exactly the same manner as the current.
size of the inductance. Since our inductance here When a rectangular wave and a saw -tooth wave
is fixed in value, ariy change in the voltage wave- are combined as shown in Fig. 9-10 (C), the result-
form across it must be due to a change in the cur- ant is known as a trapezoidal wave. This, then is the
rent through it. And the current does change ; first voltage wave that must exist across the deflection
it rises fairly slowly, that is, it changes in the posi- coil when the necessary saw -tooth current passes
tive direction. As a result, there is a positive volt- through it. In the coils ordinarily used, the resist-
age across the inductance which endures for exact- ance is small, and so the resistive component of
ly the same length of time as the current rises. voltage is much smaller than the inductive. The
Then, abruptly, the current starts to decrease, that resultant trapezoidal wave, therefore, does not have
is, it changes in the negative direction. At the time as pronounced slopes as does that shown in Fig.
that this change in the current wave takes place, 9-10(C), but is almost a rectangular wave. How-
there is a corresponding change in the voltage wave ; ever, its difference from a simple wave is sufficient-
and since the current is changing steadily in the ly great to be significant.
negative direction, the voltage is steady at a nega-
Having determined the voltage waveform that
tive value. The duration of the negative swing of
must exist across the coil in order to achieve the
the voltage is exactly the same as the duration of
desired saw -tooth deflection, our next problem is
the negative change of current that produces it. Thus
how it is generated. This is where the mysterious
alternating fixed positive and negative voltages are
resistor, R-144 (see Fig. 9-9), enters the picture.
produced by the passage of a saw -tooth wave of
Fig. 9-11 illustrates its function. Part (A) of this
current through an inductance. This voltage wave-
figure indicates the charging current of C-128 as a
form is sometimes known as a rectangular wave ;
positive current, which flows for a relatively long
it is also sometimes called a "square wave," but it
time, and the discharge current as a negative flow
is more exact to restrict this latter term to a wave
having a short duration. The combination of R-144
similar to that shown in which the durations of the
and C-128 is shown at (B) for ready reference. In
positive. and negative swings are equal. This usage
part (C) are found the individual voltages existing
stems from the distinction between a rectangle and
across the components shown in (B) and the re-
a square in both all the angles in the corners of the
;
sultant voltage.
As we have already pointed out, it follows from
Ohm's law that the waveform of the voltage across
a resistor is the same as the waveform of the cur-
VOLTAGE ACROSS INDUCTANCE
rent through it. This accounts for the rectangular
CURRENT THROUGH COIL voltage wave across R-144. We have also seen that
(A) a steady flow of current charging or discharging a
EQUIVALENT capacitor produces a steadily rising or falling volt-
SERIES age across it. Hence the saw -tooth wave of voltage
VOLTAGE ACROSS RESISTANCE
INDUCTANCE across C-128. The resultant of these two waves is a
OF COIL
trapezoidal wave, just as in Fig. 9-10(C). Thus the
addition of R-144 modifies the output of the block-
EQUIVALENT ing oscillator in the manner required for a magnetic
SERIES
RESISTANCE sweep.
OF COIL RESULTANT VOLTAGE Another interesting feature of the circuit shown
ACROSS COIL. in Fig. 9-9 is the VERT. LIN. (vertical linearity)
(B)
(C) control. The function of this adjustment is to make
FIG. 9-10.-When a saw -tooth current (A) flows through the vertical deflection of the electron beam as near-
a coil, the combined waveforms of the voltages due to the ly perfect a saw -tooth wave as possible. The wave-
inductive and resistive properties of the coil (B) produce the -
resultant waveform shown at the bottom of (C). forms shown in Figs. 9-10 and 9-11 are, of course,
SWEEP CIRCUITS 161
idealized. Actually, there will be some curvature in movable arm is above or below the fixed tap, and
them; but if distortion of the picture on the face how much flows depends upon how far the arm is
of the cathode-ray tube is to be minimized, this cur- from the tap. In this way a fixed component of cur-
vature must be made as slight as possible. rent through the coil can be added to the sweep cur-
To reduce the curvature, advantage is taken of rent, and this determines where the center of the
a characteristic of vacuum -tube amplifiers that is picture will lie in the vertical axis. Hence the name
usually a source of trouble. This characteristic is of the control.
the curvature of the grid voltage-plate current char- Resistor R-148 is a damping resistor, used to re-
acteristic, which causes amplitude (or harmonic) duce the tendency of the vertical deflection coil to
distortion. By good fortune, this curvature is op- produce unwanted oscillations. When a sharply
changing voltage wave (such as the trapezoidal
wave used here) is applied to coil, there is a likeli-
hood that oscillations will be produced, since the
CNARGEe DISCHARGE
coil is tuned by its distributed capacitance. This
CURRENT VOLTAGE ACROSS R-144
effect is known as ringing.
(A)
The GE model 802 television receiver employs
VOLTAGE ACROSS C-128
magnetic deflection of the electron beam in the picture
tube. The vertical deflection circuit is quite dissimilar
R-144
in detail to that used in the Stromberg -Carlson sets
just described. However, the principles are very much
TC -128 the same. For example, the output tube is a 6V6-GT
RESULTANT OUTPUT 1
in the GE set, instead of a 6SN7 with its two sections
(a) VOLTAGE OF DISCHARGE". in parallel as in the Stromberg -Carlson receivers. But
TUBE
in both circuits, the vertical linearity control is lo-
FIG. 9-11.-When a rectangular current wave (A) is ap-
cated in the cathode circuit of this stage; and in
plied to a resistor and capacitor in series (B), the individual both there is a small corrective signal fed back from
waveforms are as shown in (C) with the resultant waveform the cathode to the grid. A somewhat more pro-
at the bottom.
nounced difference is the use of a multivibrator in
the GE receiver as the sweep generator ; and an-
posite to the curvature of the sweep waveform gen- other is the manner in which a trapezoidal sweep
erated by the blocking oscillator, and so the two waveform is obtained. Here the sweep generator
tend to Cancel. You will recall that the amplitude produces an unmodified saw -tooth wave, and this
distortion produced in an amplifier is dependent is applied without change (except for the small
on the bias, and that the bias can be controlled by corrective feedback) to the grid of the output amp-
varying the cathode resistance. R-145 (the VERT. lifier. However, in the plate circuit of this tube ân
LIN. control) accomplishes this function, and so effect is emphasized which is carefully avoided in
permits adjustment of the curvature of the 6SN7 the Stromberg -Carlson receivers. This is ringing in
characteristic to provide the best compensation for the inductive load (output transformer and deflec-
the curvature of the output of the blocking oscillator. tion coil.) It is emphasized by the addition of a ca-
The output of the 6SN7 is coupled to the vertical pacitor across the secondary of the transformer.
deflection coil (part of the deflection yoke, L-105) If you will look once again at the rectangular
through a transformer. This transformer provides waves shown in Figs. 9-10 and 9-11, you will see
an impedance match, just as an output transformer that they are in the nature of pulses. This resem-
in an audio amplifier matches the voice coil of the blance would be even more distinct if the duration
speaker to the plate circuit of the final stage. In the of the negative swing were a good bit shorter. On
secondary circuit of the output transformer we find the basis of this, it may be said that a trapezoidal
the VERT. CENTERING control. This is a small wave can be produced by adding suitable pulses to
potentiometer (20 ohms) with a fixed center tap as a saw -tooth wave. This is done in the GE model
well as the movable arm. Current is drawn through 802. During the retrace time the grid voltage of the
it by one of the power supplies, and by proper ad- output tube .(V20, see Fig. 9-12) changes from its
justment of the control, part of this current can be maximum positive voltage to its maximum negative
made to flow through the deflection coil. Which way voltage. This negative voltage swing is sufficient to
this current will flow depends upon whether the cut the tube off momentarily and cause ringing. The
capacitor (C44) across the secondary of the output tion coil is stepped up in an autotransformer to a
transformer (T20) is rather large (0.1 µf), so that very high voltage. It is then rectified, and provides
a half cycle of the ringing frequency lasts long 8200 volts for the second anode of the picture tube.)
enough for the output tube to return to its conduc- The HORIZONTAL LINEARITY control
tive state. When the tube conducts, its plate resist- changes the time constant of the circuit consisting
ance is sufficiently low to damp out further oscilla- of capacitor C94 and resistors R90 and R115. In
tion. In this manner a good approach to a trape- this manner the shape of the signal applied to the
zoidal wave is obtained, using half sine-wave pulses grids of the damping tube is modified, and this in
turn affects the shape of the sweep signal during
FIG. 9-12.-In the GE and immediately after the retrace. Resistors R120
model 802, this circuit is
TO VERT.
DEFLECTION used to produce a trape- and R121 prevent the grid current in the damping
COIL zoidal waveform by add- tube from rising to an excessive value when the
ing suitable pulses to a
saw -tooth wave. positive pulses due to ringing are applied to it.
The WIDTH control is simply a variable induc-
instead of flat-topped. straight -sided pulses in com- tance, similar to the deflection coil and in series
bination with a saw -tooth wave. This deviation from with it. Because of the series connection, part of
the ideal trapezoidal shape of the sweep voltage the sweep voltage appears across one coil, and part
produces some distortion of the sweep. However, across the other. By varying the inductance of
this distortion occurs during the retrace time, when L23 the operator of the set can change the propor-
the beam is blanked, and so its effect on the picture tion of the sweep signal appearing across the de-
is negligible. flection coil and, therefore, the amplitude of the
Ringing is put to similar use in the horizontal horizontal sweep.
sweep. This is quite common practice in modern C 94
television receivers using magnetic sweep, but it 47

R120 8121
has been found necessary in this case to use a sep-
arate tube to provide damping action. The necessity
R90
47K
2200
FAN 2200
NOR.
for the separate tube lies in the much higher sweep FROM
V2 3 DEF.
frequency used in the horizontal circuit, 525 times NOR. ans 6Á57G COIL,
250K PART
the vertical sweep frequency. The damping tube and OUTPUT
TR ANSF. HOR.LIN. R122 OF Ill
100 K
its associated components are shown in Fig. 9-13.
This circuit is treated at soiree length in chapter 11, L23 C96 R123 CSOC
300 - 401JF
but there are certain points in connection with its .01 {14
operation that it is well to mention here. (This cir- WIDTH

cuit is discussed under power supplies because the FIG. 9-13.-The damping tube and its associated compo-
sinusoidal pulse produced by ringing in the deflec- nents in the horizontal sweep circuit of the GE model 802.
CHAPTER 10
THE PICTURE TUBE
BY WILLIAM BOUIE
As stated in chapter 1, the picture tube which used in test oscilloscopes and other devices used
displays the televised images has essentially the in studying and photographing observed electrical
same design as the conventional cathode-ray tube. phenomena.
For those who desire to investigate further the The screens used in television picture tubes are
theory and functioning of cathode-ray tubes, ref- formed from a coating of phosphor having a white
erence is made to Rider's The Cathode -Ray Tube or bluish -white cast. These screens have a low per-
At Work, which contains a wealth of material on sistency, the glow ceasing when the electron beam
these tubes. Our discussion here will be limited to is removed from a point. This low persistency char-
the cathode-ray tube as used in television reception. acteristic is an absolute requirement in order to
Two types of picture tube are in popular use at prevent blurring of moving objects in the television
the present time. These are the direct -viewing-type picture. As stated in chapter 1, the screen material
and the projection -type tube. In the direct -viewing is usually coated with a thin aluminum backing
tube, the received image is displayed on the face which increases the brightness of the television
of the tube and is viewed directly. In the projec- image and stops the heavy ions in the electron beam,
tion -type tube, the image on the tube screen is di- thereby preventing formation of the ion spot. The
rected through a lens system and is then projected formation of the ion spot will be discussed later in
upon a large viewing screen. Both of these tube this chapter.
types will be discussed in this chapter.
The type of picture tube used for direct viewing
Persistency of Vision
varies in size from the 5 -inch tube which is incor-
porated in the small table -size models to the 20 -inch The electron beam causes a fluorescent spot to
tube used in the large console television receivers, appear upon the screen only at the point where the
with even larger tubes being used for experimental beam strikes the screen. It is this single spot, acting
work. An illustration of a 10 -inch picture tube, a under the influence of the deflecting fields and the
popular size in the moderate -priced television re- television signal on the grid, that traces the pic-
ceiver, is shown in Fig..1-5. ture on the face of the tube.
The action of the deflecting fields is such that
the electron beam is made to shift its position by
Screens
small increments and to produce a fluorescent spot
The luminescent screen material which is coated which appears periodically at every point on the
on the inner surface of the large end of the picture screen. This spot varies in intensity with the sig-
tube performs a very important function in the nal voltage applied to the grid of the tube, and rap-
operation of the tube. It is this material that con- idly traces the television picture. The progressive
verts the electrical energy in the accelerated elec- appearance of the fluorescent spot at all points on
tron beam into light whose intensity varies with the screen creates the illusion of an instantaneous
the intensity of the television signal modulating the picture by persistence of vision and, while it is
electron beam. The efficiency of this conversion is true that the fluorescent spot exists at any one point
largely dependent upon the type of material used in for a very small fraction of an instant, it seems to
the formation of the screen, and the thickness of the observer that the entire area of the screen is
the coating. Many different types of oxides and made fluorescent at the same instant. In other
phosphors are used in forming the screens of words, the entire picture on the tube appears to
cathode-ray tubes and each has its value in specific occur instantaneously instead of being constructed
applications. The screen can be made to glow for of a series of infinitely small fluorescent spots of
long or short periods of time (persistency), usually light moving with great rapidity horizontally and
in shades of white, green, or blue. These screens are vertically across the face of the tube.
163
This persistency of vision, which is a character- less the beam is properly focused. An incorrectly
istic of the human eye, is an optical illusion created focused beam results in a large spot with a halo
by the inability of the eye to respond to changes around it. In some instances, improper focusing re-
which occur faster than a certain rate, usually one - sults in a very wide spot which, upon being de-
tenth of a second. Persistence of vision can be illus- flected, produces extremely wide lines of non -uni-
trated by whirling a lighted flashlight in a circular form width on the screen.
motion in a dark room. As the speed of rotation is
increased beyond a certain point, a ring of light Optical Analogy of Focusing
will be observed in the darkness. The circular speed
of the flashlight and the inability of the eye to spot Suppose that we consider a simple optical sys-
the light beam at any one position, because of its tem, such as that shown in Fig. 10-2. The variables
rate -of travel, causes the illusion described as per- which present themselves in this analogy are very
sistence of vision. The motion picture is, itself, an much like the variables which are found in the tele-
optical illusion. Actually the subjects on the screen vision picture tube. This will become evident as we
do not move, although the eye sees continuous mo- continue the discussion. This optical system con-
tion. What really happens is that a series of still sists of a source of light, L, a stop, or light gate, S,
pictures is presented, each of which is stationary for two lenses, O and P, and a screen Sc.
a very short time, usually 1/24th of a second. The The purpose of the system is to project the light
eye cannot follow change at this speed, hence the from the lamp, L, upon the screen, Sc, in the form
motion appears continuous. of a small but intense spot of light. The nature of
the light gate, or stop, is like that of a stop on a
camera; that is, it is a device with an opening of
Focusing variable size. By means of this device the amount
Fig. 10-1 shows the elements contained in the of light passed from the lamp, L, to the lens, O, is
picture tube. These elements, or electrodes, all indefinitely under control. At the same time, we can
fluence the stream of electrons in the formation of further assume that the intensiy of lamp L, is also
the television picture. Upon application of the prop- under control. In other words, a rheostat can be as-
er voltages, electrons are emitted from the cathode sumed to exist in the power supply circuit feeding
and they spread outward in all directions. In order the lamp.
to form the spot of light which is used to trace out The two lenses are of the type which will pick
the image on the tube screen, these electrons must up the light from some source, concentrate the
be concentrated into a beam. This action is called beam, and project a pin point of light upon a screen
focusing the electron beam. Proper acceleration at a definite distance from the lens. In the process
of these electrons is also required so that when of projection from lens P, the light from lens P is
they strike the viewing screen the impact will be caused to converge into a spot upon the the screen.
sufficiently great to cause fluorescence. Focusing is In projecting the spot, let us consider the first
also a function of this acceleration. requirement, namely, the intensity of the spot. This,
If the emitted electrons are not concentrated into quite obviously, depends upon the amount of light
a beam, the image position will not be under con- available from L. If the original light source is not
trol. Neither will it be possible for the electrostatic sufficiently luminous, all the concentration in the
or magnetic deflection fields, whichever are used, world is not going to produce a sufficiently lumi-
to act uniformly to produce an image of value un- nous_ spot. On the other hand, although the light
Fic. 10-1.-The elements of

a cathode-ray tube: H is the
heater; C the cathode; G the
control grid; A, and AE, the
first and second anodes re-
spectively; H1, Ht and V1,
V2, the horizontal and verti-
cal deflection plates, respec-
tively and SC, the screen.
;
THE PICTURE TUBE 165
source may be sufficiently intense, any adjustment, and lens O will result in an optimum setting, which
for example some setting of the stop S may reduce will provide greatest intensity and greatest defi-
the amount of light conveyed from L to lens O, nition in other words, a spot of the proper inten-
;
will interfere with the intensity of the spot ; and if sity is now properly focused upon the screen. In-
this adjustment is such that the light transferred correct adjustment of stop S with correct adjust-
is not sufficient, the intensity level of the spot upon ment of lens O will result in a properly focused
the screen will not be suitable. Bear in mind that at spot of light upon the screen, but one of insufficient
this time we are concerned solely with the intensity brilliance. On the other hand, correct adjustment of
or the brightness of the spot or image. the stop and incorrect adjustment of lens O will re-
sult in light upon the screen, but not of the correct
intensity or focus. The intensity also suffers when
the light is out of focus because proper concentra-
tion does not take place. Instead of having all the
light concentrated into one spot, the rays picked up
o P
by the lens are diffused over a greater portion of
FIG. 10-2.-An optical system which is analogous to the ele- the screen, with reduced light at any one point.
ments of a cathode-ray tube which bring the electron stream
to a focus on the screen. It is significant to note that with the position of
the light source L, the opening of stop S, and the
screen position permanently fixed, the projection
Two variables are directly associated with the in- of the best defined spot is determined by the rela-
tensity or luminosity of the spot. If we desire to tive positions of lenses O and P. Such focusing can
vary the intensity of the light, we have two variable also be done with a single lens in the place of O
controls, either or both of which may be used. If and P.
we should decide to keep the brilliancy of the lamp
fixed, adjustment of the opening of the stop S
Electrostatic Focusing
would give us the desired control. On the other
hand, if we desire to keep the adjustment of the Let us now see how closely the optical analogy
stop opening fixed, the filament control in the lamp of focusing holds for the electronic method em-
circuit would give us the intensity control. There ployed in the television picture tube. Referring to
is no reason for a dual control of this type, so we Fig. 10-1, H is the heater which raises the cathode
can forget for the moment any variation in the temperature to the proper operating point for the
brilliancy of the lamp and use only the stop open- emission of electrons. C is the cathode, which emits
ing as our intensity control. When the stop is the electrons and is equivalent to the light source in
ofkned wide, the full intensity is realized, because Fig. 10-2. G is the control grid and is equivalent to
the maximum amount of light goes through. When the light gate, or stop. A, and As are the first and
the stop opening is closed, no light passes through. second anodes which are operated at high positive
Any setting of the stop opening, between wide open potentials. These electrodes operating in unison are
and fully closed, will provide a light intensity vary- called an electron gun because they concentrate the
ing between the two limits. electrons into a narrow beam and direct this beam
For any one particular distance between the light against the screen with sufficient force to cause the
source and the screen, there is one setting of the required degree of fluorescence.
lens, which will result in the most sharply defined The control grid differs from the conventional
area of light. This operation or adjustment of the grid in an ordinary vacuum tube, being a small hol-
lens to most sharply define the rays of light upon low cylinder. The end of the cylinder toward the
the screen is known as focusing. cathode is completely enclosed except for a small pin-
In connection with the optical analogy offered to hole through which the electrons pass. Just as stop S
illustrate focusing in the television picture tube, we in Fig. 10-2 serves as a light gate, the control grid
are concerned with light rays which converge to functions as an electron gate or stop because of the
a pin point. The adjustment of lenses O and P con- following. The electrons which are emitted from
stitutes focusing with two variables. If lens O is the cathode are negative particles. For any one
fixed in position, focusing can be accomplished by temperature of the cathode, the maximum number
moving lens P. On the other hand, if lens P, is of electrons are emitted for that temperature. If we
fixed, all the focusing can be done with lens O. desire to control the intensity of the spot, we must
With lens P fixed in position, adjustment of stop S have some means of controlling the number of elec-
trons allowed to pass into what finally becomes the the screen irrespective of the adjustment of the
beam. The intensity of the spot upon the viewing anode A, voltage. This corresponds to the case of
screen is dependent upon the number of electrons in the simple optical system shown in Fig. 10-2, where,
the beam, just as the intensity of the spot in the op- if light is passed through the stop, it will be visible
tical analogy is dependent upon the amount of light upon the screen even if the adjustment of lens O
allowed to pass to the lenses. is not correct.
Control of the number of electrons which are The voltage applied to anode A, is the primary
permitted to pass through the electron gate is ac- focusing voltage, since the first focusing action
complished somewhat in the manner employed in takes place in the region of the first anode. Anode
the regular vacuum tube, that is, by the application A, is a cylinder having a small aperture in the end
of a negative bias to the control grid. The nega- nearest to A2. The initial beam of electrons is con-
tive bias applied to the grid repels the electrons verged in A, for passage through this aperture.
emitted from the cathode and definitely controls the Those electrons which are too divergent from the
number which pass through the pinhole aperture. beam do not pass through the aperture. A compari-
Reducing the bias naturally allows a greater number son between the simple optical system and the elec-
of electrons to pass through aperture and thus pass trical optical system in the picture tube shows a
under the influence of the adjacent electrode. A,. parallel between L, S, and O in Fig. 10-2 and C, G,
The greater the number of electrons permitted to and A, in Fig. 10-1.
pass through the control grid, the greater the in- The main focusing action in the picture tube
tensity of the spot which appears upon the viewing takes place between anodes A, and A2. A fixed high
screen. voltage is applied to anode A2. This voltage is pos-
The electrons emitted from the cathode have a itive with respect to the cathode and much greater
certain velocity but not the velocity required. There- in value than the voltage applied to anode A,. This
fore, a positive voltage is applied to A, to attract voltage on A, accelerates the electrons to the velo-
sufficient quantities of electrons and to accelerate city required to produce fluorescence when they
their movement through the electron gate G towards strike the screen. The electrons which pass through
the viewing screen. The pinhole aperture and the the beam -defining aperture in A, are brought to a
negative bias on the control grid plus the attracting focus as a result of the combined action of A, and
influence of anode A, combine to produce the initial Af, but at this time they are caused to converge into
converging force, for the diameter of the electron a spot at a specific distance from A2, namely, upon
beam passing through the control grid is much less the screen. Anode A, also has a beam -defining aper-
than the beam diameter at the cathode. ture, so that electrons which are too divergent from
The control grid electrode serves another pur- the main beam are not permitted to pass througii.
pose in addition to controlling the number of elec- During the discussion of the optical system, lens
trons permitted to pass through the aperture. As P was fixed and focusing was accomplished by ad-
a result of the negative charge applied, this grid limits justment of lens O. With the position of L and the
electron emission from the outer portions of the adjustment of S also fixed, it is obvious that fo-
cathode. Only the center portion of the cathode, cusing of the light upon the screen Sc is governed
parallel to the aperture, emits freely. Of course, by the relative positions of the two lenses, O and P.
the exact extent of this condition depends upon the A corresponding condition is true in the picture
bias adjustment. The control grid, in conjunction tube, but in this case, with cathode emission fixed,
with the cathode and the first anode A, which ap- the focusing of the beam upon the screen is deter-
plies the initial acceleration force to the emitted mined by the ratio of voltages applied to anodes
electrons, constitute an electronic lens. A, and A,. With a fixed voltage on A2 the proper
From what has been said, it is clearly evident focusing adjustment is made by varying the volt-
that the primary control of the electron beam is the age applied to anode Al.
intensity control, that is, the control grid electrode We made the statement that if the intensity ad-
bias adjustment. When the bias adjustment of the justment (the bias applied to the control grid) was
control grid is such that no electrons, or insufficient correct, electrons would strike the viewing screen re-
electrons, pass through the electron gate aperture, gardless of the setting of the focus control which
no electron beam/ will be formed. On the other varies the voltage applied to A,. It is also impor-
hand, if sufficient electrons are made to pass tant to remember that if the intensity adjustment
through the electron gate, a spot will appear upon is such that insufficient electrons are passed, or if
the emission from the cathode is insufficient, varia- low -voltage power circuit, and the current through
tion of the focusing voltage may cause the spot to the coil is controlled by a rheostat. Focusing is ac-
disappear from the screen. There are several pos- complished by varying the amount of current flow-
sible reasons for this. One is that as a result of in- ing through the coil until the point of optimum
correct focusing, insufficient acceleration is given focus is reached. This magnetic method of focusing
to the electrons and comparatively few reach the has the advantage of reducing certain detrimental
high voltage anode At. Another reason is the diffu- effects which occur on the screen of the tube. These
sion of the beam, which results in less than the re- detrimental effects caused by a condition known as
quired amount of light appearing on the screen. ion spot will be covered in more detail later.
The production of a bright and sharply defined In some instances the internal structure of the
spot upon the screen is accomplished by correctly picture tube may be considerably different from the
controlling the intensity and the focusing of the tube shown in Fig. 10-1. Some types of picture
electron beam. The size of the spot is influenced by tubes have a graphite coating on the inner surface
both the intensity and focusing controls. beginning near the fluorescent screen and extend-
ing back to anode A,. This conductive coating is in
some cases coñnected internally to the second anode
Magnetic Focusing
and in other cases is brought out to a separate ter-
In our discussion on focusing the electron beam minal located on the side of the tube. A positive
in the picture tube, we have described the method voltage equal to, and at times greater than, the
using electrostatic fields. However, the same results voltage on A2 is applied to this coating. The coating
can be obtained by means of magnetic fields. This serves as a return circuit for the electrons that
reference to magnetic focusing must not be con- bounce back after contacting the screen, and also as
fused with the application of magnetic fields for the an additional anode providing greater accelerating
purpose of deflecting the beam in the formation of force to the electron beam.
the image. This reference relates to the proper con- Quite often where magnetic focusing is em-
centration of the electron beam in order to provide ployed, anode A, is omitted entirely in the fabrica-
a small, sharply defined spot on the screen. tion of the tube, and this coating, which is then
Most of you will recall from your study of ele- brought out to a separate terminal, serves as anode
mentary electrical theory, that a circular magnetic A2. The focusing coil is positioned in the neck of
field is established around a conductor carrying an the tube at the point where the electron beam nor-
electric current. By placing a current -carrying coil mally comes under the influence of anode A2.
around the neck of the picture tube near the loca-
tion of anode A,, the magnetic field can be made to
Deflection of the Beam
act upon the beam of electrons within .the tube. The
effect of this action is to force divergent electrons There are two very important functions per-
to assume a helclal movement which terminates in a formed in the picture tube. The first is the focusing
sharply defined spot on the screen when the beam of the electron stream into a narrow concentrated
is focused properly. This helical movement is the beam of electrons. The second is the subjection of
result of two forces. The first force is a property this beam to the influence of an electrostatic or
of the electrons diverging from the electron beam magnetic, or combination of both, field in order to
which is being accelerated by the positive voltages secure the proper deflection of the beam. Deflection
on anodes A, and At. The second force is that of of the beam results in a scanning pattern, or raster,
the magnetic field which acts at right angles to the on the screen of the picture tube.
first force. As long as the electron beam travels We made the statement that two types of deflec-
along the axis of the picture tube, it is not affected tion are available, namely electrostatic and mag-
by the lines of force of the magnetic field. However, netic. Electrostatic deflection is employed in the
when the beam passes out of the converging in- smaller picture tubes (5 to 7 inches), but because
fluence of the positive anode, electrons on the sur- of the extremely high deflecting voltages that would
face of the beam tend to spread out. It is these be required, magnetic deflection is used in tubes
electrons that come under the influence of the mag- having larger diameters. Some of the factors to be
netic field which throws them into a helical move- considered in deflection are the velocity of the elec-
ment in the effort to force them back into the origi- tron beam and the intensity of the electrostatic or
nal axis of the beam. magnetic fields employed. Also, it must be remem-
The focusing coil is connected in series with the bered that electrons are particles bearing a negative
charge and as such are repulsed by a negative field age to the upper plate V, and a negative voltage to
or attracted by a positive one. In addition, electrons the lower plate V,. With such connections, the posi-
in motion are surrounded by a magnetic field in the tive charge upon V, will attract the stream of elec-
same manner as a current -carrying conductor, and trons towards it, whereas the negative charge upon
can be affected by external magnetic forces. the lower plate V, will repel the stream, so that the
In the discussion of the deflection of the electron entire beam will be deflected toward the upper
beam, we will consider first deflection with an elec- plate V,.
trostatic field and then with magnetic fields. As far The normal direction of the beam is as shown by
as deflection is concerned, the nature of the electron
the middle dotted line in Fig. 10-3 and the position
emitter and the type of accelerating anode or anodes of the spot upon the screen is O. When the differ-
used are of no consequence. The primary consid-
ence of potential previously described exists between
eration is the existence of the electron stream with- the plates, the beam is pulled out of line as it passes
in the deflecting field.
between the plates and the spot upon the screen is
moved to the point indicated by the number 1, the
Electrostatic Deflection beam being shown as a dotted line. The extent that
the spot is moved up away from its normal position
Electrostatic deflection is accomplished by the is determined by the magnitude of the positive
application of voltages to the deflecting plates in
charge upon this plate, with respect to the lower
the picture tube. These plates are identified in Fig.
plate.
10-1 as H,, HE, V,, and V2. The position of the
The limitation of the movement of the spot, or
electron beam without any voltages applied to these
deflection of the beam, with a certain potential dif-
plates is midway between them ; this places the spot
ference between the two plates, is a result of the
in the center of the viewing screen. This position
voltages applied to the accelerating anodes. These
of the spot midway between the plates also exists
voltages project the electrons in a beam toward
if no potential difference exists between the plates.
the screen. Since the velocity of these electrons in
In other words, if both plates of each pair are at the
motion is very great, a certain force is necessary
same potential, the effect upon the beam is as
though there were no voltage whatsoever applied to to deflect them or pull them away from their nor-
mal line of travel. The potential difference between
either pair of plates.
To appreciate properly what happens in the pic- the deflecting plates is the force which tends to pull
ture tube when a potential difference exists between the stream out of its normal axis of travel. The
the plates of each pair of deflecting plates, it is higher the accelerating voltage which starts the
necessary to consider each pair separately. Particu- beam in motion, the greater the force required to
lar reference was made to the fact that the electron pull it out of line. It should be understood that
stream or beam consists of a group of moving neg- there are also other factor present in the tube
ative charges ; also, that these charges are repelled which may offset the effect of the Sigh accelerating
voltage, so that two tubes of unlike accelerating
by similar negative charges or by a negative field
and that they are attracted by positive charges or a voltages may require a like value of deflecting volt-
age in order to move the beam the same distance.
positive field. Suppose then, that the beam of elec-
trons is moving between the two deflecting plates The degree of sensitivity to deflection is known as
the deflection factor of the tube and varies with
as shown in Fig. 10-3. By means of connections to
these plates, it is possible to apply a positive volt -
different tubes.
If the voltage to the two deflecting plates is re-
versed as shown in Fig. 10-3, so that V, is negative
and V, is positive, the beam and the spot shift in a
direction opposite to that which took place before.
The basis of this shift is exactly as stated before,
except that the polarity of the deflecting plates has
NOTION OF
BEAN AND SPOT;
OR ELECTRIC
been changed. Now the electron beam is attracted
FIELD DIRECTION
by V, and is repelled by V,. Since the hypothetical
voltage applied to the two deflecting plates is the
FIG. 10-3.-When a positive voltage is applied to V, and a same value as before, the potential difference be-
negative voltage to Vt, the spot caused by the beam will rise tween the plates is the same and the spot moves the
to point 1 from 0 at the center of the screen. When the volt-
ages are reversed in polarity, the spot will descend to point 2. same distance in the opposite direction to point 2.
If a device is used whereby the voltage applied to force of strong magnetic fields in order to deflect
the two plates is automatically reversed at a very it across the screen of the picture tube. In most
slow rate, you would see the beam shift from posi- tubes using magnetic deflection, the two pairs of de-
tion O to position 1, back through position O to flecting plates are left out of the tube, and a mag-
position 2, thus tracing a line of light on the screen. netic deflecting yoke is placed around the neck of
At this time we want to call to your attention one the tube at approximately the position occupied by
very significant fact. The electric field existing be- the deflecting plates.
tween these plates is like the electric field existing This magnetic yoke consists of two pairs of de-
between the plates of a two -plate capacitor and a flecting coils mounted at right angles to each other,
specific amount of capacitance does exist between a in the manner of the deflecting plates used for
pair of deflecting plates. The electric field exists electrostatic deflection. The two coils comprising
perpendicular to the plane of the plates, sa that the each pair are connected in series and are mounted
plates which are horizontal with respect to the front opposite to each other on the neck of the tube. When
of the tube cause the vertical deflection. The motion current is passed through the coils, the magnetic
of the beam is parallel with the electric field. The field which is created bends the electron beam in
term "vertical deflection plates" refers to the motion the same manner as the electrostatic field previously
given to the beam by those plates, with respect to discussed.
the normal position of the tube, and not to the In operation, deflecting currents from the sweep
orientation of the plates inside the tube. This ap- circuits are passed through the coils, causing the
plies to the term "horizontal deflection plates" as electron beam in the tube to trace out the raster
well. on the face of the tube. The deflecting yoke must be
The action we have described in regard to the adjusted in position on the neck of the tube so that
vertical deflecting plates also occurs when a poten- the picture will appear in correct position for the
tial difference exists between the horizontal deflec- viewing audience.
tion plates, H, and H2 in Fig. 10-4. With H, posi -
Combined Methods of Focusing and Deflection
Quite often combinations of electrostatic focus-
ing and magnetic deflection and vice versa will be
encountered. It has been stated that on the larger
MOTION OF REAM
AND SPOT1 w picture tubes, magnetic deflection is used. One rea-
OR
ELECTRIC FIELD
DIRECTION
son for this is that magnetic deflection provides
uniformity of focus at all points on the viewing
10-4.-If horizontal deflection plate H, be made posi- screen, which is not true in the case of electrostatic
tive and Hs negative, the spot will move to the right from the deflection. Another reason is that the hazard of
center of the screen 0 to point 1. When the polarity of the
voltages is reversed, the spot will move toward point 2. electrical shock is reduced since the positioning
controls, which are normally connected to the de-
tive with respect to H2, the spot will move from O flection plates in electrostatic deflection circuits, are
to position 1. If the polarity of the charges is re- no longer in the high -voltage circuit.
versed, so that H2 attracts the stream, the beam is If electrostatic focusing is used, electrostatic de-
pulled in the opposite direction and the spot moves flection should also be used to minimize defects to
to position 2. the screen of the tube caused by ion bombardment.
This is basically what happens in the tube when Magnetic focusing and electrostatic deflection is
electrostatic deflection is employed. With this un- not entirely satisfactory because of the nonuniformity
derstanding, it is not difficult to visualize what oc- of focus at all points on the screen, as was mentioned
curs within the tube when the sweep voltages de- previously.
scribed in chapter 9 are applied to the deflecting Magnetic focusing reduces the effect of ion bom-
plates. bardment, and magnetic deflection provides a more
uniform focus over the area of the screen. From
this it can be seen that magnetic focusing combined
Magnetic Deflection with magnetic deflection provides a better television
As in the case of magnetic focusing previously picture than does either of the combined methods.
discussed, the electron beam is subjected to the This last is the favored method ; however, certain
design problems in particular sets will determine eliminate the detrimental action of the ion spot if
which method is incorporated in the television receiver. prolonged and efficient service is desired.
The problem involves either the removal of these
heavy negative particles from the electron beam by
ION SPOT
means of a trap method which would reject the
After a picture tube which employs electrostatic ions while permitting the electrons to pass freely,
focusing and magnetic deflecting methods has been or, in some other way, to prevent the ions in the
in operation for a time, a small round spot may beam from striking the fluorescent screen. As a
appear on the screen of the tube. This spot consti- matter of fact, both of these methods are utilized
tutes a blemish in the screen material and impairs on occasion and have been very effective in elimi-
the quality of the television picture. The cause of nating the ion spot. These two methods, both of
this spot has been traced to negatively charged which involved a change in tube structure methods,
heavy particles in the electron beam which strike will be described in detail in the following discus-
the screen with great force, causing eventual dete- sion.
rioration of the screen material. These heavy parti-
cles are negative ions and the blemish on the screen,
which appears dark brown in color, is called an ion Ion Trap Methods
spot.
One of the methods developed to eliminate ion
When the cathode of the picture tube reaches
spot from the screen of the tube is the bent gun
operating temperature, electrons and negative ions
method. The electrodes within the tube which con-
are emitted from the cathode material. Of course,
stitute the electron gun are bent so that the path
all of the negative ions may not be emitted from
the cathode material, some may be formed by other of the electrons is toward the side of the anode as
means. However, all of these negative particles are shown in Fig. 10-5. However, by means of the
formed into a concentrated beam as a result of the magnetic field from a "bending" coil placed around
the neck of the tube, the electrons are bent back
electronic lens system within the tube. The high
into line so that their path lies along the axis of the
accelerating voltage on the second anode increases
the velocity of the beam to a high value. The action tube. The heavy ions are not affected by the mag-
of the magnetic fields of the deflecting coils easily netic field and, therefore, travel straight along the
deflects the electrons in the beam, but because of original path and strike the side of the anode, or,
in some cases, a disk which serves to collect the
their relatively greater mass, the ions are not de-
ions. In this manner, the undesired negative ions
flected and continue to travel straight along the
axis of the tube and strike the center of the screen are eliminated entirely from the electron beam. Fol-
lowing this, the stream of ion -free electrons is then
at high velocities.
It can readily be seen that if this ion action were focused, accelerated, and deflected in the usual
allowed to continue, rapid deterioration of the
manner.
screen material, with attendant short life span for In chapter 1 mention was made of the practice
of coating the inner surface of the luminescent
the tube, would result. Since picture tubes are
rather expensive items, it becomes necessary to screen material in the picture tube with a thin back-
ing of aluminum film in order to increase the
brightness of the images on the face of the tube.
In addition, the application of this metallic film
provides another method of correcting the condi-
tion caused by the ion spot on the screen. It has
been discovered that a film of aluminum coated
over the fluorescent material will readily pass the
electron beam but will completely stop the heavy
ions, thus preventing them from bombarding the
screen material. Of course, a number of factors
must be taken into consideration, such as the thick-
ness of the metallic film, the gas content of the
Fm. 10-5.-In order to get rid of the negative ions in the
electron stream, the "gun" is placed at an angle so that the tube, and the operating potentials applied to the
ions are aimed at one side of the anode while a bending coil high-voltage anode. However, prolonged tests have
causes the ion -free electrons to turn and go along the axis of
the tube. proved that tubes having a metallic backing of the
proper thickness offer no restrictions to passage of the electron beam, and this is exactly what is done.
the electron beam and yet do not develop an ion For home receivers, accelerating voltages as high
spot when the applied operäting voltages are nor- as 30,000 volts are applied to the accelerating anode ;
mal. while for theater projection, voltages as high as
80,000 volts are used.
In projection -type television receivers used in
PROJECTION TUBES the home, a sufficiently bright, sharply defined pic-
ture as large as 12 by 16 inches can be obtained
All of the tubes discussed up to this point have from a picture tube having a diameter no greater
been of the direct -viewing type, that is, the televi- than 2/ inches. In theater projection of television
sion picture appears on the face of the tube and is pictures an 18 -by -24-foot picture can be obtained
viewed directly by the observers. One of the com- from a 15 -inch tube.
plaints about the presentation of television pictures It is not our purpose in this discussion of picture
in earlier receivers was the small size of the repro- tubes to become involved in a description of the
duced image, since the tubes then available were various projection methods resorted to in order
usually not larger than 7 inches in diameter. In an to obtain a picture of the desired size, brightness,
attempt to satisfy the demand for a larger picture, and sharpness of detail. However, it seems that a
the television industry turned to the development brief reference to one of the most popular systems
of receivers using picture tubes with larger di- in use at the present time will illustrate more gra-
ameters. However; this offers only a partial solu- phically how the television tube functions in such
tion, for as the size of the tube is increased, the pro- a system.
duction costs and the costs of storage and trans- Fig. 10-6 illustrates this projection system which
portation increase rapidly. These tubes are of the is an adaptation of the Schmidt principle developed
very' high -vacuum type and must be tabricated to originally for use in telescopes used by astronomers.
withstand extremely high air pressures, with the Referring to the figure, the image on the face of the
attendant hazard of injury by implosion or break- tube is directed onto the spherical mirror in the
age of the tube. Sufficiently large pictures for large
coverage cannot be attained even with use of the
20 -inch picture tube, which is the largest tube in use
in commercial television receivers today. PLANE H/í.e04
The most practical solution from all viewpoints
seems to be the projection of the television picture
upon a viewing screen of the desired size, similar FIG. 10-6.-The opti-
/fW/n'6
cal system of a projec-
to the manner in which motion pictures are pro- tion television receiver, ICQEEN
jected. In this method, the image on the face of the which employs the prin-
ciples originated by
tube is directed through an optical system consist- Schmidt for use in tele-
ing of correcting lenses and reflecting mirrors and scopes. The image of
then projected to the screen. This permits the use the picture on the DEfL EC T/N6
yOKE
screen of the picture
of a small size tube, as the original picture can be tube is reflected upward
by the spherical mirror C044fCT/MG
greatly enlarged in the optical system. However, going through the cor- L ENI
another problem arises, and that is the creation of recting lens into the in-
a sufficiently bright image on the face of the tube. clined mirror. From
here it is reflected onto
There is a loss in light intensity when light is the viewing screen. ;W./EC-7701Y
q'inESCPPF
passed through an optical system. This must be Courtesy RCA
taken into account when designing such a system
for projection television. A limit is thus quite ob- JPiE4/GAL
viously placed upon the minimum size of the pic- Mi44oe
ture tube.
We know from our discussion of the electron
beam that the intensity of light produced on the bottom of the cabinet. From there the image is
flourescent screen is directly proportional to the projected through the correcting lens which cor-
velocity of the beam at the point of impact. The rects the aberration caused when the light strikes
obvious answer to the question of screen brightness the spherical mirror. The light then strikes the in-
would seem to be an increase in the acceleration of clined mirror at the top pf the cabinet and the pic-
ture is reflected onto the viewing screen in enlarged The second anode is coated on the inside of the
size. Modifications of this system in which the tube, and a glass cup is sealed to the tube around
light is "folded" back once or twice has resulted in the second anode terminal to insure corona-free
a greater utilization of the available light with re- operation. A second coating of Aquadag on the out-
duced losses. side of the tube is grounded and used as a static
The most currently popular projection tube used shield. The two coatings form a capacitor of ap-
in commercial television receivers is the 5TP4 pro- proximately 300. µµf which provides final smooth-
jection Kinescope. This is a 5 -inch tube having ing action of the high voltage applied to the tube.
electrostatic focusing and magnetic deflection. The The average beam current is 90 microamperes and
accelerating voltage applied to the second anode is on highlight peaks it reaches the value of 500 micro-
27,000 volts (maximum) with a maximum of 6,000 amperes. The required grid drive voltage is about
volts being applied to anode number one. The re- 50 volts peak.
quired grid drive voltage is about 40 volts. This tube is employed in a projection system
The tube has an internal coating brought out to using "folded" Schmidt optical principles in a tri-
a terminal on the side of the tube which serves as angular arrangement. Because the tube itself is an
anode one. An external insulating coating, which is element in the optical system, the face plate of the
grounded, in operation forms a capacitance with tube must be optically correct. This face plate is
anode two in the order of 500 µµf. This capacitance molded of special glass which will be not discolored
serves as a filter capacitor to the high voltage applied by the low intensity, soft X radiation which is pro-
to anode two. The curved screen is coated with phos- duced by the 25,000 -volt electron -beam bombard-
phor which fluoresces white and has a medium per- ment. A phosphor screen with an aluminized back-
sistency. The tube can be mounted in any position and ing is used which increases the light output, im-
is ideally suited for employment in the Schmidt pro- proves over-all contrast, and eliminates the neces-
jection system previously described. sity for an ion trap. The tube is powered from a
A recent development in a small projection tube special 25,000 -volt pulse -type voltage -tripling power
is the MW -6 (nomenclature for experimental tube) supply which uses a special control circuit for im-
which has been developed by the North American proved voltage regulation.
Philips Company. This tube has a 2.5 -inch diameter This new projection tube with its optical system
and an over-all length of 10.5 inches. The image and power supply is incorporated in a special kit
reproduced on the face of the tube measures 1.4 by of small size and is intended for installation in
1.86 inches and from this image, a 12 -by -16 -inch pic- television receivers having the power supply require-
ture with satisfactory brightness, resolution, and ments of a 10 -inch direct -viewing picture tube such
contrast is projected onto the viewing screen. The as the 10BP4. This permits the conversion of exist-
tube utilizes magnetic focusing, magnetic deflection, ing receivers, using direct -viewing 10 -inch tubes with
and a moderate deflection angle. It features a small the proper voltage requirements, into projection sets
neck with an accurately positioned electron gun. giving larger and more satisfactory group coverage.
The accelerating voltage applied to the second anode Various sizes of projection tubes are used in
through a special type terminal brought out to the receivers designed for television theater projection,
side of the tube is 25,000 volts. By proper design but these tubes function basically in the same man-
of the electron gun a very small diameter spot ner as the picture tubes described here. The im-
(0.003 inch) is produced on the screen of the tube, portant thing to remember in the operaion of any
permitting adequate resolution of the small size picture tube is the formation of a sharply focused
image appearing on the tube face. The focusing and electron beam and its deflection across the screen of
deflecting requirements are the same as for the the tube. These functions are the same regardless
10BP4 direct -viewing picture tube operated at of the shape and size of the tube, any modifications
9,000 volts, which permits the substitution of this in the structure of the tube, and the application to
projection tube in place of the direct -viewing which the tube is finally put.
10BP4.
CHAPTER 11
POWER SUPPLIES
BY HENRY CHANES
As a general rule, two separate power supplies The low -voltage power supply used in the RCA
are used to supply the cl -c voltages for television model 621TS receiver'is shown in Fig. 11-1. The
receivers. A low -voltage supply provides approxi- transformer T106 is a conventional power trans-
mately 300 volts for the tubes associated with the former with sufficient capacity to supply the plate
amplifier, deflection, and sync circuits of the re- and heater currents required by the receiver. To
ceiver. A separate high -voltage power supply pro- prevent any undesirable coupling between the pic-
vides the high voltages required for the picture ture tube and the other circuits in the receiver, there
tube. These voltages vary from 2,000 to 10,000 is a separate heater winding for the picture tube.
volts for direct -viewing tubes and as high as 30,000 The transformer al -so has a separate 5 -volt wind-
volts for projection tubes. ing for the heater of the 5V4 -G damper tube to pre-
vent any interaction between the damping circuit
Low -Voltage Power Supplies and the power supply. The capacitors C143 and
The low -voltage power supplies used in television C144 across the primary are the usual filter capacitors
receivers are very similar to those used in radio re- to prevent any noise on the power line from coming
ceivers. However, the regulation in a television low - through the receiver. The filtering in this power
voltage power supply is generally better than that supply is quite extensive as can be noticed from the
in a radio receiver because of the importance of large number and high value of the filter capacitors
preventing variations in the numerous circuits from used. This filtering is important not only to keep the
interfering with each other and with the picture. hum clown to a very low level. but also to prevent any
The hum level is also kept down to a lower value coupling between the many different circuits supplied
because of the large number of circuits and stages, by this one power supply.
and because hum which would be inaudible in a The voltage drop across potentiometers R166 and
sound receiver may cause appreciable distortion of R152 are used to control the horizontal and vertical
the picture. centering respectively. Direct current is made to
TO HOR . TO VERT.
OUTPUT OUTPUT
-'
V4 -G SPEAKER I HOR. CENT. VERT. CENTERING
TO
FIELD CONTROL CONTROL
HEATER
+245V. R166 R152
+225V.
5101
C1285 C131A
40J1f 40).1f
C1431: R1535
.01 C127A _ 1200
+135V.
R154C
C144 -T 9000
.01
3.5V,
TO KIN.
-
-
R154B
120
R192
82K
Y2 15V.
-s
X TO R153A R189 R154A
17V. HEATERS 620 56 50
20V.
60^' T106
-110V.
TO ION TRAP
MAGNET COIL
Fic. 11-I.-The low -voltage power supply circuit used in the RCA model 621TS receiver.
173
flow through the yoke in addition to the saw -tooth that are necessary when a common supply is used
sweep current. The saw -tooth current makes the spot for all the circuits, resulting in greater efficiency
sweep back and forth, or scan, the face of the tube, and lower operating temperature. Undesirable coup-
but d.c. merely makes the spot move to one side or ling between different circuits of a receiver due to
the other and stay there as long as the current a common power supply also is reduced by the use
flows. The way the spot moves depends on the pol- of more than one supply.
arity of the current. The voltage between the arm 3. The selenium rectifier does not require heater
and the center tap of the vertical centering control current. It is small, light, and easy to install. It runs
R152 can be made either positive or negative, de- cooler than a tube and generally has a much longer
pending upon which side of the center tap the arm life.
is placed. This voltage will cause d.c. of either The operation of the selenium rectifier is easily
polarity to flow in the vertical yoke and enable the understood. Its rectifying action depends on a prop-
picture to be moved up or down erty often found at a junction of two dissimilar
In the horizontal centering control circuit, the metals, i.e., that electrons flow more readily in one
situation is somewhat different. The damping tube direction than in the opposite direction.
in the horizontal sweep circuit will, because of its A typical unit found in present-day television
rectifier action, produce d.c. The action of this tube
is discussed in more detail in chapter 9 which covers FIG. 11-2, right.-The elec- METALLIC
COATING
sweep circuits. Because of this d.c. produced by the tron flow through a selenium ALUMINUM
PLATE
rectifier is from the metallic
damping tube, it is necessary to produce d.c. of op- coating -selenium surface to- SELENIUM
SURFACE
posite polarity in order to center the picture. This - ward the aluminum plate, as
indicated by the arrow.
is done by using the voltage drop between the arm ELECTRON FLOW
and one side of the horizontal centering control -*-SELENIUM RECTIFIER SYMBOL
R166, since d.c. of a constant polarity is required FIG. 11-3, left.-In the sche-
by this circuit. matic symbol for the selenium
rectifier on the right, the elec-
The string of resistors following the centering tron flow is from 2 (identified
controls make up a bleeder to supply the different by a red dot) upward to 1
(identified by a yellow or no
positive and negative voltages required by the dot).
different circuits in the receiver. The voltages shown
at the different points in the bleeder circuit are de- receivers consists of several small plates, usually
termined by the bleeder current and the currents square in form, stacked together at their centers,
drawn by the load on each output of the power and utilizing two lugs for connection to the circuit.
supply. The power supply is grounded only at the Each element is simply a supporting aluminum
junction of the resistors R154C and R 154B, so that plate, coated on one side with selenium, and then
negative as well as positive voltages with respect to having a metallic coating over the selenium. This
ground are available. selenium "sandwich" ensures adequate electric con-
The low -voltage power supply in the Consolida- tact with both sides of the selenium. By proper
ted Television model 2315 consists of a single heat- heat processing, a rectifying film is formed between
er supply plus' three separate d -c supplies that use the selenium and the metallic coating. The metallic
selenium rectifiers instead of vacuum tubes. The use coating has a great number of free electrons that
of selenium rectifiers in television receivers has can flow through the selenium, and into the alumi-
several advantages over the conventional power sup- num plate. However, the selenium has very few
plies. free electrons available, so that the electron flow
1. By the use of voltage -multiplying circuits in the opposite direction is very limited.
where higher voltages than the power lines are nec- If the negative terminal of a power source is
essary, the power transformer can be eliminated. connected to the metallic coating, the flow of cur-
This appreciably cuts down the weight and size of rent toward it is very small, for we have seen that
the receiver. Also, the danger of stray fields from there are few free electrons available. Consequently,
the transformer affecting the video circuits and pro- the electron flow in Fig. 11-2 is from right to left.
ducing interference in the picture is removed. The symbol used to represent the selenium recti-
2. Separate supplies for different circuits in the fier and similar types, shown in Fig. 11-2, was
receiver are practical with the use of selenium rec- adopted before the electron -flow theory was accepted,
tifiers. This eliminates the large bleeder resistors but has never been revised. Therefore, when a sele-
POWER SUPPLIES 175
nium rectifier replaces a diode rectifier in a circuit When the side of the line that is connected to
diagram, the corresponding terminals are as indicated R66 goes positive, the rectifier RX-1 will conduct
in Fig. 11-3. and charge C61 to the peak value of the line voltage.
A typical type of selenium rectifier is the Fed- Let us call this voltage E,,,. The polarity of the
eral Type 403D2625, which has a maximum d-c charge on C61 is shown in Fig. 11-4. On the next
output of 100 ma and a drop of 5 volts across it. half cycle RX-1 will not conduct, but capacitor C61
The two lugs may be marked positive and negative, will act in series with the line and, with the recti-
or they may be color coded ; positive is indicated fier RX-2 now conducting, will charge the capaci-
by a red dot, negative by a yellow dot or blank, as
tor C60 to 2Em with the polarity shown in Fig. 11-4.
ln Fig. 11-3. Care must be taken that the proper
polarity connections are made, remembering that On the next positive half cycle, C60 will act in
B+ is obtained from the cathode of the rectifier series with the line and a total voltage of 3Em will
tube, and so the positive lug (red dot) must be be applied to the rectifier RX-3 which is now con-
where the cathode was formerly connected. ducting. Capacitor C62 will be charged up to 3Em.
The low -voltage power supply of the Consolida- This is one of the output voltages. The other
ted Television model 2315 is shown in Fig. 11-4. is taken from the other side of the filter which con-
The purpose of the heater transformer is obvious, sists of C62, R82, and C63, and due to the voltage
so let us look at the first of the selenium rectifier drop in R82, is equal to 380 volts.
supplies. This one has an output of 440 volts. To The second selenium rectifier supply delivers 105
get this high voltage without the use of a step-up volts. This section contains the limiting resistor
transformer, a voltage -tripling circuit is used. This R67, the selenium rectifier RX-4, and the filter con-
circuit will theoretically have an output of 3 times sisting of CM, R82A, and C65. The rectifiers used
the peak value of the input voltage. If we assume in this circuit are the same type that are used in the
the line voltage to be 115 volts rtes, the peak value voltage tripler supply. They are Federal Type 404-
is 115 x 1.4 or approximately 160 volts. The output D2795 and have a d -c capacity of 200 milliamperes.
of the voltage tripler then should be 480 volts. Act- The third section of the power supply is a bias
ually this circuit delivers 440 volts. The loss in supply. It has an output of -105 volts. Notice that
voltage from the ideal value is due to the fact that the selenium rectifier RX-5 is connected with the
current is being drawn by the load on the supply terminals reversed in order to obtain an output
and the drop in the rectifiers themselves. There is which is negative with respect to ground. The filter
also some voltage drop in the current -limiting re- capacitors have their positive terminals grounded
sistors R65 and R66. for the same reason. The selenium rectifier RX-5
6.3V. TO ALL HEATERS

EXCEPT H.V. RECTIFIERS
117v.
60
+440 VOLTS
4 TO AUDIO OUTPI)T
AND C.R.T. FOCUS
C60 R82
R65
+ _ -_.+ + 380 VOLTS TO
HOR. OUTPUT
RX-1 RX-2 RX-3 +
C63
EM c61 3EMT C62
T R82A +105 VOLTS TO
Rb7 +
R F, I -F, OSC,
RX-4 C64 C65 SYNC. CIRCUITS
R68 R69
T
-105 VOLTS TO
HOR. OUTPUT
RX-5 C67 Cbb
T
+ +
T
Fic. 11-4.-The low -voltage power supply of the Consolidated Television model 2315. Note the use of the selenium rectifiers.
used in this supply is a Federal Type 403D2787, cycles from the power line for the input of the step-
which is rated at 150 millamperes d-c output. up transformer, a separate r -f oscillator is used to
supply the input of the step-up transformer. The
High -Voltage Power Supplies frequency of this oscillator is usually in the range
of 50 to 500 kc. The transformer can be made much
There are three types of high -voltage power sup-
lighter and smaller than a 60 -cycle power trans-
plies that are used in television receivers. They are :
former because of the high frequency used. It is
1. The 60-cycle high -voltage supply
also much easier to filter this higher frequency,
2. The r -f high-voltage supply therefore, the filter capacitors can be made smaller.
3. The kick -back high -voltage supply The oscillator is capable of delivering little more
The basic circuit of a 60 -cycle power supply is power than is actually required by the anode of the
shown in Fig. 11-5. This circuit is very similar to
the conventional circuits used in the low -voltage R -F STEP-UP RECTIFIER FILTER HIGH
supply. The transformer T1 steps up the 60 -cycle OSCILLATOR TRANSFORMER VOLTAGE
line voltage to the high voltage required, usually in

the order of several thousand volts. This voltage is FIG. 11-6.-Block diagram of the r-f high -voltage supply,
the frequency of the oscillator being in the range of 50 to
rectified by the half -wave rectifier tube and then 500 kc.
filtered by the RC filter. The current required by
the anode is usually about 200 microamperes, there- picture tube this is usually about two watts. The
;
fore the voltage drop across the resistor is not regulation of the power supply is, therefore, very
too great. The resistor R should be quite large, at poor; that is, the output voltage falls off rapidly
least a megohm, and the capacitors as small as will with an increase in load. This is a very desirable
give satisfactory filtering. Large values of filter re- feature in a high -voltage power supply. If a serv-
sistance rather than large values of capacitance are iceman accidentally came in contact with a lead car-
used to obtain the required filtering. The advantage rying the high voltage, he would load clown the
of this is the reduced cost and size of the capaci- power supply and the voltage would drop, lessen-
tors ; in addition, the smaller the size of the filter ing the chances of a fatal shock. However, this does
capacitor, the less the danger of fatal shock if one not mean that it is safe to touch a high -voltage lead.
accidently comes in contact with a charged or a part- Do not work on any high -voltage circuits unless the
ly discharged capacitor. Nevertheless, the high -volt- power has been turned off and the high -voltage fil-
age filter capacitors should always be discharged be- ter capacitors have been discharged.
fore any measurements are made so as to guard The r -f power supply used in the Belmont models
against the possibility of an open bleeder circuit. A- 22A21 and 22AX21 is shown in Fig. 11-7. The
bove 400 volts, the transformer necessary for this oscillator circuit uses a 6V6 as a tuned plate, tickler
type of power supply becomes heavy, bulky, and ex- feedback oscillator. The frequency of oscillation is
pensive. Also, the danger of fatal shock becomes determined by the setting of the trimmer capacitor
greater. Because of these objections, this type of C86. The secondary winding together with a wind-
power supply is generally not used except in the ing for the filament of the rectifier tube are both
smaller television receivers, usually those employ- wound on the same form as the oscillator primary
ing a 7 -inch picture tube or smaller. and tickler coil, thus combining the oscillator coil
Most of the disadvantages of the 60 -cycle supply and the step-up transformer. The secondary winding
are eliminated by the use of the r -f high -voltage has many more turns than the primary winding,
power supply. A block diagram of this type of sup- therefore the voltage induced in the secondary will
ply is shown in Fig. 11-6. Instead of using the 60 be many times greater than the voltage applied
across the primary by the oscillator. The frequency
of the oscillator is determined by the setting of the
trimmer capacitor C86. The resonant frequency of
the secondary winding is determined by the shunt-
ing capacitance due to the rectifier tube. The volt-
age across the secondary will be greatest when the
frequency of the oscillator is equal to the resonant
frequency of the secondary winding in parallel with
FIG. 11-5.-The basic 60 -cycle power supply circuit. Here the shunting capacitance. The circuit is designed so
large values of filter resistance rather than large values of
capacitance are used to obtain the necessary filtering. that the maximum voltage that is obtainable at res-
POWER SUPPLIES 177
onance will be greater than that required by the tele- ing a high voltage which is rectified by the 8016, as
vision picture tube. In actual operation, the trim- shown in Fig. 11-8. The 6BG6G is a beam tetrode,
mer is adjusted until the output voltage falls off to capable of handling heavy current ; its main func-
the required value. The LC combination of L6 and tion is to supply current with the proper waveform
C84 makes up a decoupling filter to keep r.f._out of to the horizontal sweep coils so that a horizontal
the low -voltage power supply.
The method of .heating the high -voltage rectifier H.V. 8000V.
RECTIFIER TO
tube is interesting. A small winding on the same (8016) C. R.T.
form as the other windings has sufficient r -f volt- SHAPED
MULTIVIBRATOR
age induced in it to heat the filament of the recti- OUTPUT
fier tube. This has a decided advantage over heating
the filament with 60-cycle current. If 60 cycles HORIZONTAL
OUTPUT
HORIZONTAL
SWEEP
were used, the high -voltage filter would have to 6BGGG COILS
filter out the 60 cycles due to the filament current.

8+
This would mean very large capacitors. By heating NOR. SYNC.
CONTROL
the filament with r.f., the filter has to take care of SYSTEM
only the high -frequency r -f voltage. This is very Fic. 11-8.-The "kick -back" type of power supply such as is
easily done, as can be seen from the small capaci- used in the G.E. model 802. Here the rectifier plate voltage is
obtained from the horizontal deflection system during the fly-
tors used in the filter circuit. In this circuit, C89 back periods.
and C90 are only 0.001 µ f each.
The third type of high -voltage power supplies trace of the proper length is applied to the viewing
used in television receivers is the kick -back type tube. The 6AS7G is a dual triode, having an ampli-
of supply. The distinguishing feature of this power fication factor of only 2.1 and a plate resistance of
supply is that the rectifier plate voltage is obtained 140 ohms when the two sections are cónnected in
from the horizontal deflection system during the re- parallel. This tube damps the high -frequency oscil-
trace or flyback of the sweep, rather than from the lation set up during the flyback period of the elec-
a -c line as in the case of the 60 -cycle. supply or tron beam.
from a separate high -frequency oscillator as in the The usual high -voltage rectifier, such as an 878,
case of the r -f power supply. A block diagram of requires considerable heater power and is not de-
this type of supply is shown in Fig. 11-8. The high - signed for r -f operation. The 8016 diode was de-
voltage supply in the G.E. model 802 is typical of veloped for this purpose and requires only a quar-
the kick -back high -voltage supplies used in televi- ter of a watt for the heater. Fig. 11-8 shows that
sion receivers. the heater voltage for the 8016 is obtained by a
To analyze the operation of this circuit, we must very small secondary winding on the transformer,
see how the horizontal output tube, a 6BG6G, and T25. The 10BP4 picture tube is scanned by a mag-
the damping tube, a 6AS7G, generate a pulse that netic coil system, but only the horizontal sweep coils
excites the step-up transformer, 725, thus provid- enter our study of the high -voltage power supply.
+300
FIG. 11-7.-The r -f power supply

used in the Belmont models 22A21
and 22AX21, in which the 6V6-
I BOGT/ 8016 GT/G is used in the tuned -plate,
H.V. RECT.
tickler feedback oscillator circuit,
the frequency of which is controlled
R80 by C86.
10 K
R82
C83 I 00 K
02Nf C87
+5000 VOLTS
330fyuf C90
TICK.
6V6-GT/G 001/uf
R. F. OS C.
For proper scanning, the current in the sweep coils a multivibrator (6SN7-GT) which is a two -tube
must have a saw -tooth form, such as shown in Fig. oscillator that normally produces a rectangular out-
11-9, the forward trace across the screen occurring put. However, the network composed of the capaci-
during period a -b and the retrace or flyback during tors and resistors in the input circuit of the 6BG6G
is so designed that we obtain a saw -tooth waveform
b with a negative pulse, such as shown in Fig. 11-10.
11-9.-The current in the This is the input that is applied to the grid of the
sweep coils must have a saw -
tooth form, a -b being the trace 6BG6G. The collapsing field in the horizontal de-
period and b -c the retrace. flection coils produces a positive voltage pulse on
aL C the primary of T25 and consequently on the 6BG6G
plate. The negative pulse on the 6BG6G grid, a -b
the much shorter period b -c. The forward trace in Fig. 11-10, ensures that the 6BG6G is cut off
must be as linear as possible so that distortion is during the flyback period, in spite of the high plate
prevented. voltage.
The sweep trace takes place when the output of During the trace period (c -a in Fig. 11-10) when
the 6BG6G tube is applied to the horizontal sweep the 6BG6G is conducting, the saw-tooth grid volt-
coils, and the transformer T25 is so designed that
aL al.
a proper impedance match exists between the tube
and the coils. When the 6BG6G stops conducting,
FIG. 11-10.-The saw -tooth
the components to the right of T25 (See Fig. 11-11) waveform with a negative
are excited into violent oscillation, and the oscilla- pulse, that is applied to the grid
tion is used to obtain the rapid flyback, b -c in Fig. of the 6BG6G tube.
11-9. Any oscillation beyond the first half cycle is
undesirable, as it will affect the linearity of trace b
a -b in Fig. 11-9, and the 6AS7G is used to damp age produces a saw -tooth plate current. The effec-
it out. Very little of the magnetic energy is con- tive plate load consists of the horizontal deflection
sumed by the flyback, and the collapsing field pro- coils, T25 being used only as an impedance trans-
duces a positive voltage pulse on the primary of former to match the coils to the tube. Since the
T25. This pulse is stepped up by the additional plate load is inductive, if a sudden change in cur-
winding shown on the primary ; it is then rectified rent takes place, we obtain a high -voltage pulse.
by the 8016 and delivered to the viewing tube. This sudden change takes place during a -b of Fig.
The detailed functioning of each component may 11-10. During c -a of Fig. 11-10 the energy supplied
now be examined, keeping the previous discussion to the yoke builds up, and continues to do so until
in mind. The input to the 6BG6G is obtained from the negative pulse occurs. At this instant, when the
HORIZ. 8016
MU IT IV IB R.AT OR H.V. RECTIFIER
OUTPUT C97 HORIZ. OUTPUT
. 68666 R 113
8,000 V.
TO 10BP4
C95
R121
6A57G
CISSr R85 DAMPING
CSOC
TO HORIZ. L23
SYNC.CONTROL WIDTH TO CONTRAST HORIZ-LINEARITY
SYSTEM CONTROL CONTROL CONTROL
Fic. 11-11.-Schematic diagram of the high -voltage supply circuit of the G.E. model 802. The block diagram of this circuit
is shown in Fig. 11-8.
POWER SUPPLIES 179
6BG6G is cut off, the components between T25 and the second anode and about 5000 volts on the first
the 6AS7G are shocked into violent oscillation anode. It would be very difficult to obtain these very
(L23, horizontal deflection coils, part of T25, and high voltages by the use of only one rectifier. As-
distributed capacitances). During the first half cycle, suming that a kick -back type of supply is used, the
which is negative, the current in the hórizontal de- high -voltage rectifier tube, the horizontal output
flection coils reaches a maximum in the direction transformer, and all associated wiring would have
opposite to that in which it is flowing and the flyback to withstand this very high voltage. This, of course,
consequently takes place. At the end of the first would increase the cost of the power supply tre-
half cycle, the voltage starts to go positive. The mendously. For these reasons the high -voltage sup-
6AS7G now conducts, since its plate is positive, plies for the projection type picture tubes generally
and the oscillation is rapidly damped. are of the voltage, multiplying type. The high -volt-
Transformer T25 has four functions. Its main age supply of the GE model 901 consists of a kick-
function is to transform the inductance of the hori- back type of power supply plus a voltage multiplier.
zontal deflection coils to a value that meets the Except for the voltage multiplier, this supply is
operating conditions of the 6BG6G. A small sec- very similar to the kick -back supply of the GE
ondary tap supplies heater voltage for the 8016. It model 802 (see Fig. 11-11) which we have already
also takes the inductive "kick" voltage from the analyzed. Therefore it will be sufficient for us to
collapsing magnetic field flyback and places it on discuss the theory and analyze the operation of the
the primary. Finally, this voltage pulse on the pri- voltage multiplier circuit, a simplified schematic of
mary is raised by autotransformer action to 8000 which is shown in Fig. 11-13.
volts and is applied to the plate of the 8016. (In First, let us look at the basic circuit shown in
an autotransformer the input is applied to a portion Fig. 11-12. In this circuit, we have a source of d -c
of the winding, terminals 1-2 .of Fig. 11-11, and the voltage as represented by the battery and a switch
output is taken across the entire winding, terminals which will put the capacitor C, either across the
1-3. It has these advantages over the ordinary two - battery for charging or across R and C2 for dis-
circuit transformer: better voltage regulation, great- charging. At the beginning of the experiment we
er efficiency, and smaller size ; it does have disad- will assume that neither capacitor has any charge
vantages, such as lack of d-c separation, which pre- on it. Now, we will place the switch in position 1
vent its more universal use.) for a very short time. Since capacitor C, does not
The 8016 rectifies this high -voltage pulse. Due to have any resistance in series with it, it will charge
the high frequency at which this takes place, a 500
µµf capacitor, C118, is sufficient for filtering; and
Fic. 11-12.-Equivalent cir-
as its small value means small energy storage, there cuit showing charging action
is consequent reduction of danger when servicing occurring in the voltage mul-
tiplier shown in Fig. 11-13. CZT
the high -voltage supply. The 8000 -volt output is ap-
plied to the picture tube through the filter section
C118 and R113. up almost immediately to the full voltage of the
Some of the signal on the secondary of the trans- battery, which is E. Then the switch is turned to
former T25 is used to operate the horizontal sync position 2 and held there for a much longer time
control system. This system is also known as auto- than in position 1, since the resistor in series with
matic frequency control (afc) and is discussed in the capacitor C2 prevents it from charging up as
detail in chapter 8. The negative bias developed in fast as it would without the resistor. While the
the grid circuit of the 6AS7G damping tube is used switch is held in position 2, some of the charge
to supply a steady negative d -c voltage for the con- from C, will go to C2. If the two capacitances are
trast control circuit. The contrast control varies the equal, the voltage across each will become half of
gain of the i -f amplifiers by varying the grid bias E. Now, suppose the switch is again placed in posi-
on the first two stages and thereby controls the con- tion 1 for a very short time. Capacitor C, will again
trast of the picture. This is discussed in chapter 6. charge up to E. When the switch is placed in posi-
The GE models 901 and 910 are projection type tion 2, C, will discharge into Cs until the voltage
television receivers. In order to obtain an image on across each capacitor is equal to 34 of the voltage
the picture tube bright enough so that it can be E. If this is kept up for a number of times, the
projected onto a screen, very high voltages on the voltage across Cs will become greater and greater
second anode are necessary. The 5TP4 projection until eventually both capacitors are charged up. to
tube used in this receiver requires 27,000 volts on the voltage E. Actually the voltage across C. never
180 RIDER'S-HOW IT WORKS"
becomes exactly equal to E, but it is very close to it, ground at the horizontal sweep frequency due to
so that for all practical purposes we may say it be- the large filter capacitors in the low -voltage supply.
comes equal to E. Since this pulse is positive, the rectifier V207 will
If we replace the components of the circuit of Fig. conduct and charge C214 to the peak voltage E,.
11-12 with the corresponding components of the During the forward trace, C214 will discharge
simplified schematic shown in Fig. 11-13, we will through R219-R222 to charge C211 to E.
The dis-
have the first stage of the voltage multiplier used charge path of C214 is through R219 -R222, C211,
in this high -voltage supply. The transformer T202
R215, the primary of the transformer, L204, B+
and the first high -voltage rectifier, V207 are the
and then to ground. When we say that C211 will
equivalent of the d -c supply in the basic circuit of
Fig. 11-12. The capacitor C214 corresponds to C1, charge up to E., we mean after a number of pulses
and the RC combination of R219 -R222 and C211 in much the same manner as was done in the basic
corresponds to R and Cs in Fig. 11-12. The switch circuit in Fig. 11-12.
action in the basic circuit is performed by the pulse On the first positive pulse after the capacitors
that is developed in the primary of the transformer are fully charged, capacitor C211 will act in series
during flyback. The capacitance C8 represents the with the pulse and a total voltage of 2E,,, will be ap-
capacity between the second anode of the picture plied to the second rectifier tube, V208. This tube will
tube and the external conductive coating which is conduct and will charge the series combination of
grounded. This capacity is usually around 250 µµf C215 and C214 to a voltage of 2E,. Since the capa-
and at the sweep frequency is large enough to be citances are equal, each one will be charged to a
used as a filter capacitor. voltage of E,. The polarity of these voltages is
When the flyback occurs, a positive pulse is pro- indicated in Fig. 11-13. During the forward trace,
duced in the primary of the transformer T202. the capacitors C214 and C215 will discharge through
Since the flyback period is a very small part of the the resistors R223 -R226 to charge the capacitors
entire sweep cycle, the pulse will be a comparatively C212 ,and C211. The rest of the discharge path is
short one, but due to the step-up in the transformer the same as in the first stage. C212 and C211 will
it will be about 8000 volts at the output of the trans- now charge up to a total voltage of 2E,, but since
former. Let us call the peak voltage of this pulse they are equal, the voltage across each will be E.
E,,,. This voltage can be applied between the plate On the next positive pulse, C211 and C212 will
of the first rectifier tube, V207, and ground since ict in series with the positive pulse and apply a
the other end of the primary of the transformer total voltage of 3E,,, to the plate of the third recti-
T202 is connected to B'+, which is an effective fier tube V209. This tube will conduct and charge
I B3GT IB3GT IB3GT IB3GT

H.V. R E CT. HV.RECT. H.V RECT. H.V RECT.
V 207 V208 V 209 V 210
EM EM
-11+ IIM+
C2I1 C212 C213
C219 R215
T202
r IIC215
-
+27KV.
+EM *TO
IB3GT 5TP4.
HEATER EM
SUPPLIES - C214 R245,6
R217 R23I - R233
TO FOCUS POT.
OUTPUT
TO DAMPING
CIRCUIT 11-13.-Simplified schematic dia-
AND YOKE gram of high -voltage supply using volt-
6BG6-G age multipliers as used in G.E. models
HOR. 901 and 910.
OUTPUT B+ TO AFC
POWER SUPPLIES 181
the capacitor C216 in a similar manner to the quite 4E, due to the current drawn by the picture
second stage. During the forward trace the three tube and the voltage drops in the rectifier tubes. The
capacitors C214, C215, and C216 will discharge output voltage is usually about 3.7E, instead of the
through R227-R230 to charge C213 to E,. On the theoretical 4E,. The first anode of the 5TP4 re-
the next positive pulse C211, C212, and C213 will quires a voltage of only about 5000 volts as com-
all act in series with the pulse so that a total volt- pared with the 27,000 volts required by the second
age of 4E, is applied to the plate of the fourth anode. The voltage at the cathode of the first recti-
rectifier tube V210. This tube will conduct and fier tube V207, which is about 7000 volts, is tapped
produce an output voltage of nearly 4E,. across the off through the resistors R231 -R233 and is applied
bleeder resistors R245,6 and the shunting capa- to the focus control potentiometer which supplies
citance Cs. The output voltage will not be equal to the voltage for the first anode.
CHAPTER 12
ALIGNMENT AND SERVICING

BY WILLIAM BOUIE and HENRY CHANES
When defects occur in a television receiver caused video sections of the receiver. An examination of the
by misalignment or breakdown of parts, the defect schematic and. parts -layout diagrams for the particu-
must be isolated and corrections made to restore nor- lar receiver is required to locate the various trimmers
mal operation of the set in the least amount of time. and slugs to be adjusted.
This chapter is written with that end in view. A simple block diagram of a conventional television
With experience, the average radio serviceman receiver is shown in Fig. 12-1. This diagram shows
usually works out his own pet troubleshooting pro- the television receiver to consist of a number of indi-
cedure with which he can quickly find the defects in vidual sections performing specific functions. For the
inoperative radio receivers. This also applies to television receiver to operate properly and with the
troubleshooting television receivers, and it is hoped greatest efficiency, the tuned circuits in each section
that the information contained here will be of con- must be adjusted to certain frequencies. Adjustment
siderable help in establishing a fast working method of these tuned circuits in a definite sequence consti-
consistent with the circuit specifications of the re- tutes what is known as alignment.
ceiver being serviced. As regards alignment, each From the block diagram, it can be seen that the
television manufacturer provides alignment data for function of the r -f section of the receiver is to select
his particular receiver and it is not intended that the the desired station and to amplify the incoming signal
general information contained herein should be sub- before passing it on to the mixer stage. In order to
stituted for these specific alignment data. However, if perform both of these functions, it is essential that the
for any reason the manufacturer's alignment proce- tuned circuits present in the r -f unit be tuned ac-
dure is not available, this general procedure outlined curately to the frequency of the incoming signal. The
here will enable the serviceman to align the set. oscillator coil for each channel must be adjusted to
The troubleshooting section covers localizing meth- produce a frequency which, upon being beat with the
ods used to isolate the defect to a particular section frequencies of the picture and sound carriers of the
and stage. For television receivers incorporating con- selected television channel, will produce the i.f.'s for
ventional circuits, best results will be obtained by fol- the sound and video sections of the receiver.
lowing the localizing procedure presented here. How-
Correspondingly, the sound section and the video
ever, occasions may arise where an alternative method
section of the set require adjustment in order to pass
may prove more satisfactory. A troubleshooting chart
their respective i.f.'s through to the detector stages.
is included at the end of the chapter to facilitate fur-
Additional tuned circuits are placed in the video sec-
ther investigation into the. causes of probable break-
tion to keep interfering frequencies from appearing in
downs.
the picture with resulting distortion effects. These
ALIGNMENT latter circuits are known as sound or video traps, de-
A television receiver is able to select one particular pending upon the particular interfering frequencies
station at a time to the exclusion of all other stations to be suppressed.
operating on different channels in a manner similar It must be determined whether the faulty operation
to the conventional superheterodyne radio receiver. of a television receiver is due to poor alignment or to
Accurate tuning to a particular channel and the re- some defect in the set before any alignment is at-
ception of high -quality sound and picture detail is tempted. In entirely too many instances there is a
dependent upon optimum adjustment, or alignment, tendency to blame poor receiver operation upon the
of the various tuned circuits within the television set. alignment and to upset a perfectly good alignment
These tuned circuits, which consist of acceptance and without first having investigated for defects in the set.
rejectance circuits (the latter known as traps), are This tendency should be discouraged and an exami-
incorporated into the r -f, mixer -oscillator, sound, and nation of the set should be made to determine the true
182
ALIGNMENT AND SERVICING 1.83
cause of the defective operation before any alignment tuned circuits of the receiver. Due to vibration, the
is attempted. movement of parts, and the effects of humidity, tem-
perature, and age, the capacitance and inductance
An incorrect alignment condition in a television set
associated with these tuned circuits change their
is generally accompanied by low sensitivity, loss of
values, and the tuned circuits go out of alignment.
sound or picture or both, and distortion in sound
Aside from changes in the tuned circuit itself, there
and/or picture. Misalignment of the trap circuits is
are a number of other factors which make realign-
indicated by interference in. the picture caused by
ment necessary. Among these can be mentioned the
sound frequencies of the channel being received, or
movement of r -f and i -f wiring, since the movement
by picture and/or sound frequencies of an adjacent
of these leads changes the relative capacitances and
channel. These conditions may occur on one or more
inductances associated with the tuned circuits. A
channels and may be present in varying degrees de-
slight change in position of the wiring may cause a
pending upon which tuned circuits are out of align-
channel, or even the entire receiver, to be inoperative
ment and the extent of the misalignment.
since the leads constitute a large part of the induct-
Considerable time .may be saved by analyzing ob- ance and capacitance of the tuned circuits.
served indications to determine which sections of the Particular mention in this connection must be made
receiver are out of alignment. For example, if mis- of the importance of using exact replacement parts
alignment is indicated on only one channel, it would where replacements of resistors, capacitors, and other
be safe to assume that only the local oscillator or the parts becomes necessary in the r -f amplifier, mixer,
r -f circuit of this channel requires alignment. Proper oscillator, or i -f sections of the television receiver. The
response on the other channels shows that the sound use of a part which has the same rating as the part
and video sections are properly aligned and do not being replaced but which has different physical char-
require realignment. Conversely, if the receiver acteristics or size will sometimes throw the receiver
shows low sensitivity to sound and/or picture on all out of alignment, and in other cases even cause insta-
channels, misalignment of the sound or video i -f sec- bility and oscillation. It also should be kept in mind
tions is indicated because misalignment of this part that many replacements parts are rated at low fre-
of the receiver will drop the sensitivity on all chan- quencies where stray capacitances and inductances are
nels uniformly. While it is possible for the alignment not significant. If an exact replacement part is not
on all the channels to be out, the more probable con- used, these characteristics must be taken into consid-
dition is that the i -f sections require realignment, and eration. In addition, replacement of tubes in the front
these sections should be checked first. end and i -f sections of some receivers may cause suffi-
There are a number of different factors which make cient change in the tuned circuit to require realign-
realignment necessary at more or less frequent inter- ment. Occasionally when replacing an oscillator tube,
vals. Perhaps the factor which is responsible for more it may be necessary to try several tubes before one is
realignment jobs than any other is the change in the found that will operate properly in the oscillatory
characteristics of the components associated with the circuit.
SOUND SECTION
2" Ist
ANT.
SOUND
F
DISC. --- AUDIO
AM P L
AUDIO
OUTPUT
SPEAKER
R F
MIXER
AMPL VIDEO SECTION
IST 2". .0
VIDEO VIDEO VIDEO
VIDEO VIDEO VIDEO
FIG. 12-1.-Block diagram of a tele- I F F I F.
DET AMPI OUTPUT
vision receiver, which can be divided osc.
into five sections: the radio frequency
or front end, the sound, video, sweep
and sync sections, and the power sup- FRONT END SWEEP SECTION
plies. In order for the set to function VERT VERT DC
properly, the tuned circuits in these sec-
tions must be aligned to certain fre- SYNC.
OSC OUTPUT
TO DEFLECTION
YOKE OF
RESTORER
PICTURE
SEP.
quencies. TO ALL PICTURE TUBE TUBE
STAGES .HOR. HOR
OSC. OUTPUT
LOW R P HIGH
120v
60r. POWER
VOLE- AS- E
SUPPLY
- VOLTAGE
POWE R
SUPPLY
POWER SUPPLIES
I 8 RIDER'S-"HOW IT WORKS"
Equipment Required oscilloscope. The double trace on the oscilloscope,
The basic test equipment required for the alignment resulting from phase shift, can thus be resolved into
of a television receiver should consist of the following: a single response pattern.
1. Sweep Generator. The marker generator is an r -f signal generator

used to supply marker pips on the response pattern
2. Marker Generator.
being observed, in order to indicate the exact fre-
3. Oscilloscope.
quency of any point on the pattern. This generator
4. Vacuum-tube Voltmeter (VTVM) also may be used to peak the i -f stages and align traps.
The sweep generator is an r -f signal generator It should contain the following features :
whose output frequency can be made to vary con-
a. The marker generator should cover the same
tinuously over a range of frequencies. This variation
frequency ranges as the sweep generator, but in this
in frequency over a specific band is automatic and is
case the accuracy of frequency calibration is of ex-
called "sweeping." This action, in effect, constitutes
treme importance. Frequency stability and calibration
f.m., and in some manufacturers' literature the sweep
in the marker generator must be exceptionally good
generator is referred to as an f -m signal generator.
for proper alignment of television receivers.
This sweep generator in conjunction with an oscillo-
scope facilitates alignment by providing a visual indi- b. An adjustable attenuator control should be in-
cation of the response characteristic of the circuit corporated in the output circuit enabling the output
being aligned. to be varied from 0 to 0.1 volt. In some generators
the attenuator control is calibrated in microvolts ; in
The sweep generator should have the following
others the level of the output is indicated by a meter.
characteristics :
c. In addition to supplying an unmodulated signal,

a. The frequency range should cover all the tele-
vision channels from about 40 megacycles to 220 the marker generator must have provision for an a -m
megacycles. It must also cover the i.f.'s of the video r -f output signal. The modulating frequency is not
and sound sections, which range from approximately critical but most manufacturers provide a 400 -cycle
20 to 30 megacycles for later models of receivers. In signal. A calibrated indication of the percentage of
early models, i.f.'s as low as 8 mc may be encountered. modulation up to approximately 80 percent is a
Accuracy of frequency calibration is not too impor- desirable feature.
tant in this generator as the exact frequency will be The oscilloscope used in the alignment of television
determined by the marker generator. receivers can be any of the conventional types used
b. The sweep width should be variable, with a
for test purposes. However, oscilloscopes with 5 -inch
maximum width of at least 10 mc. The output of the tubes may facilitate the alignment by virtue of the
sweep generator should be constant over the entire larger pattern obtained.
range of frequencies being swept. If this is not so, it The VTVM used in the alignment should have a
will be impossible to obtain a true response pattern. low d-c scale of the order of three volts and an input
c. .The sweep generator should have an adjustable impedance of about ten megohms. A zero -center scale
attenuator control in the output circuit with a maxi- will be found useful in balancing the discriminator
mum available output of at least 0.1 volt. Provision output and in adjusting the local oscillator.
should be made for either a balanced or unbalanced
output. A balanced output is desirable when coupling Alignment Procedures
the generator to the antenna terminals of a receiver In the actual alignment of a television receiver, the
having a balanced input circuit. An unbalanced out- following pertinent facts should be borne in mind :
put is generally used when the generator is connected Quite often response curves are obtained that appear
to other points in the television receiver. to be correct but that do not indicate the true re-
d. Provision must be made in the sweep generator sponse of the circuit. If a marked change in the shape
to supply a horizontal sweep voltage or sync signal of the response curve is obtained with changes in
for the oscilloscope. If a horizontal sweep voltage is output level or sweep width of the sweep generator,
provided, this voltage must be of the same frequency the curve is, in all probability, not a true indication
and waveform as the voltage used to modulate the of circuit conditions. This false response may be
carrier frequency of the sweep generator. Usually a caused by oscillation or regeneration in the circuit, or
phase adjustment control is included in this circuit to by overloading of one or more of the i -f amplifiers.
shift the phase of the sweep voltage supplied to the The solution for overloading is self-evident ; always
ALIGNMENT AND SERVICING 185
use the smallest signal possible to provide a suitable should be kept in mind when comparing the indicated
indication on the output device which should be used pattern on the oscilloscope with reference patterns.
at maximum sensitivity, whether it be an oscilloscope It is important that all of the test equipment and
or VTVM. the receiver being aligned be bonded together and
Regeneration and oscillation can be caused by too connected to a common ground.
close coupling of the generator leads since pickup or For over-all alignment of a television receiver, a
radiation from these leads can cause undesirable in- definite sequence of alignment must be followed for
teraction between the stage to which the generator is several reasons. First, the response of some circuits
coupled and the other stages in the circuit. Regenera- depends upon the proper adjustment of others. Sec-
tion or oscillation can be prevented by the use of well - ond, when an orderly sequence is followed, a com-
shielded leads from the signal generator and by means plete section of the receiver can be aligned without
of loose coupling. Sufficient coupling between the sig- too many changes in the settings of the test equipment
nal generator and the circuit may very often be ob- used. This saves time and, in addition, will reduce
tained by merely clipping the generator lead to the errors that might occur when attempting to reset the
insulated wiring of the circuit. signal generator exactly to a frequency previously
All that has been said with regard to the sweep used. This last reason is. especially important when
generator is applicable to the marker generator as making adjustments in the trap circuits.
well. In addition, using the two signal generators When manufacturers' alignment instructions are
together may increase the clanger of regeneration and supplied, these instructions follow the sequence rec-
oscillation. This can be prevented by isolating the two ommended by that particular manufacturer. It is our
signal generators from each other by keeping their suggestion that this sequence should be followed.
connecting points at least one stage apart whenever However, in the absence of manufacturers' data, the
possible. It is usually possible to leave the marker alignment procedure outlined here will be of consid-
generator connected to the mixer grid while the sweep erable value. This procedure is arranged in the fol-
generator is moved progressively through the i -f sec- lowing sequence :
tion of the receiver. The output of the marker gen- 1. Sound i -f section.
erator should be kept as low as possible ; only a 2. Trap circuits.
small marker pip is required on the oscilloscope. If 3. Video i -f section.
the marker pip is too large, the response pattern will 4. Oscillator.
be distorted. 5. R -f amplifier and mixer.
The response curve on the oscilloscope may appear
in reverse, that is, the horizontal sweep line may be Sound I -F Section
deflected either upward or downward depending upon
the arrangement of the video detector circuit. The The coupling circuits in the sound i -f section may
polarity of the pattern is of no consequence ; either be either single- or double -tuned as shown in Fig.
pattern is equally useful. The low -frequency end of 12-2. The coils are usually slug tuned and are reso-
the response curve may be on either the right or left nated with a shunt capacitance, which may be an
side of the pattern. This is dependent upon the polar- actual capacitor or may be the output or input capaci-
ity of the horizontal sweep voltage supplied to the tance of the tube. The single -tuned circuit shown in
oscilloscope by the sweep generator. The low- and (A) of Fig. 12-2 has a single -peaked response char-
high -frequency extremes of the pattern should be acteristic as shown. This response curve is broadened
determined by means of the marker generator and by the comparatively small value of resistance, R,
SOUND LF. SOUND I.F. SOUND I.F.
8+ B+
(A) SINGLE- TUNED (B) DOUBLE -TUNED, UNDERCOUPLED (C)DOUBLE-TUNED, OVERCOUPLED
as indicated above. Note the simi-

Frc. 12-2.-The coupling circuits in the sound i -f section may be single- or double -tuned
double -tuned circuit in (C) results in a
larity between the response curves of (A) and (B) and how the overcoupling of the
widened response curve.
which is in the grid circuit of the following stage. The peaks should be of equal amplitude. This completes
circuit shown in (B) of Fig. 12-2 is an undercoupled, the adjustment of the sound i -f section.
double -tuned circuit having the saine essential char- At this point it is advantageous to check visually
acteristics as that shown in (A). As will be noted, the response characteristic of the- sound section by
the response curve is also similar. Iti (C) the circuit means of a sweep generator and the oscilloscope. To
is also double tuned, but in this case it is overcoupled do this, connect the sweep .generator to the grid of
to provide a wide response characteristic. This curve the mixer tube. The marker generator is left at the
has two peaks, with the sound i -f center frequency same point, but is now loosely coupled. This loose
occurring between the two peaks. coupling may be accomplished by means of a small
Inasmuch as the circuits shown in (-A) and (B) series capacitor or by merely clipping the marker
of Fig. 12-2 have a single peak, alignment can be generator output lead to the insulated grid lead.
performed with the marker generator and the VTVM. Note: The two generators should never be con-
However, clue to the double-peak response of the cir- nected directly together. Connect the vertical input of
cuit shown in (C), these instruments alone are not the oscilloscope to the discriminator output. The hori-
sufficient to adjust this latter circuit, visual alignment zontal input of the oscilloscope is connected to the
by means of a sweep generator and an oscilloscope horizontal sweep voltage supplied by the sweep
being required. generator.
We will now consider the alignment procedure of Set the sweep generator to the frequency of the
a sound section having single -tuned coupling circuits. sound i -f section with a sweep frequency of approxi-
This procedure will also apply to the double -tuned,
undercoupled circuit in (B), with the exception that
.two adjustments are made instead of one. The pro- Fir.. 12-3.-The response
cedure is as follows : curve of the sound section ob-
tained at the output of the dis-
1. Connect the marker generator to the grid of the criminator. A, B, and C are
mixer tube. Set the marker generator exactly to the pips obtained .from the marker
generator.
sound i.f.
C
2. Connect the VTVM to the discriminator output
circuit. Set the VTVM to a low scale. mately two megacycles. The pattern shown in Fig. 12-3
Caution : Adjust the signal output from the should be obtained on the oscilloscope. The frequen-
marker generator to a value sufficiently low to cies corresponding to points A, B, and C in the figure
provide the minimum indication required on the can be determined by marker pips from the marker
VTVM. As the stages are peaked, the output of generator. To obtain these marker pips, set the marker
the marker generator must be reduced further. generator to the i.f. of the sound section and vary the
Do not advance the VTVM to a higher scale to marker frequency along the slope of the response
prevent the meter pointer from going off scale curve from peak to peak to determine the frequency
due to alignment of the circuits. Always reduce limits of the curve. These frequency limits are usually
the output of the marker generator, working of the order of 200 to 600 kc apart, depending upon
with the least possible signal to insure correct the particular receiver. If the response curve is not
results. symmetrical or is of insufficient width, it will be nec-
3. First detune the discriminator secondary and essary to retouch the discriminator adjustments.
then adjust the discriminator transformer primary The alignment of the double-tuned, overcoupled
for maximum indication on the VTVM, then pro- circuit shown in (C) of Fig. 12-2 is as follows:
gressively adjust the i -f coupling units, working back 1. Connect the- sweep generator to the grid of the
from the discriminator stage to the first sound i -f mixer tube. Loosely couple the marker generator to
stage: This procedure may be repeated for greater the same point.
accuracy.
2. Set the two generators to the sound i.f. The
4. Adjust the discriminator transformer secondary
sweep generator is set to a sweep of about two mega-
for a balanced condition as indicated by zero reading
cycles.
on the VTVM. This zero point will lie between the
positive and negative peaks and is quite critical. When 3. Connect the oscilloscope to the grid of the limiter
the discriminator is properly balanced, positive and tube. If no limiter tube is employed, completely de -
negative peaks will be obtained as the marker gen- tune the discriminator transformer secondary and
erator is tuned to either side of the sound i.f. These place the oscilloscope across the discriminator output.
4. Adjust the primary and secondary of each coup- channel being received. This beating occurs in the
ling circuit in turn, working back from the last i -f video detector and produces a difference frequency
tube to the first i -f stage. Proper response is indicated equal to 4.5 mc. In receivers having a wide response
by a symmetrical curve peaked on either side of the in the video amplifier stages, it may be necessary to
sound i.f., as shown in (C) of Fig. 12-2. The object employ traps to eliminate this interference. It cannot
is to obtain maximum peaking of the curve while be too greatly stressed that the accurate setting of
maintaining symmetry on either side of the sound i.f. these traps is extremely important for interference -
5. The oscilloscope is, if necessary, moved to the free operation of the television receiver. To insure the
output of the discriminator and the discriminator accurate setting of these traps, the signal generator
transformer is adjusted for the response curve shown must be set exactly to the rejection frequency of the
in Fig. 12-3. The primary of the discriminator trans- trap being adjusted.
former essentially determines the positions of points The following is a suggested procedure for aligning
A and C on the response characteristic. The position the trap circuits :
of point B, which should correspond exactly to the 1. Connect the marker generator to the grid of the
sound i.f. is determined by the adjustment of the dis- mixer tube.
criminator transformer secondary. Both of these ad- 2. Connect the oscilloscope to the output of the
justments should be retouched to obtain fine over-all video detector.
response. 3. Set the marker generator to the sound i.f. with
modulation on. If the trap circuit is not properly ad-
Trap Circuits justed, a signal will appear on the oscilloscope; this
signal will be the modulating signal from the marker
Trap circuits are used in the video section of the
television receiver to reject frequencies that might generator which has been detected in the video de-
cause interference in the picture. Fig. 12-4 shows a tector stage.
4. Adjust the accompanying sound trap for mini-
typical television channel with the two adjacent chan-
mum signal.indication on the oscilloscope.
5. Set the marker generator to the adjacent sound
rub-CI.ANNEL
54 -60 MC
CHANNEL
60-66 MC
3 CHANNEL
66-72
4-{ i.f. This frequency is 6 mc above the sound i.f. of the
MC
channel being received.
5 9.7 5 MC
SOUND
66.75
SOUND
MC 71.75
SOUND
MC 6. Adjust the adjacent sound trap for minimum
CARRIER CARRIER CARRIER
output indication on the,oscilloscope.
55.25 MC 61.25 MC 61 2.514C
7. If the receiver has an adjacent video trap, set
PIX PIS PIS
CARRIER CARRIER CARRIER
the marker generator to the adjacent video i.f. which
54
MC
60
MC
66 is 6 mc below the picture i.f. of the channel being re-
MC MC
Ildebee
ceived. Adjust this trap for minimum output on the
FIG. 12-4.-A typical television channel, No. 3, with adja- oscilloscope.
cent channels 2 and 4.
ADJACENT ADJACENT
nels. Interference with the picture being received may CHANNEL CHANNEL
be caused by its own sound carrier, the sound carrier
of the next lower channel, or the picture carrier of
the next higher channel. These carriers beat with the
local oscillator and form i.f.'s which may lie within 26.75
NC
the pass band of the video i -f section of the receiver
and, thereby, cause interference unless rejected by Vfl iI.25 MC
.9 73 27.26
trap circuits. This is illustrated in Fig. 12-5. These M CHANNEL MC
interfering frequencies are known as accompanying

sound, adjacent sound, and adjacent video frequen-
CHAMNiL
PIE CARRIER
SOUND CARRIER CUANNEL
SOUND
i
2
CARRIER
cies. Most receivers utilize a trap in the video i -f sec- FIG. 12-5.-An over-all video response characteristic for
tion to eliminate adjacent and accompanying sound channel 3, showing the interfering frequencies which must be
eliminated by means of traps.
i.f.'s. Some receivers, in addition to these traps, em-
ploy a trap for thé adjacent picture i.f.'s. In elimination of the 4.5 -mc video interfering fre-
Another form of interference may be caused by the quency, connection of the marker generator and the
beating together of the sound and picture i.f.'s of the oscilloscope will be determined by the location of the
video trap in the receiver circuit. Usually the marker 2. Connect the VTVM to the output of the video
generator can be connected to the grid of the first detector ; set to a low d -c scale.
video amplifier, with the oscilloscope being connected 3. Set the marker generator to the exact frequency
to the grid of the picture tube. With the marker gen- of each i -f coupling circuit and peak each in turn,
erator set at 4.5 me and modulated, a signal will ap- working progressively back from the video detector
pear on the oscilloscope if the trap is misaligned. If to the first i -f stage. The exact frequency of each
no indication is received on the oscilloscope, it may coupling circuit will be determined by the specifica-
also mean that detection of the modulated signal is tions of the particular receiver.
not occurring. In this case, use a crystal detector probe
NOTE In order to check the over-all i -f response,
:
in series with the input lead of the oscilloscope in

it will be necessary to obtain a visual indication of
order to detect the modulation on the 4.5 -mc signal circuit conditions following the above adjustment.
input. With a signal indication on the oscilloscope, This is done as follows
:
adjust the 4.5 -mc video trap or traps for minimum
4. Connect the sweep generator to the grid of the
indication.
mixer tube. The marker generator is left at this point
Video I -F Section also, but is loosely coupled to prevent interaction.
5. Replace the VTVM with the oscilloscope at the
The alignment method used in making adjust-
output of the video detector.
ments in the video i -f section of the television receiver
depends upon the type of interstage coupling used. 6. Set the sweep generator to the center frequency
There are two principal types of coupling circuits in of the i -f pass band with a 10 -mc sweep. The over-all
i -f response curve will appear on the oscilloscope. Use
use at the present time, the stagger -tuned and the
overcoupled circuits. In most stagger -tuned circuits, the marker generator to determine the frequency lim-
each of the interstage coupling circuits is tuned to a its and identifying points on the pattern.
different frequency. The over-all response character- 7. If necessary, retouch the alignment adjustments
istic of the i -f section is determined by the combina- to obtain the proper over-all i -f response characteristic
tion of the individual characteristics of each stage and for the particular receiver.
will, therefore, have the required frequency bandwidth In the alignment of a video i -f section incorporating
for the video i-f section. In most overcoupled circuits, overcoupled interstage transformers, a set of re-
the individual stages are all tuned to the same fre- sponse characteristic curves will be required. The
quency. However, the coupling between primary and following procedure may be used :
secondary of each interstage i -f transformer is great 1. Connect the marker generator to the grid of the
enough to produce a broad double -peaked response mixer tube where it remains during the entire align-
curve. This response curve can be further broadened ment procedure. The marker generator is used to
by shunting the transformer windings with loading provide marker pips for frequency identification.
resistors to provide the required response charac- 2. Connect the oscilloscope to the output of the
teristic. video detector.
In comparing the two types of coupling circuits, it 3. Connect the sweep generator to the grid of the
should be noted that the individual stages in the stag- final i -f amplifier. Set the sweep generator to the cen-
ger -tuned method of coupling are single peaked, while ter frequency of the i -f pass band with a 10 -mc sweep.
those of the overcoupled systems are double peaked. 4. Adjust the primary and secondary of the final
As in the alignment of the sound i -f section, the stag- i -f coupling transformer for the proper response pat-
ger -tuned coupling circuits, being single -peaked, can tern. Use the marker generator to identify limits and
be aligned by means of the marker generator and the peaks of the response curve.
VTVM. Since the overcoupled circuits are double - 5. Move the sweep generator progressively back
peaked, the use of the sweep generator and the oscil- from stage to stage, adjusting each interstage trans-
loscope is required in order to obtain a visual response former in turn until all are aligned for proper re-
characteristic. A typical over-all video i -f response sponse.
curve is shown in Fig. 12-5. 6. In order to obtain the over-all i -f response, the
The following procedure is a suggested method for sweep generator is connected to the mixer grid. The
aligning a video i-f section containing stagger -tuned i -f response characteristic obtained should conform to
coupling circuits. the reference characteristic. Note particularly the
1. Connect the marker generator to the grid of the position of the video i -f carrier on the response curve
mixer tube. as indicated by the marker pip. The video i -f carrier
should occur about halfway up the sloping side of the positive and a negative peak. The adjustment is criti-
response curve. Also check to see that the trap fre- cal and must be carefully performed.
quencies are outside of the pass band and properly 5. Repeat steps 3 and 4 for each channel to be
attenuated. If these frequencies are not attenuated, aligned.
realignment of the trap circuits will be required. If a sufficiently accurate signal generator is not
available, the oscillator may fte aligned by using the
Local Oscillator signal from the television station. To do this, substi-
tute an antenna for the signal generator and set the
The local oscillator used in television receivers is
receiver to the channel to be aligned. If the station is
operated at a frequency that is higher than the carrier
on the air at the time, perform steps 2 and 4 in the
frequencies of the television channel. The signal of
above procedure.
the oscillator beats with the incoming sound and video
carriers to produce difference frequencies which are
R-F Amplifier and Mixer
equal to the sound and video i.f.'s of the receiver.
For example, in a typical television receiver, the 'l'he number of tuned circuits and adjustments re-
video i.f. may be equal to 26.6 mc and the sound i.f. quired in the r -f amplifier and mixer circuits depends
equal to 22.1 hic. When this receiver is tuned to chan- upon the particular receiver design. A number of
nel two, the local oscillator will operate at a frequency receivers employ an untuned antenna input circuit,
of 81.85 mc, since the video carrier is 55.25 mc and but have a double -tuned coupling circuit between the
the sound carrier is 59.75 mc. r -f amplifier and the mixer. In other receivers both
Since the receiver i -f sections have already been coupling circuits are tuned, but only single tuning in
aligned for proper response, the frequency of the local each circuit is used. These circuits can be tuned to the
oscillator must now be adjusted so that the proper different channels by varying either the circuit capaci-
i.f.'s can be produced. This adjustment is especially tance or inductance, or by switching in different
critical for the sound section, as the sound i -f section values of inductance or capacitance. The circuits are
has a limited bandwidth in the order of several hun- tuned to provide maximum response to each channel,
dred kilocycles as compared to the 3- or 4 -megacycle consistent with a sufficiently wide bandwidth to pass
bandwidth of the video i -f section. An error of a few all frequencies of the television channel. Some re-
hundred kilocycles in the frequency of the local oscil- ceivers require adjustments for each channel, while
lator will have a relatively small effect on the video others may require adjustments for only one or two
response but it may be sufficient to completely elimi- channels ; the other channels being preset in accord-
nate the sound. Because of this, the sound i.f. gener- ance with the design of the receiver. While the exact
ally is used as a reference when adjusting the oscil- aligning method of the tuned circuits in the r -f ampli-
lator. The sound i.f. is also used as a reference because fier and mixer circuits may vary somewhat with dif-
the discriminator provides an accurate indication of ferent receivers, the following general procedure can
frequency variation. It should be understood that if be used with satisfactory results.
the receiver has been properly aligned up to this point, 1. Connect the sweep generator to the antenna ter-
adjusting the oscillator for the correct sound i -f re- minals. If the receiver input is balanced, use the bal-
sponse will automatically provide the correct video anced output from the sweep generator. Set the re-
i -f response because of the fixed 4.5 -mc difference ceiver to the channel to be aligned.
between the sound and video carriers. The following 2. Loosely couple the marker generator to the an-
procedure can be used for aligning the oscillator : tenna terminals. Do not directly connect the two gen-
1. Connect the marker generator to the antenna erators. The marker generator is used to provide
terminals of the receiver. If the receiver has a bal- marker pips for frequency identification.
anced input circuit, the output of the marker gen- 3. Connect the oscilloscope to the output of the
erator should also be balanced. video detector.
2. Connect the VTVM to the output of the dis- 4. Set the sweep generator to the center frequency
criminator. of the channel with a 10 -mc sweep. The over-all video
3. Set the marker generator to the exact sound car- response characteristic should appear on the oscillo-
rier frequency of the channel being aligned. The ac- scope.
curacy of this frequency setting is very important. 5. Adjust the tuned circuits, working back from
4. Adjust the oscillator frequency for zero output the mixer to the antenna terminals, for maximum in-
from the discriminator. This zero will lie between a dication consistent with proper response. The re-
sponse pattern obtained should coincide with the Safety Precautions
over-all video i -f response characteristic previously In view of what has just been said about cautions in
obtained. A typical example of this response curve is regard to high voltage, a few additional precautions
shown in Fig. 12-5. are in order at this point.
In some receivers, it is possible to align the tuned 1. Do not attempt to remove the chassis from its
circuits of the r -f amplifier and mixer by observing cabinet without first turning off the power switch and
the response pattern of these tuned circuits alone, removing the power plug from the a -c outlet.
rather than the over-all i -f response curve. With this 2. Handle the picture tube with extreme care.
method, the oscilloscope is connected to the mixer When placing it on the work bench, make sure that it
tube instead of to the video detector. Detection occur- is placed so that it cannot roll off the bench or bump
ring in the grid circuit of the mixer tube will provide against other objects. If the supporting mount is not
a signal for the oscilloscope. The sweep and marker removed from the cabinet with the tube, a jig should
be fashioned to support the tube on the work bench.
generators are used as before, but the response pat-
Use a thick soft padding under the tube when stand-
tern now obtained should be wide enough to include
ing the tube on its face, to prevent scratching of the
surface. Retain the cartons from new picture tubes to
PICTURE SOUND
CARRI E R CARRIER Fic. 12-6.-The response protect any tubes not in immediate use.
pattern of the tuned cir-
cuits of the r-f and mixer 3. Before doing any soldering within the set, be
circuits. sure the power is turned off and the plug removed
from the a-c outlet. Discharge all filter capacitors by
both picture and sound frequencies of the channel. shorting them to ground by means of an insulated
The adjustments are made for maximum amplitude lead. Always connect the lead to ground first.
of the response curve consistent with sufficient band- 4. As a further caution in regard to the picture
width, as shown in Fig. 12-6. tube, do not touch the high -voltage terminal on the
tube even though the tube is disconnected. Some tubes
TROUBLESHOOTING have a coating on the outer as well as on the inner
surface which is usually the second anode. These two
While troubleshooting and repair of a television re-
coatings act as the two plates of a capacitor and will
ceiver may seem at first to present some rather insur-
develop a charge that may last for a considerable time
mountable problems to the average serviceman, in if it is not discharged.
actuality he will find himself dealing with many cir-
5. Do not attempt to reach into tight places inside
cuits which are common to conventional radio re- the television cabinet to make any adjustments. In
ceivers. \Vith a good basic knowledge of radio receiver other words, always allow sufficient room in which to
circuits, and an understanding of the circuits peculiar work. Also, do not stand on chairs, tables, or insecure
to television receivers, which have been covered in objects when making adjustments on sets having an
this book, the servicing of these receivers should pre- elevated mounting.
sent no extraordinary problems. In the final analysis,
a television receiver can be reduced basically to the Test Equipment
sistors, and capacitors -

usual electronic circuit components - tubes, coils, re-
with which every service-
The test equipment used in the servicing of tele-
vision receivers should include the following :
man is familiar. 1. Signal generator.
Of course, the operating frequencies used in tele- 2. Oscilloscope.
vision are much higher than those used in radio re- 3. Vacuum -tube voltmeter or 20,000
ceivers, necessitating greater care in checking circuits ohms/volt meter.
and replacement of parts. This is especially true of the 4. Tube tester.
circuits in the front end of the receiver which operate 5. Hand tools, soldering irons, etc.
at the high frequencies of the various television chan- The signal generator should have a frequency range
nels. Also, the higher operating voltages used in some of 8 to 220 me with provision for a.m. and f.m. and
circuits which range from 7,000 to 30,000 volts, de- separate audio output. This generator can be the one
mand more than usual caution when checking or serv- used as the marker generator for alignment adjust-
icing these circuits. ments.
The oscilloscope can be any conventional type 'used great extent, the owner will be aware that the sound
for test purposes. However, for observing waveforms and picture are not coming through satisfactorily.
an oscilloscope with a wide -band frequency response Since it is quite possible that the trouble may be origi-
will give a more accurate indication. nating at the transmitter, or that interference, such as
The vacuum -tube voltmeter should have a low -scale diathermy, may be distorting the picture temporarily,
range in the order of 3 volts and an input impedance the serviceman should have available in the shop a
of at least 10 megohms. External multipliers should television receiver that can be used for monitoring and
be available when measuring higher voltages beyond should check whether or not the trouble is coming in
the range of the VTVM. It will also be helpful if the from the transmitter of the station being received.
VTVM has an r -f probe which may be either of the If examination of the set reveals that the defect is
diode type or the crystal type. High -voltage test leads caused by conditions external to the receiver, such as
should be used in making voltage measurements in interfering signals from other sources, repositioning
high -voltage circuits. of the antenna or transmission line may be required.
Any good tube tester with facilities for testing all In the case of diathermy very little can be accom-
the later -type tubes will be suitable. It should be kept plished other than an explanation to the set owner re-
in mind, however, that checking a tube in a tube tester garding the cause of the defective condition. In any
is not always conclusive. In many instances, tubes event the serviceman must determine definitely
such as the 6AC7, 6AB7, 6J5, and 6N7 which are whether or not the defect is inside the television re-
used in television receivers have tested sati'sfaçtorily ceiver and proceed accordingly.
in tube testers of the better grade, but performed very After the defect has been localized to the receiver
unsatisfactorily when placed in some circuits of the proper, the next step is to determine which of the
receiver. However, these same tubes when placed in basic five sections within the receiver is not operating
other circuits gave the desired performance. For ex- normally. For facility in isolating the defective stage,
ample, a 6N7 tube may not operate properly as a sync the receiver can be divided into sections consisting of
separator, but will be all right as a deflection oscillator the front end, sound, video, sweep, and power supply
and discharge tube ; or a tube that will not operate sections, as shown in the block diagram of Fig. 12-1.
satisfactorily as an oscillator will function in an am- Quite often the defect can be isolated to one of the sec-
plifier stage. Therefore, the final test of tube merit will tions by observation of the sound and picture and
always be its actual performance in the circuit. without removing the chassis from the cabinet.
A word of caution is in order here. Before inserting If the sound comes through clearly but no picture
a new tube into a suspect stage, be sure that condi- appears on the picture tube, this indicates that the
tions within the circuit will not burn out the new tube. video, sweep, or high -voltage section is defective.
If the tube that has been removed from the set is not The front end which supplies the signal to the sound
burned out, it can be assumed that insertion of the new section, as well as to the video section, is evidently
tube will not damage it, and the substitution test of operating normally ; this is also true of the low -
tube merit can be made in satety. voltage power supply which supplies operating poten-
Solder connections in television receivers are very tials to all these sections. If a raster can be obtained
critical and must be made carefully. Do not use an on the picture tube by turning up the brightness con-
iron so hot that the solder crystalizes, nor so cool trol, the trouble is quite definitely isolated to the video
that resin joints result. It is recommended that the hot section. If no raster is obtained, then the trouble very
iron be used to heat the joint and that the solder be likely is in either the sweep or the high -voltage sec-
applied directly to the heated work so that it flows tions. Proper high -voltage supply will be denoted by
evenly into the connection. some sort of indication on the picture tube, whereas
lack of high voltage will result in no indication.
Localizing the Trouble If no sound is present, or if the sound is distorted,
but a normal picture is received, then the trouble is
When a serviceman receives a call to repair a tele- isolated to the sound section, inasmuch as normal re-
vision set, he usually receives some hint as to the pos- ception of the picture indicates proper operation of the
sible location of the defect. He may be informed that other sections of the receiver. However, defects in the
no picture and/or sound can be obtained, or that the sound signal can occur in the front end without notice-
quality of one or both is poor. Though trouble indi- ably affecting the picture. An example of this would
cations given by the set owner cannot be expected to be a slight detuning of the oscillator.
be very accurate and cannot be relied upon to any If no sound or picture is received, the trouble must
be in some section of the'receiver that is common to The systematic procedure for isolating the defective
both. This can be either the front end or the low -volt- section is as follows :
age power supply sections. If turning up the bright- Check 1. Low-Voltage Power Supply. Using
ness control provides a raster on the picture tube, the the voltmeter, check for proper voltage at the output
low -voltage supply is eliminated as the source of of the power supply.
trouble which is then isolated to the front end. Of Check 2. High -Voltage Power Supply. Using
course, in sets incorporating a separate low -voltage the VTVM and external multipliers, if necessary,
supply for the sweep section, this test is valueless. check for proper voltage at the output of the high -
Lack of both sound and picture combined with no voltage power supply. Caution: Use extreme care
raster on the picture tube indicates a completely dead when making this check.
set, and this is probably caused by a blown fuse, de- Check 3. Sweep Section. Connect the oscillo-
fective power transformer, defective line switch or scope to the horizontal and vertical deflecting plates,
cord, or no line voltage. in turn, (or to the deflecting yoke terminals) and
With experience, the serviceman will develop a check for proper waveform and peak -to-peak voltage.
keener sense for interpreting the indications observed These should conform to the reference waveform and
in defective television receivers and the ability for peak voltages for the particular receiver.
speedy localization of the defective section simply by Check 4. Video Section. Set the signal gen-
looking at the picture, listening to the sound, and erator to the video i.f. and connect it to the input of
manipulating the panel controls. the video i -f section. Turn on the a.m. of the signal
When the serviceman has isolated the trouble to a generator and check for a series of light and dark bars
particular section of the receiver, his next step should in the picture tube.
be to check the tubes in the defective section, inas- Check 5. Sound Section. Set the signal gen-
much as the tubes are the first suspects in any inoper- erator to the frequency of the sound i-f section and
ative receiver. Often this can be done without remov- connect it to the input of the sound i -f section. Set for
ing the chassis from the cabinet by reaching in and f.m. with a deviation of ±25 kc and listen for a tone
feeling for a cold tube. On occasion, a whole series of in the speaker.
tubes may be found cold indicating series wiring of Check 6. Front End. Set the signal generator
their heaters ; so that the whole series has gone dead to the sound carrier frequency of the channel to which
as a result of one tube's burning out. If no burned -out the set is tuned and connect it to the antenna input
tube is found, all the tubes in the suspected section terminals. Set for f.m. with a deviation of ±25 kc and
should be tested. listen for a tone in the speaker. An alternative method
If the trouble cannot be determined by an observa- is to set the signal generator to the video carrier fre-
tion of the picture and sound, it is necessary to requency with a.m. and check for light and dark bars in
move the chassis and perform a series of systematic the picture. If necessary, either of these two check
checks with test instruments to isolate the defective methods may be repeated for each of the other
section. channels.
These systematic checks are performed at signifi- Check 7. Sync Circuits. A further check may
cant points in the receiver by means of voltage or be required in the sync circuits to determine whether
resistance measurements, or by injecting a signal into or not these circuits are operating properly. Connect
the input of a section and observing the output indi- the test oscilloscope to the output of the sync sepa-
cation. The output indication can be observed either rator, or the sync amplifier tube if one is used. Check
on the picture tube, the test oscilloscope, or by listen- for proper waveform and peak -to -peak voltage of the
ing to the speaker of the receiver. The checks should sync signals. These should conform to the reference
be performed in the indicated sequence, eliminating waveform and peak voltages for the particular re-
one section at a time until the defective section is dis- ceiver.
covered. Once a defect has been localized to a particu- Correct voltages and waveforms at the stated check
lar section, further checks using signal -tracing meth- points will vary with the type df television receiver.
ods must be performed to determine the faulty stage Obviously, it is not possible in a general servicing
within the section. After the trouble has been isolated procedure such as this to include reference waveforms,
to a particular stage, conventional voltage and re- voltages, and resistance measurements for the many
sistance measurements should be made to determine types of television receivers. In some cases, it may be
the defective component. possible to service a defective television receiver with-
out these reference data. However, if this information input of the video i -f strip, then the fault quite ob-
is available, it will facilitate servicing and greatly re- viously lies in the stages following the detector. In
duce the amount of time spent in repairing a receiver. which case the oscilloscope should be connected to the
These reference waveforms, voltages, and resistance grid of the picture tube. Set the signal generator for
measurements may be found in the complete data on 400 -cycle output and connect it to the grid of the video
current television receivers contained in Rider's Tele- output tube. If an abnormal indication is obtained on
vision Manual 1. the oscilloscope, the trouble is in the video output
When it has been established that the defect is in a stage. If a normal signal is obtained, the signal gen-
particular section of the television receiver it must erator should be moved back to the grid of each video
then be traced to a specific stage. In the high- and amplifier stage in turn and the output observed. If all
low -voltage power supply sections, voltage and resist- video amplifier stages check normal, the signal gen-
ance measurements are made to isolate the fault. How- erator should be left at the grid of the first video am-
ever, in the sweep, video, sound, and front-end sec- plifier and set for a high-level 400 -cycle output. A
tions, signal tracing will facilitate the location of the pattern of light and dark bars should appear on the
faulty stage. picture tube. If this pattern is absent, and if the sweep
If an abnormal indication has been obtained at the section and both power supplies check normal, a bad
output of either the vertical or horizontal sweep cir- picture tube is indicated.
cuits, the oscilloscope should be moved back progres- The sound i -f section can be checked similarly to
sively through the section from the output to the input the video i -f section. However, the signal generator
of each stage until a normal indication appears on the is set to the sound i.f., and f.m. is used. The oscillo-
oscilloscope. The defect will lie in the stage immedi- scope is connected to the output of the discriminator,
atelÿ following the point where the normal indication and the trouble is isolated to either the sound i -f stages
is obtained. Further voltage and resistance measure- or to the audio amplifier stages that follow the dis-
ments will isolate the defective part within the stage. criminator. Localization of the defective stage can be
This part should be replaced with one having the same accomplished as before. However, in checking the
characteristics. audio amplifier stages the loudspeaker may be used
If the defective circuit is traced to the video section, as an output indicator, unless the loudspeaker itself
the stage can be isolated by the following procedure. has been found defective.
The first step is to determine whether or not the If checks 1 to 5 of the procedure for isolating the
trouble lies ahead of the video detector or in the stages defective section are performed with normal results,
following it ; that is, whether it is in the video i -f or in the fault would appear to be in the front end of the
the videb amplifier stages. This can be done .by con- receiver. Check 6 of this procedure will provide defi-
necting the oscilloscope to the output of the video de- nite information in this direction. If no tone can be
tector and the signal generator, amplitude modulated obtained in the speaker with the signal generator con-
and set to the video i.f., to the input of the video i -f nected to the antenna input terminals and set to the
strip. A detected audio signal should be obtained on sound carrier frequency (frequency modulated) , move
the oscilloscope if the section is operating normally up the signal generator to the grid of the mixer tube. If
to the video detector. If an abnormal indication is ob- a normal response is obtained, the trouble definitely
tained, the signal generator should be connected to the lies in the r -f amplifier stage. If a normal response is
grid of the final video i -f amplifier. If the output is still not obtained, the fault is in the mixer or oscillator
abnormal, then a check of the final video i -f amplifier stages. A quick check of the oscillator can be per-
and video detector stages is in order. If a normal indi- formed by measuring the bias voltage developed across
cation is obtained at the final stages, the signal gen- the grid resistor ; exceptionally low or no bias voltage
erator should be moved back stage by stage until the indicates a defect in the oscillator stage. If the oscil-
normal indication is no longer obtained, which will lator checks normal, the fault is in the mixer stage,
indicate that the signal generator is connected to the and voltage and resistance measurements, as pre-
defective stage While moving through the video i -f viously stated, should be made to determine the faulty
stages, a rough check on the gain per stage can be part in the circuit.
made by noting how much the signal generator output
must be reduced in order to maintain the same ampli- Troubleshooting Chart
tude of signal on the oscilloscope. The following troubleshooting chart is included to
If a normal indication from the video detector is supplement the systematic servicing procedure pre-
obtained with the signal generator connected to the viously outlined. When the defect has been traced to
RIDER'S "HOW IT WORKS"
,,articular stage, the chart may be used to check first apply only to an r -f type of high -voltage supply, and
/hose parts in the circuit which are more subject to not to a kick -back or 60 -cycle type of supply. A high -
breakdown. This chart cannot include every possible voltage transformer will be found only in a 60 -cycle
type of trouble that can be encountered in servicing type of supply. The fact that no low voltage is present
television receivers, but an attempt has been made fo will not affect a 60 -cycle type of power supply but will
cover those troubles which experience has shown are render the other two types inoperative. In utilizing
most likely to develop. This chart covers the receiver this chart, the serviceman will select only those prob-
by sections, as in the case of the localizing procedure. able causes that apply to the particular receiver under
These sections are the low -voltage power supply, high - test.
voltage lower supply, sweep, video, sound, and front- In making checks on suspected components, the
end sections. The chart is divided into sections each
: conventional voltage and resistance measurements,
lists the faulty indication, and below this are listed supplemented with signal -tracing methods, are used to
the most probable causes of the observed symptom. In determine the defective part. It should be stressed
a number of instances, the probable causes listed will here that this chart does not represent a cure-all, but
apply only to particular types of receivers incorporat- is meant to supplement the regular servicing proce-
ing specific circuits. For example, in the high -voltage dure and to afford a quick check of the most probable
power supply reference made to an oscillator coil will causes of circuit failure.
TROUBLESHOOTING CHART
ì. LOW -VOLTAGE POWER SUPPLY
Faulty Indication
No Output Voltage
Probable Causes:
No line voltage Short-circuited filter capacitor
Defective line switch Open choke
Blown fuse Loose or broken connections
Defective power transformer Defective interlock switch
Defective rectifier tube
Faulty Indication
Low Output Voltage

Probable Causes:
Low line voltage Filter capacitors changed value
Open filter capacitor Poor solder connections
Leaky filter capacitor Shorted bleeder resistor
Weak rectifier tube Grounded choke
2. HIGH-VOLTAGE POWER SUPPLY
Faulty Indication
No Output Voltage
Probable Causes:
No low -voltage output Open oscillator coil
Defective tubes Open sweep output transformer
Defective high -voltage power transformer No horizontal sweep voltage (check sweep circuit)
Shorted filter capacitor Loose or broken connection
Open filter resistor
Faulty Indication
Low Output Voltage
Probable Causes:
Defective low -voltage supply Open filter capacitor
Weak rectifier tube Leaky sweep output coupling capacitor
Weak oscillator tube Poor solder connection
Leaky filter capacitor Insulation breakdown
195
19(i RIDER'S-"HOW IT WORKS"
3. SWEEP SECTION
Faulty Indication
No Horizontal or Vertical Sweep

Probable Causes:
Defective tubes Open resistors
Open coupling capacitors Open deflection coils
Open or shorted bypass capacitors Open sweep output transformer
Open blocking -oscillator transformer I.00se or broken connections
Faulty Indication
Insufficient Sweep Amplitude
Horizontal Width Control Vertical Height Control

Misadjusted. Misadjusted.
Probable Causes:
Height or width control improperly adjusted Defective deflection coils
Weak tubes Defective sweep output transformer
Resistors have changed value
Faulty Indication
No Raster - Sound Satisfactory

Probable Causes:
No high voltage to picture tube Beam -bending control improperly adjusted
Defective picture tube Loose or broken connection
Beam -bending coil improperly adjusted
TROUBLESHOOTING CHART 197
3. SWEEP SECTION - Continued
Faulty Indication
Nonlinear Horizontal or Vertical Sweep
Vertical Linearity Control Horizontal Linearity Control Vertical Linearity Control

Misadjusted. \lisadjusted. Misadj usted.
Horizontal Linearity Control

Misadjusted.
Probable Causes:
Horizontal or vertical linearity controls improperly Iefective sweep output transformer
adjusted Defective deflection coils or shunt resistors
Defective tubes Capacitors have changed value
Faulty Indication
No Synchronization of Horizontal or Vertical Sweep
Vertical Hold Control Horizontal Hold Control Horizontal Hold Control

Misadjusted. Misadjusted. Misadjusted.
3. SWEEP SECTION - Continued
Horizontal Hold Control Horizontal Hold Control

Misadjusted. Misadjusted.
Probable Causes:
Horizontal or vertical hold controls improperly Defective tubes in sync circuits
adjusted Open coupling capacitor
Insufficient signal pickup at the antenna Resistors or capacitors have changed values in
Low gain in video i -f section sweep oscillators
Defective sync separator stage
4. VIDEO SECTION
Faulty Indication
Poor Picture Definition
Contrast Too High. Contrast Too Low. Misalignment or Improper

Antenna Orientation.
Probable Causes:
Improper adjustment of focus, contrast, or Weak tubes
brightness controls Incorrect alignment of video i -f stages
Weak input signal Defective peaking coils
Incorrect channel tuning adjustment
4. VIDEO SECTION - Continued

Faulty Indication
Noise in Picture
Strong Diathermy Interference. Ignition Interference. Diathermy Interference.
Transients.
Probable Causes:
Interference due to external causes. stich as, Noisy tubes
diathermy. ignition, electric equipment, etc. Noisy resistors or capacitors
Dirty or worn contacts on switches, sockets, etc.
Faulty Indication
Sound in Picture
(Indicated by lines or bands in picture which vary with the audio signal.)
Probable Causes:
Incorrect trap alignment
Oscillator improperly adjusted
Oscillations in video i -f stages
Sound and video i -f leads too close together
Sound Bars or Mlicrophonics.

2(X) RIDER'S - "HOW IT WORKS"
Faulty Indication
Hum in Picture
(Wide dark horizontal bands.)
Hum in Video and Sync Excessive Ripple in Video Excessive Ripple in Video
Amplifier. Amplifier.
120 -Cycle Hum in Video and

Horizontal Scanning.
Probable Causes:
Open or leaky filter capacitor in low -voltage supply Defective decoupling filters in video amplifier stages
Defective tubes Pickup due to misplaced leads
Faulty Indication
Smeared Picture
Probable Causes:
Defective load resistors or peaking coils in video Video i -f stages misaligned
amplifier stages Excessive signal input to receiver
Open bypass capacitors

Faulty Indication
Blurred Picture
Focus Uisadjusted. Focus sadjusted.

Probable Causes:
Improper adjustment of focus control Defective focus control
Defective focus coil
Faulty Indication
Narrow Vertical or Diagonal Lines in Picture
Interference from Another Beat Frequency.

Signal.
Probable Causes: Interference from nearby radio transmitters for other

services such as amateur, communications, f.m., etc.
Faulty Indication
Ghosts
Probable Causes:
Poor antenna orientation
Reflector element required
Ghosts.
>> RIDER'S - "HOW IT WORKS"
5. SOUND SECTION
Faulty Indication
Weak or Distorted Audio

Probable Causes:
Fine tuning control improperly adjusted Sound i -f or discriminator stages misaligned
Defective tube Oscillator misaligned
Open filter capacitor Defective speaker
Open coupling capacitor Faulty connections
Open bypass capacitor
Faz lty Indication
Noise in Audio Output

Probable Causes:
Dirty or worn channel -switch contacts Arcing in high -voltage circuits
Noisy tube Dirty or worn volume control
Overheated resistors I.00se tube -socket contacts
Faulty capacitors Faulty connections
Defective speaker
Faulty Indication
Hum in Audio Output
Probable Causes:
Open filter capacitor Poor ground connections
Defective tube Pickup duc to misplaced leads
6. FRONT END
Faulty Indication
No Sound and No Picture; Raster OK
(on all channels)
Probable Causes:
Defective tubes Short-circuited bypass capacitors
Open antenna input circuit Open load resistors
Defective channel switch Faulty connections
Open coupling capacitors
M
6. FRONT END - Continued
Faulty Indication
No Sound and No Picture; Raster OK

(on one channel)
Probable Causes:
Defective antenna, r -f, or oscillator coil corresponding Oscillator for this channel incorrectly aligned
to dead channel
Faulty Indication
No Sound; Picture Satisfactory

(on one channel)
Probable Causes:
Oscillator for this channel slightly misaligned
Normal Picture.
The patterns used in the Troubleshooting Chart are reproduced through the courtesy of the following
companies : Radio Corporation of America, General Electric Co., Allen B. Du Mont Laboratories,
Inc., Farnsworth Television and Radio Corporation, Electro -Technical Industries, Motorola. Inc.
-
1= -
kti
47'

Television How It Workes Rider 1948

Uploaded by

Copyright:

Available Formats

Television How It Workes Rider 1948

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Television How It Workes Rider 1948

Uploaded by

Copyright:

Available Formats

TELEVISION

JOHN F. RIDER PUBLISHER, INC.

JOHN F. RIDER PUBLISHER, INC.

First Printing, June, 1948

All rights reserved including that of translation into the

Printed in the United States of America

stone on which to build -

GENERAL ASPECTS OF THE TELEVISION SYSTEM

(A)-SOUNO TRANSMISSION SYSTEM

(B)-TELEVISION TRANSMISSION SYSTEM

have "arrived" many years ago. In television, we

an area 1.5 by 2 inches, giving 150 X 200 - 30,000

TRAN5M/TTER RECE/ VER

/,l ¡25 27 ,ßp429 30 /.( /

ELECTRICAL SIGNAL Fig. 1-4.-The photoelectric tube scans

SECONO ANOOE CONOUCTOR

Fig. 1-6.-The different elements of a picture tube are

pEfE TEp E.T/_A_L~

Recently it was discovered that a metallic film

surface of the Kinescope tube improved the distri- CCrLECTING COIL

bution of light over the screen. This metallic film

through the aluminum and strike the luminescent PAPER TUBE

material causing it to fluoresce in the usual manner. Courtesy RCA

1-10 are easily identified. Light images formed on the signal

Fig, 1-13.-An Image Orthicon is shown in Fig. 1-13(A).

CATHODE (ZERO) DECELERATING RING

SIGNAL OUTPUT PHOTO -CATHODE

frame between two rasters (called "fields") , each

a, changes gradually to a brilliant white at b, and Signal and Sync Pulses

VERTICAL. BLANK/NG TOP OF -

525 LINES. JO FRAMES PER SEC., 60 FIELDS PER SEC.. INTERLACED

.5N W%MlY CARNIER qXTK(

(-DIAGRAM C SHOWS ENLARGED DETAIL VIEW OF SIGNAL IN VIEW a BETWEEN LINES C -C

3 HTIME FROM START OF ONE LINE TO START OF NEXT LINE.I/IS,750 SEC.

6 -RECEIVER VERTICAL RETRACE SHALL BE COMPLETE AT END OF .07 V

ED) DETAR. BETWEEN D-O

faithful reproduction because of the rapid iriation

The Modulated Wave V/pEO

wave produced when a scene is scanned and the

V/DEO VIDEO CHANNEL

FREQUENCY CHARACTERISTICS OF THE TELEVISION SYSTEM

In order to understand many of the features of CHANNEL FREQUENCY CHANNEL FREQUENCY

designated as f0. If the frequency of the carrier is

curve is assumed to be that for channel number 1, Operating Bandwidth Characteristics

TELEVISION RECEIVING ANTENNAS

CURRENT VOLTAGE DIPOLE

called the real resistance; the other type is often called

type, and the coaxial line. The characteristic imped-

a direct function of the length of the line, although

where D is the distance between centers of the wires

For example, let us compute the characteristic im- Thus

and measures 22/64 of an inch between centers with 138 19/64

electric material. For air E = 1, for polyethy- Impedance Matching

characteristic impedance of the line are each about

FIG. 3-11.-A folded half -wave dipole antenna, where d, the

where L is equal to the length of a half wavelength in 492 x .915 450

one-half wavelength long, the actual physical length

The distance D that the reflector is spaced from the REFLECTOR

dipole is a factor in the amount of increased signal GUIDE LOOPS