Camera Chains - American Radio History
TEE [VISION
BROADOASTINO
Camera Chains
by Harold
r
II
pRO 1
II
U18700 3828588
II
A
E. Ennes
Television Broadcasting
Camera Chains
by
Harold E.
Fennes
HOWARD W. SAMS & CO., INC.
THE BOBBS -MERRILL CO., INC.
INDIANAPOLIS
KANSAS CITY
NEW YORK
FIRST EDITION
FIRST PRINTING -1971
Copyright © 1971 by Howard W. Sams & Co., Inc., Indianapolis,
Indiana 46206. Printed in the United States of America.
All rights reserved. Reproduction or use, without express permission, of editorial or pictorial content, in any manner, is prohibited.
No patent liability is assumed with respect to the use of the infor-
mation contained herein.
International Standard Book Number: 0- 672 -20833 -4
Library of Congress Catalog Card Number: 71- 157802
Preface
The manufacturers of television camera chains normally provide instruction manuals that range from preliminary (sketchy) data to elaborate
coverage of circuit theory, operations, and maintenance of the specific
equipment. Such manuals obviously cannot delve into certain training
programs that are usually necessary before adequate comprehension of
modern broadcast technology can be gained.
It is the purpose of this book to provide the fundamental and advanced
training that is necessary if full benefit is to be obtained from the information in modern instruction books. To do this most effectively, where
possible, complete, detailed schematics have been avoided, and, instead,
use has been made of block diagrams with simplified diagrams of individual blocks under discussion. The overall system concept is stressed so
that the reader can more readily grasp the meaning of a specific circuit
adjustment in terms of its effect on system performance.
For the serious student, or for the practicing engineer, the information
contained in the author's previous two books should be considered pre requisities to a study of this volume. These books are Workshop in Solid
State and Television Broadcasting: Equipment, Systems, and Operating
Fundamentals. Both are published by Howard W. Sams & Co., Inc. These
books and the present volume serve a dual purpose, as basic textbooks for
students or beginners and as factual guidebooks for practicing technicians.
The author extends his appreciation to the following organizations for
providing information and photographs used in this book: Albion Optical
Co.; Ampex Corporation; Amphenol Corporation; Belden Corporation;
Cohu Electronics, Inc.; Hewlett- Packard Co.; International Video Corporation; Kliegl Bros. Lighting; Philips Broadcast Equipment Corp.; RCA;
Shibaden Corp. of America; Tektronix, Inc.; TeleMation, Inc.; Telesync
Corp.; Visual Electronics Corporation; WBBM-TV; and WTAE -TV.
HAROLD E. ENNES
f
Contents
CHAPTER
1
11
STUDIO LIGHTING
Amount of Light
Air Conditioning
1 -3. Power Requirements for Lighting
1 -4. The Suspension System and Power Distribution
1 -5. Lighting
Control
1 -1.
11
1 -2.
23
24
25
CHAPTER
26
2
THE SYSTEM CONCEPT
36
NTSC and FCC Color Standards in Practical Form
Evolution of the Color-Bar Signal
2 -3. Defining and Recognizing "Distortions" in NTSC Color
2 -4. Digital Concepts
2 -5. A Digitally Controlled Color Camera
2 -1.
36
2 -2.
55
57
CHAPTER
72
81
3
CAMERA MOUNTING, INTERCONNECTION FACILITIES,
AND POWER SUPPLIES
3 -1.
3 -2.
3 -3.
3 -4.
3 -5.
3 -6.
3 -7.
3 -8.
The Camera Pan and Tilt Cradle
Pedestal Dollies
The Crane Dolly
Prompting Equipment
Camera -Chain Power Supplies
Camera -Chain Power Distribution
Power -Supply Maintenance
Interconnecting Facilities
90
90
93
96
97
99
107
112
123
CONTENTS
CHAPTER 4
134
THE CAMERA PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
4 -1. The Image Orthicon
4 -2. The Yoke Assembly
4 -3. Yoke Maintenance
4 -4. Special Notes on the 3 -Inch Field -Mesh I.O.
4 -5. The Image-Orthicon Cooling System
4 -6. The Vidicon
4 -7. The Lead -Oxide Vidicon
4 -8. Camera Optics
4 -9. The Iris -Control Servo
CHAPTER
134
152
157
159
161
162
180
182
190
5
193
VIDEO PREAMPLIFIERS
The Vacuum-Tube Video Preamplifier
Solid -State Video Preamplifiers
5 -3. Maintenance of Video Preamplifiers
5 -1.
193
5 -2.
198
205
CHAPTER 6
222
VIDEO PROCESSING
6 -1. The True Meaning of Bandwidth
6 -2. The Television -Camera Resolution Chart
236
244
6 -3. Clamping Circuitry
6 -4. Aperture -Correction Circuitry
6 -5. Cable Equalization
251
251
6 -6. Gamma Correction
6 -7. Blanking and Sync Insertion
6 -8.
258
260
Testing and Maintenance
CHAPTER
222
233
7
271
PULSE PROCESSING AND TIMING SYSTEMS
7 -1.
7 -2.
7 -3.
7 -4.
7 -5.
7 -6.
The Automatic Timing Technique
Level-Control Pulses (Manual and Automatic)
Level Calibration and Test Pulses
Camera Deflection Circuitry
Pickup-Tube Protection
Shading -Signal Formation
271
279
284
292
299
303
CONTENTS
CHAPTER
8
308
CAMERA CONTROL AND SETUP CIRCUITRY
8 -2. The Marconi Mark VII Four-Plumbicon Color Camera
308
315
The RCA TK -44A Camera Control
The Camera Waveform Monitor
8 -5. Camera Interphone
321
327
333
8 -1.
The RCA TK -60 Monochrome Camera Chain
8 -3.
8 -4.
CHAPTER 9
337
THE SUBCARRIER AND ENCODING SYSTEM
The Dot Structure in NTSC Color
Power -Line Crawl on Color Standards
Adjusting the Subcarrier Countdown
9 -4. Solid -State Counters
9 -5. Final Subcarrier Generator Countdown Check
9 -6. Final Check on Sync- Generator Color Lock
9 -7. Setting the Color-Subcarrier Frequency
9 -8. The Color -Sync Timing System
9 -9. Adjustment of Burst -Key Generator With Encoded Signal
9 -10. Adjustment of Burst -Key Generator by Itself
9-11. The Encoding Process
9 -12. The Vectorscope
9 -1.
9 -2.
9 -3.
..
337
339
340
345
348
349
349
350
354
356
358
379
CHAPTER 10
394
COLOR PICTURE MONITORING SYSTEMS
10 -1.
10 -2.
10 -3.
10 -4.
10 -5.
10-6.
Analysis of Basic I -and -Q Decoder
Color Picture -Tube Circuitry
Adjustment of Color Monitors
The RCA TM -21 Color Monitor
The X and Z Demodulator
Use of the Color Monitor in Matching Techniques
CHAPTER
394
413
418
423
432
434
11
437
PREVENTIVE MAINTENANCE
11 -1.
The Vidicon Film Chain
11 -2. General Preventive Maintenance
11 -3.
Troubleshooting
437
442
445
CONTENTS
APPENDIX
ANSWERS TO EXERCISES
INDEX
k
15
-i6
1
CHAPTER
1
Studio Lighting
The television camera depends primarily on the reflected light it "sees"
through the optical system. Optimum performance is possible in the studio,
where light can be controlled to shape the "taking" characteristics. In an
"adequate" studio lighting system, we are concerned with the following
basic factors:
1.
Amount of light, and color temperature
2. Flexibility
3. Light control
It is assumed that the reader has basic knowledge of the types and techniques of lighting.'
1
-1.
AMOUNT OF LIGHT
It is possible to relate foot -candles (fc) to watts, lumens, etc., but such
a study, while interesting from an academic viewpoint, is of little value
to the practicing operator. Information regarding the amount of light at a
given distance and direction from a given source is furnished by manu-
facturers of lighting equipment. Any conversion of watts to foot- candles
to lumens is useless unless the particular angle of throw, type of lamp,
and type of reflector are accounted for.
The number of lighting fixtures may vary from 20 to 25 in a small
limited-program studio to more than 200 in a larger multiple -purpose
studio. The amount of light required depends on the size and purpose
of the studio, whether programs are in monochrome or color, and the degree of flexibility required. The key word for good studio lighting systems is flexibility.
'See, for example, Harold E. Ennes, Television Broadcasting: Equipment,
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co.,
Inc., 1971).
11
12
TELEVISION BROADCASTING CAMERA CHAINS
From previous experience, the foundation for an adequate amount of
light can be laid as in Table 1 -1. The higher values are representative of
more complex productions, such as large variety shows with liberal use of
effects and modeling lighting both foreground and background. Note in
connection with this table that the quartz lamp has more light output per
watt of input power over a broader area than the incandescent lamp has.
Also, the size and shape of the quartz lamp permits design of reflectors
of higher efficiency.
The net production area to which Table 1 -1 applies is the actual area
used within the cyclorama curtain. The net production area can be approximated by subtracting an area 4 feet wide all around the perimeter of the
studio. For instance, a 40' X 60' studio (2400 square feet) has a net production area of 32' X 52', or 1664 square feet. However, smaller studios,
such as the one in the following example (20' X 30') , do not always employ cycloramas, and the wall -to-wall dimensions normally are used for
these studios.
Table
1
-1. Light Requirements for Net Production Area
Monochrome
100 -125 fc
Color
250 -350 fc
100% Incandescent
100% Quartz
25 -35 W /ft2
(Nominal: 30 W /ft2)
75-110 W /ft2
(Nominal: 90 W/ft2)
(Nominal: 15 W /ft2)
40-60 W /ft2
(Nominal: 50 W /ft2)
10 -20 W /ft2
A typical incandescent lighting package for a 20' X 30' studio with a
ceiling is illustrated in Fig. 1 -1. On the basis of the nominal
values shown for incandescent lights in Table 1 -1, this 600- square-foot
studio would require a total of 18,000 watts for monochrome, or 54,000
watts for color. Note from the equipment list in Fig. 1 -1 that two wattages
are listed for each fixture. In most instances, the lower-wattage lamps
would be used for monochrome, and the higher- wattage lamps would be
used for color.
Fig. 1 -2A gives the photometric data for a 2000 -watt incandescent unit
with a 12 -inch round lens. Note that for the flood operating position,
about 65 foot -candles are available at 20 feet directly on the axis. Since
the flood beam is broad (about 55° ) , essentially the same number of footcandles exists up to about 7 feet off the axis at this 20 -foot distance. In
the spot operating position (beam about 10° ) , almost 760 foot- candles
are available at 20 feet directly on the axis, but only about 80 foot- candles
are available just 2 feet off the axis at the 20 -foot distance.
For illustrative purposes, Fig. 1 -2B gives the same general data for a
5000 -watt unit with a 16 -inch round -beam lens. The graph for the flood
position is for a 30 -foot throw, and the spot -position graph is for a 60 -foot
throw.
15 -foot
13
STUDIO LIGHTING
Fig. 1 -2C gives the photometric data for a 1000 -watt unit equipped
with an 8 -inch oval -beam lens. Note that because of the beam shape,
horizontal and vertical coverage must be considered separately.
See Table 1 -2 for multiplying factors for distances other than those
shown in Fig. 1 -2. To use this table to find photometric data for throws
other than those shown on test charts, apply the following procedure:
Locate the throw distance of the test in the group headed "Throw
Distances Shown in Tests."
2. Directly below this figure is a series of actual throw distances desired. Pick the figure in the vertical series which corresponds to the
distance at which you want photometric data.
1.
Table
1
-2.
Multiplying Factors for Lighting Data
Multiplying Factors
Throw Distances Shown in Tests
10'
15'
20'
30'
50'
Multiply
Multiply
Dimensions
by
Illumi-
20'
0.33
9.0
30'
0.5
4.0
0.66
2.25
0.67
2.25
0.75
1.8
0.8
1.5
50'
0.83
1.44
60'
1.0
1.0
70'
1.16
0.73
1.2
0.69
1.25
0.64
80'
1.3
0.56
90'
1.5
0.44
100'
1.7
0.36
1.75
0.32
2.0
0.25
2.5
0.16
2.7
0.14
3.0
0.11
3.3
0.09
60'
Actual Throw Distances Desired
10'
10'
15'
25'
10'
40'
20'
15'
7.5'
40'
25'
10'
15'
20'
30'
50'
35'
60'
25'
40'
20'
30'
15'
45'
75'
50'
25'
35'
20'
30'
40'
50'
25'
40'
30'
50'
60'
100'
120'
nation
by
Courtesy Klieg! Bros. Lighting
14
TELEVISION BROADCASTING CAMERA CHAINS
When this figure is chosen, refer to the "Multiplying Factors" column directly to the right for the proper factors to use in multiplying dimension data and illumination data.
4. Example: If a test was calculated at a throw of 30' and you require
data for this unit at a throw of 50', find the column headed 30' and
go down this column until the figure 50' is located. Then, by extending directly to the right, you will find the multiplying factors 1.7
for dimension and 0.36 for illumination.
3.
30'
r
o
ó
ó
Jti
o
0
44
Legend
Studio Plan
Control Room
0
44N6TVC 500-Watt
O44N8TVC 1000 -Watt
C1155TVG 1000 -Watt
Studio Elevation
\
>-C Lazy-Boy Pantograph
\\
/
/
I
/
Fig.
1
-1. Incandescent
15
STUDIO LIGHTING
Table
1
-3. Performance of
"Standard" Quartz Fresnel
Performance at 20 Feet
Minimum Spot
Catalog
Number
Lens Wattage
Hours
Life
Spread
l'
Unit
Dimensions
Maximum Flood
FC
Spread
FC
Length Diameter
3518
3518
12"
1000
2000
150
150
31/2' x 31/2'
500
400
15'
15'
40
8"
55
131/4"
131/4"
13"
13"
3521
12"
2000
150
3'
500
17'
60
18"
18"
x 4'
Courtesy Kliegl Bros. Lighting
The variation of foot -candles per watt from luminaires is shown by
Tables 1 -3 through 1 -5. These tables illustrate not only the higher light
output per watt for quartz lamps as compared to incandescent lamps, but
also that the design considerably affects the actual foot -candles per watt
of input power.
EQUIPMENT LIST
Fixtures
10- 1155TVG
750 /2000 -Watt Standard Scoop 18"
75 /100 -Watt Fresnel Camera Light 3" Lens
10- 44N6TVG 500 /750-Watt Fresnel Spotlight 6" Lens
4- 44N8TVG 1000 /1500-Watt Fresnel Spotlight 8" Lens
1- 44N3TVB
1- 1365PTVG 500 /750-Watt Pattern Projector
1- 1365TVG /Iris 500 /750 -Watt Iris Klieglight
Klieglight
Accessories
Door for 44N62TVG
Door for 44N8TVG
-1097 Set for 16 Patterns for Projector
5-1078X 18" Diffuser Frame
2 -1421 Caster Stand 25" Base
4 -111TV Pantograph Hanger
4- 10E955G Extension Cable 10'
2- 25E955G Extension Cable 25'
5 -1080A 4 -Way Barn
2 -1081A 4 -Way Barn
1
Wiring
6- 619G/10/5 Connector Strips
4- 2433G/2 Wall Outlet Boxes
10' Long With
5
20 A Pigtail Outlets.
Control
3- 2500-Watt
9- 2500-Watt
Dimmers & Space For 3 Future Additions
Nondims
-100 A 3-Pole Main Breaker
1- Cross-Connecting System for 34 20 A Circuits Using Either Rotolector Or Patching
Devices
1
NOTE: Fixtures not shown are on portable floor stands and on camera.
Courtesy Klieg! Bros. Lighting
lighting plan for studio.
16
TELEVISION BROADCASTING CAMERA CHAINS
80
800
Spot Position 20' Throw
60
600
>ô
3
a0
ú
20
600
200
Flood
Position 20' Throw
0
5
10
2
Distance From Mechanical Axis In Feet
Distance From Mechanical Axis in Feet
(A) Type 44N12.
200
aoo
150
300
60' Throw
C
Ú
100
3
zoo
8
50
100
Flood Positi
5
30' Throw
10
0
Distance From Mechanical Axis In Feet
5
10
Distance From Mechanical Axis in Feet
(B) Type 44N16.
40
200
Flood Position 20' Throw
Spot Position 20' Throw
30
150
C
5
ú20
100
V
li
so
10
5
10
5
0
Distance From Mechanical Axis in Feet
Distance From Mechanical Axis in Feet
(C) Type 44N08.
Courtesy Klieg! Bros. Lighting
Fig.
1
-2. Photometric data for representative luminaires.
STUDIO LIGHTING
17
When comparing the lamps shown by Tables 1 -3 through 1 -5 with the
lamps of Fig. 1 -2, it is necessary to note the wider angle of throw for the
quartz luminaires. For example, the 2000 -watt incandenscent lamp of Fig.
1 -2A provides essentially 60 foot-candles (flood position) over a spread
of about 6 feet at a distance of 20 feet. The "standard" quartz luminaire
of Table 1 -3 (2000 watts, 12 -inch lens) has 60 foot-candles over a spread
of 17 feet (in flood position) at a distance of 20 feet (an equivalent of
almost three incandescent lamps for the same coverage). The equivalent
luminaire in the high -efficiency design (Table 1 -4) gives 225 foot -candles
over a spread of 17 feet at a distance of 20 feet.
Table -4. Performance of High- Efficiency "XKE" Fresnel
1
Performance at 20 Feet
Unit
Dimensions
Minimum Spot Maximum Flood
Catalog
Number
Lens
Wattage
3507
3507
3525
3525
3527
63/e"
63/e"
400
650
1000
2000
2000
8"
8"
12"
Hours
Spread
Life
FC
2000
4'
100
100
150
150
150
4'
2'
2'
300
300
500
600
1'6"
Spread
20'
20'
17'
17'
17'
Length
FC
Diameter
12"
12"
25
50
100
175
225
81/4"
81/4"
13"
13"
18"
131/4"
131/4"
18"
Courtesy Klieg! Bros. Lighting
Table 1 -5 shows that two lamps are available for the same luminaire:
One is a 30 -hour lamp that has an additional 40 foot -candles, compared
to the 150 -hour lamp, under the condition shown. This is normal in all
luminaires; shorter -life lamps such as a 110 -volt bulb on a 120 -volt circuit
give higher light output at a reduced life expectancy.
Table
1
-5.
Performance of Follow Spots
Catalog
Number
Wattage
Hours
Life
1393
1393
1000
1000
32
150
Performance at 75 Feet
Unit Dimensions
Spread
FC
Length
Diameter
6'
6'
180
140
46"
46"
16"
16"
Courtesy Klieg! Bros. Lighting
A 40' X 60' studio (net production area of 1664 square feet) lighted
for color with incandescent lamps would require 1664 X 90, or 150 kW
of power (Table 1 -1). Fig. 1 -3 illustrates a typical quartz studio-lighting
package for a 40' X 60' studio. In this case, a total of 72.2 kW is required.
Additional lighting for the cyclorama brings the total to about 100 kW.
Note that these power inputs are close to those computed from the data
of Table 1 -1.
TELEVISION BROADCASTING CAMERA CHAINS
18
r
Fig.
1
-3.
Quartz -iodine
STUDIO LIGHTING
19
40'
I
1
I
I
I
I
I
'
í
i
180-20 A Wi es
20-50 A Wi es
i
I
rb
20'
SAF
Patch
B
oard
SCR Ba nk
14'
I.
11
4
Wall
Outlets
Pipe or
Track
12/8
&&3 Extra
Flexible Cable
12/12 Extra
2406-G
Flexible Cable
Terminal Boxes
Cyc Track
Elevation
I
6
3' 9"
3' 9"
3' 9"
3' 9"
I
l
b
18
6 No.
12
Wires
Wires
For Control
I
Idb
30 -60 A Wires
18 No.
r
6"
l
d
dIbI
16'
Typical Connector Strip
Legend
3451,16' Scoop
3507, 6" Fresnel
3525D, 8" Fresnel
CD
1357,
6" Klieglight (Ellipsoidal)
3500FC,
i
Striplight
20 A
Connector
50 A
Connector
Cyc Track
Note: See equipment list on page 20.
Courtesy Klieg) Bros. Lighting
lighting plan for studio.
TELEVISION BROADCASTING CAMERA CHAINS
20
EQUIPMENT LIST
Quartz- Iodine Lighted 40' X 60' Color Studio (Package No. 27CQ)
For 100%
Fixtures & Accessories
Control Center
27 -No. 3451 1000 W Quartz Scoop
1- Composite
one scene, two subscene
lighting preset system containing: 108
-No.
6 -No.
36 -No.
3 -No.
44N3TVB 150 W Cameralight
3507 650 W Quartz Fresnel
3525D 1000 W Quartz Fresnel
1357P/6W 1000 W Quartz Pattern
Klieg
1000 W Quartz Iris
2 -No. 1357/6/1
2
20- ampere and 16 50- ampere counter-
plugs, 75 Automatic
Cold- Patching 20- ampere female jacks
with associated circuit breakers and 15
Boardlight, 8 7000
50- ampere jacks.
W SCR dimmers, 7 7000 W plug -in non -
weighted male
1
Klieg
4 -No. 1106A 4 -Way Barn Door for 650
W Fresnel
18 -No. 1081A 4 -Way Barn Door for 1000
dims which permit future insertion of
dimmers,
300 -ampere 3 -pole main
breaker, a preset section with 15 pots
with selector switches and 2 submasters and lock and key switch
1
W Fresnel
14 -No. 585A 16 -Inch Color
Frame
/Diffuser
for Scoop
-No.
1097 Sets of 16 Patterns for
Kliegs
2 -No. 1421 Castered Floor Stands
9 -No. 11 TV Pantograph Hangers
9 -No. 10E955G 10 -foot Extension Cables
for Pantographs
2-No. 25E955G 25 -foot Extension Cables
for Floor Stands
3
Lamp Package
27 -No. Q1000T3/4 Quartz Lamps
for
Scoops
-No.
2
1
150G161/2/3DC Lamps for
Camera lights
6 -No. FAD -650 Quartz Lamps
for
Fresnels
36-No. DXW -1000 Quartz Lamps for
Fresnels
Wiring Devices
10 -No.
16
5
619G/16/7 Connector Strips (each
feet long and wired with 2 double
and 3 single 3 -foot pigtail outlets on
five 20-ampere circuits)
10 -No.
6190/16/6/1X Connector Strips
with 2
single 20-ampere 3 -foot
(Each 16 -feet long and wired
double and 2
pigtail outlets and one 50- ampere out-
1
1
2406G/1OX Ceiling Terminal Boxes
100 -foot Multiconductor Drop
Cable
-No. 6/3 100-foot Multiconductor Drop
-No. 12/8
1
1000 W Lamps
Cyc Package
(Sufficient to light an "L" shaped cyc covcurve, and 1/2
ering one 40 -foot wall,
of the 60 -foot wall)
8 -No. 3500FC 7 -foot Striplights With 4
1
Reflectors and Glass Filters
3500AFC 31/2-foot Striplights With
2 Reflectors (For Curve)
-No.
10-No. 24060/10 Ceiling Terminal Boxes
10 -No.
1000T6Q /RCL/
for Kliegs
3
let)
-No.
2
-No. 453TVG/3 Adapters
50- ampere
(To
convert
a
outlet to three 20- ampere
outlets)
Cable
1
-No. 12/12
100 -foot Multiconductor
Drop Cable
6 -No.
24330/3/1X Wall Outlet Boxes
with three 20- ampere and one
50- ampere pigtail outlets)
(Each
Courtesy Kliegl Bros. Lighting
Fig.
1
-3. Quartz- iodine lighting plan for studio. (Cont'd.)
STUDIO LIGHTING
21
There are three possible ways of lighting for color television:2
The 100 -percent use of standard incandescent spotlights and floodlights. This has been the general practice, but, although it produces
satisfactory results, it has a high cost because of greatly increased
power and air -conditioning requirements.
2. The 100 -percent use of quartz-iodine spotlights and floodlights. This
method requires a minimum increase in power and air-conditioning
expenditures.
3. A mixture of incandescent and quartz- iodine lighting. Here the increase in power and air -conditioning requirements is somewhere between the foregoing extremes.
1.
Before considering the relative merits of these three lighting methods,
review two basic requirements of the color system. First, the lighting
level: This is 250 foot- candles plus a "holey factor." In order to understand this factor, it is essential to recognize that, in color, shadows are
more critical than in black and white. If the contrast range is not carefully controlled, shadows become an unacceptable dark color, and may
also cause video noise. Therefore, it is desirable to fill in these holes
(shadows) by using additional luminaires. These may be either Fresnellens spotlights or scoop floodlights as the situation may dictate. Since the
Fresnel in the flood position and the scoop both deliver a broad and softedged light beam, their coverage not only lightens the shadow (fills the
hole) but also overlaps the main object being lighted. With several such
"holey" luminaires being employed, it is possible to have the total overlap
add 75 to 100 foot- candles to the overall lighting level. This results, then,
in a level from 300 to 350 foot- candles of total illumination on the subject.
The second requirement is the color temperature of the light. By going
to 100 -percent quartz lighting, one has the opportunity to obtain an entire
family of lamps and luminaires in either 3000 or 3200 Kelvin temperature. By and large, the TV industry has shown a preference for 3200 -K
lamps for color work. Telecasters may now obtain scoops, Fresnels, pattern and iris projectors-all with the same color temperature. When this
is done, cameramen may move from one set to another without worrying
about the light mixtures, since they all will have the same color.
A direct comparison of light, life, and temperatures between standard
incandescent and quartz luminaires may be helpful at this point; see Table
1 -6. The first observation from this table may very well be that the quartz
lamp, although it may produce two to four times the light of its standard
2Much of the information given here was first presented at the Symposium
on Theatre- Television Lighting sponsored by the illuminating Engineering
Society (IES) , at Chicago, Ill., May 1966. It appears here by permission of
Kliegl Bros. Lighting.
TELEVISION BROADCASTING CAMERA CHAINS
22
Table
-6. Incandescent and Quartz- Iodine Comparison
1
Incandescent
Scoop
1000
500
200
Unchanged
With Age
Decreases
Unchanged
With Age
Decreases
With Age
Temperature (Color)
Decreases
With Age
(to Orange)
Quartz -Iodine
Fresnel
Fresnel
Scoop
Decreases
Life (Hours)
Light (Lumen Output)
Quartz- Iodine Incandescent
150
With Age
With Age
(to Orange)
Unchanged
With Age
Unchanged
With Age
incandescent /counterpart, has a shorter life. This may be true if the absolute term "life" is used, but it is not so if the limited term "useful life"
TV studios, in order to avoid degradation of color
and light output (see Fig. 1 -4) as lamps age, make it standard practice
to change lamps at 50 percent of the anticipated life. This practice gives
quartz lamps a longer useful life, since such lamps maintain color as well
as light output throughout their life. Furthermore, the production staff is
relieved of the burden of compensation for color degradation when the
lighting is 100 percent quartz.
In going to color with all incandescent lighting fixtures, approximately
150 percent is added to the power and air -conditioning load. Technically
speaking, a 100 -percent increase is sufficient, but the "holey factor" adds an
additional 50 percent. The luminaire complement would roughly double,
with some 5000 -watt Fresnels added to the 1000 to 2000 -watt Fresnels in
use for monochrone (only) . This method has been producing excellent
color pictures for networks and many independent stations. However, unless the station already has a large luminaire complement and has installed
the increased power and air conditioning in anticipation of the move to
color, this is the most costly way of lighting for color.
By providing 100 percent quartz- iodine fixtures in place of the incandescent fixtures-and by adding 50 percent more luminaires for the
is employed. Many
120
100
o
1000 -Watt
500 Watt
32008 Quartz -Iodine
Quartz- Iodine
go
1000 -Watt
60
500-Watt
Standard Spotlight
_Standard
Spotlight
7I
d(xl
I
800
1000
1200
1400
1600
Hours of Operation
Fig.
1
-4. Lumen -output
fall -off curves for typical lamps.
1800
2000
STUDIO LIGHTING
23
"holey factor"-one may go to color with only a 50 percent increase in
power and air conditioning. Not only does this method afford production
of color TV pictures more inexpensively, it also makes it a far easier
process for the production staff.
If existing equipment is reused rather than replaced, the method employed by many telecasters who convert to color is to use existing units
for fill and background lighting and to add quartz units for key and back
lighting. Providing for 75 percent quartz enables the station to meet color
requirements. This adds 75 percent to the power and air -conditioning costs.
Incidentally, this method produces excellent results, although it requires
color balancing and mixing, and is subject to some of the color degradation
exhibited by the old incandescent method.
1
-2.
AIR
CONDITIONING
Air conditioning in the control room and studios is mandatory for most
sections of the United States if optimum equipment performance is to be
expected. Although the effect of temperature in individual components
may be small as the temperature increases, the accumulative effects over
the average operating day can become quite noticeable. Such troubles as
poor focus, drifting linearity, waveform instability, and loss of resolution
often can be attributed to excessive temperature changes occurring after
adjustment and alignment of camera chains.
It is a fundamental truth that if equipment areas and studio areas are
maintained within reasonably comfortable temperatures for personnel,
rack equipment and components within camera and monitor housings will
remain within optimum operating temperatures. If the room temperature
is held no higher than 80 °F and the humidity remains within the 40 to 45
percent range, both personnel and equipment should operate efficiently.
Obviously, the required air -conditioning capacity depends on the size
of control rooms and studios, average seasonal temperature of the location,
and the amount of equipment and lighting installed. A competent air conditioning engineer as close to the community as possible should be
consulted. Usually, it is necessary to provide sound- isolation baffles in ducts
and outlets for the air stream to prevent sound leakage from control room
to studio and between studios, and to prevent whistles or other noise
caused by the forced air movement. This is a greater problem in television
than in aural broadcasting because TV audio techniques require greater
microphone -to- performer distances than the more intimate techniques used
in radio broadcasting.
Table 1 -7 gives the "rule -of- thumb" air -conditioning requirements for
the conditions stated, and assuming a 40' x 60' color studio. Note that a
diversity factor of 60 percent normally is used for the design of the studio
air -conditioning system. This would be 60 watts per square foot for incandescent lighting, 36 watts per square foot for quartz lighting, and about
TELEVISION BROADCASTING CAMERA CHAINS
Table
1
-7. Comparison of Lighting Design Parameters
Standard
Incandescent
Quartz
Mixture
250-350
250-350
250 -350
180
108
126
Watts per sq. ft. (NPA)*
100
60
75
Air Conditioning Heat Load ** (kw)
108
62.4
31
17.1
75.6
21.6
Lighting Levels (Foot -Candles)
Power (Kilowatts)
(Tons)
*NPA-Net Production Area.
* *Calculated on the basis of 60 percent of input power since the air conditioning is designed
on a continuous 12 -hour or longer basis, whereas the lighting is sporadic with peak lighting
(heat) loads having only a 30- minute to -hour duration.
1
45 watts per square foot for a studio in which quartz and incandescent
lamps are mixed.
1
-3. POWER REQUIREMENTS FOR LIGHTING
In figuring the power required for studio lighting, a 100 -percent utlization factor normally is used. The minimum diversity factor ever to consider
is 80 percent, but the 100 -percent factor is recommended. Thus, if the
lighting system provides 100 watts /square foot for a net production area
of 1000 square feet, the feeder requirement is 100 kW plus any additional
power required for lighting control, rear -screen projectors, cameras and
camera pedestals, etc. In general, however, the lighting-load feeders are
separate from those for other studio equipment.
It is the usual practice to order a three -wire service with a grounded
neutral. This is equivalent to two separate services, as indicated by Fig.
1 -5. This arrangement would be adequate to handle all lights in the average installation. An extra service for office equipment, air conditioning,
heating units, and the like is required.
I20 V
120V--
Main Fuse
Main Fuse
Bank
Bank A
Lighting -Load
B
Feeders
To
To
Individual Disconnects
Individual Disconnects
Fig.
1
-5. Simple power arrangement for lighting.
STUDIO LIGHTING
25
The simple arrangement shown in Fig. 1 -5 provides two individual
breaker panels, A and B. Each light fixture usually is marked on the housing in large letters such as A -6, B -12, etc. This designates the panel and
breaker involved for the individual luminaire. The individual breakers
usually are of 20- ampere capacity, with some 50- ampere breakers necessary ( for color) as specified in Fig. 1 -3.
1
-4. THE SUSPENSION SYSTEM AND POWER DISTRIBUTION
The most common system for providing current to suspended lights is
the connector -strip method illustrated in Fig. 1 -6. According to IES and
SMPTE recommendations, ceiling (grid) outlets should be spaced evenly
over the entire studio ceiling. The most convenient and least costly method
involves the use of factory assembled and wired, UL- approved connector
strips. These strips provide safe and convenient facilities for connecting
spotlights, floodlights, etc., where units are hung from an overhead track
or pipe batten. The particular strip illustrated in Fig. 1 -6 has the following
specifications: the front cover is removable for access to terminals; standard
finish is flat black (circuit designations, stenciled on the front in 2 -inch
letters, sometimes are used); each section is furnished with proper hangers;
strips are 2" X 4", 20 -gauge steel; strips are furnished with 5 pigtails,
each 3 feet long.
The sheet -steel wireway (with pigtail connectors) is mounted onto the
pipe batten as in Fig. 1 -6B. The pigtails are evenly spaced at 3' -9"
intervals along the connector strip. Each light is clamped onto the pipe
by means of a "C" clamp (Fig. 1 -7) , and the cord on the lighting unit is
plugged into the pigtail connector. A terminal box provided at one end
of the strip carries multiple- conductor flexible cable to a square duct
which, in turn, contains conductors that carry the connection back to the
lighting control center.
Another type of overhead light suspension is shown in Fig. 1 -8. Both
the fixtures and the individual cross tracks may be moved at will. Individual
cross tracks travel on master support tracks, and the fixtures travel on the
cross tracks. The ability to move individual fixtures and /or individual
cross tracks gives full location flexibility to each lighting fixture. All movement of fixtures and track may be accomplished from the floor with the
use of a pole. All fixtures have the same tilt and swing as when attached to
the pipe batten with a standard pipe clamp.
In addition to overhead lighting, some provision normally is made for
floor units. It is recommended that duplex or triple wall outlets be provided about every 30 running feet of wall. For monochrome, 20- ampere
outlets are used with 50- ampere outlets provided in at least one position.
For color, all outlets are usually of 50- ampere capacity. Such outlets provide power for floor -stand lights (such as low -level front lights) , rear screen projectors, follow spots, special- effect machines, etc.
26
TELEVISION BROADCASTING CAMERA CHAINS
Feed
Cable
Cable
Clamp
Pipe-Batten
Clamp
Hanging
Cable
Connector
Strip
Pipe -Batten
Clamp
Pigtail
Outlet
Courtesy Kliegl Bros. Lighting
(A) Typical unit.
\Ceiling
Duct
Connector Strip
Terminal
Pipe Batten
Box
Terminal Strip
Connect
at Duct
Cable Clamp
20A
50 A
20 A
20A
Grounded Pigtail Outlets
20A
Special Multiconductor Cable
Containing 3 No. 6 and
8 No. 12 Wires
(B) Power distribution.
Fig.
1
-6. Lighting connector strip.
illustrates one type of wall box that is available with either 2,
pigtails 18 inches in length. Such pigtails permit the use of the
standard heavy-duty lighting plugs required for television and theater
luminaires.
Fig.
1 -9
3, or 4
1
-5. LIGHTING CONTROL
A single circuit is brought to the lighting control center from each
lighting unit. The "control center" for a small studio may be as simple as
STUDIO LIGHTING
Fig.
1
27
-7. Double C clamp.
Courtesy Kliegl Bros. Lighting
a bank of circuit breakers, as described for Fig.
1 -5. In a larger studio,
breakers may be grouped and then mastered. Each load circuit is represented either by a cord and plug (similar to a telephone switchboard) or
by a rotary switch. Individual loads then can be grouped into any one of
12 to 24 (or more) dimmer or nondim controls.
Fig. 1 -10A illustrates the Kliegl Saf -Patch panel, in which "make" or
"break" of a "hot" circuit is prevented automatically. The system consists
of load plugs and female receptacles. Each receptacle is controlled by an
individual magnetic -type circuit breaker. These breakers serve as on -off
controls for their associated receptacles and may be used for individual
changes without affecting other lights. Individual load plugs, one for each
Courtesy Kliegl Bros. Lighting
Fig.
1
-8. Double -track mobile system for overhead lights.
28
TELEVISION BROADCASTING CAMERA CHAINS
lighting circuit, are equipped with cable clamps that eliminate strain on
internal connections. The special design of the load plugs makes it impossible for them to be inserted into a receptacle when the circuit breakers
are in the on position (Fig. 1 -10B) . When the plugs are removed from
the receptacles, they automatically trip the breakers to the off position
before the plugs are completely withdrawn (Fig. 1 -10C) eliminating any
possibility of a "hot" break. As an additional safety feature, load plugs of
different capacities cannot be interchanged and can be connected only
to properly rated female receptacles. The plugs and patch cords are
properly counterweighted for automatic restoring when not in use.
Fig.
1
-9. Typical wall box.
Courtesy Klieg! Bros. Lighting
Individual standard Saf-Patch panels are furnished with six female receptacles, although panels of three or four can be furnished upon request.
Groups of these panels form a Saf-Patch Klieg -Board (Fig. 1 -11). The female receptacles can be of 20- or 42- ampere size and freely intermingled.
Groups of female receptacles generally are wired together and fed by a
dimming or nondim circuit.
Manually operated systems may be recommended for more simplified
dimming applications where requirements are moderate. In this system,
the dimming units and their controls are mechanically manipulated, and
29
STUDIO LIGHTING
Switch Must Be Off Before
Plug Can Be Inserted
Switch Must Be Off Before
Plug Can Be Removed
(A) Panel unit.
(B) Make operation.
(C) Break operation.
Courtesy Klieg! Bros. Lighting
Fig.
1
-10. Klieg! Saf -Patch panel.
Saf -Patch
Panel
Retractable
Patch Cords!
Courtesy Klieg! Bros. Lighting
Fig.
1
-11. Sal-Patch lighting panel.
30
TELEVISION BROADCASTING CAMERA CHAINS
it should be remembered, therefore, that ease and flexibility of control will
not duplicate those of a corresponding remote system. A mechanical system may include various forms of interlocking so as to provide for control
Courtesy Kliegl Bros. Lighting
Fig.
1
-12. Autotransformer dimmer board.
of a few or all dimmers at any one time by means of submaster levers and
a grand master lever. The autotransformer dimmer (Fig. 1 -12) is the
major component of a manually operated system. The diagram of Fig.
1 -13A illustrates a typical manually operated autotransformer system. As
you will note, a cross -connect circuit-selection panel for the interconnection
Secondary
Secondary
Lighting Circuits
Lighting Circuits
Cross
11
ttt
Connect
Connect
Circuit -
Circuit -
Selection
Selection
Sy tern
System
Manually
Ope ated
Remote
Cant oiled
Dimmers
Dimmers
Cross
Control
Console or
Programmer
1
1
Power Feeders
Power Feeders
(A) Manual operation.
Fig.
1
(B) Remote control.
-13. Two basic types of lighting control.
Preset
Panel
STUDIO LIGHTING
31
of lighting loads and dimmers is located between the dimmer bank and
the lighting fixtures.
The use of remote operation is mandatory under certain circumstances:
where presetting of scenes is desired; where two or more control stations
are called for; where remote location of the control console is required.
With a remote system, finger -tip controls are assembled in a control console. This control center then may be located away from the dimming bank
in any desired location. At the same time, the power units can be located
conveniently near the lighting fixtures so that feeder runs and other wiring
may be kept to a minimum.
The diagram of Fig. 1 -13B illustrates a typical remote -control system.
As you will note, the control console and preset panel (if required) may
be located at a good vantage point away from the dimmer bank. Located
between the dimmer bank and the lighting fixtures is a cross -connect
circuit -selection panel for the interconnecting of lighting loads and
dimmers.
A cross -connect circuit-selection system that eliminates the use of patch
cords and plugs is the Kliegl Rotolector. This device is a rotary circuit
selector used to cross -connect lighting load circuits to dimming or nondim
feeder lines. Each branch lighting circuit is wired directly to its own
Rotolector, which, in turn, is connected to the various dimmer and nondim
feeders. For example, a 12 -point Rotolector can be fed by 10 dim and 2
nondim sources, or any other combination that totals 12. To patch the
load to any one of the feeders, the operation is as follows:
1.
2.
3.
The knob of the Rotolector is withdrawn. This automatically trips
the associated circuit breaker before the load contact is broken.
The knob is rotated to the selected position, as indicated by the numbers on the dial, and then pushed in until contact is made.
When contact is made, the circuit breaker can be turned on to energize the circuit. "Hot" connections and arcing are eliminated automatically.
Each 61/2" X 61/2 unit can be furnished with either 12 or 24 positions.
Units are available to handle either 20- or 50- ampere lighting loads.
In manually operated autotransformer -dimmer Klieg- Boards, each auto transformer dimmer is controlled by its own operating handle, but all units
may be coupled to submaster and grand master levers if desired. Control
levers are color coded, labelled, and grouped for quick identification and
use. Each dimmer is protected by a magnetic -type circuit breaker that can
serve also as an on -off switch. Additional breakers are provided to protect
and control secondary circuits. The board also can include circuit- breaker
switches for control of nondim circuits, as well as remote-control switches
and transfer switches. Manually operated autotransformer dimmers are
used when there are strict budgetary restrictions and there is no need or
desire for presetting of intensities or control from a remote location.
TELEVISION BROADCASTING CAMERA CHAINS
32
Fig.
1
-14. SCR dimmer module.
Courtesy Klieg! Bros. Lighting
The majority of modern remote -control dimming installations utilize
silicon controlled rectifier (SCR) dimmers. The wide -range capacity of
these dimmers makes them especially suitable for auditorium lighting and
other applications where large lighting loads are involved. They are small
in size and require no outside cooling. Dimmer banks can be located remotely or as a part of a console and cross-connect system.
Fig. 1 -14 shows an SCR dimmer module. This module is interchangeable with SCR nondim units, and can be safely inserted or removed from
the board, while under load, without tripping any protective devices. The
response to control is instantaneous, and no warm-up period is required.
The units are designed for continuous operation at ambient temperatures
up to 40 °C. When loads in excess of the rated capacity of the dimmer are
Circuit
Breaker
Line
-o
Back-to-Back Silicon
Controlled Rectifiers
Fast-Acting
Fuse
o _
Transformer
eksy
Output
lQQWJ
Control
Fuse
Control
Input'
Firing
Circuit
r
Neutral
Control Transformer
Magnetic Amplifier Circuit
The control input is completely isolated from the power input.
Courtesy Klieg! Bros. Lighting
Fig.
1
-15. Circuit of SCR dimmer.
STUDIO LIGHTING
33
hot patched into the unit, it operates at reduced voltage until the overload
is removed. Short -circuit protection is provided by means of silver -sand
fuses that operate within 1/60 second to protect the rectifiers.
Fig. 1 -15 is a schematic diagram of the SCR dimmer module of Fig.
1 -14. The back -to -back SCR's are in series with the lighting load. The
firing- circuit diodes for the SCR gates are activated from a low- voltage control input (never more than 28 volts) through a magnetic amplifier.
`
MUMMMUMM --
MM61=MMMMMMM_
Ñ
O
1
! / r rl
UMMM
NNN__,
IIMM=N
/=UUMMUMUM=MM6111/UM_
MqMMMUU=NUMMUM=
SMM=MMMIiIIIMMI11.\61ME=Mrift
\'MMMM11II11M=M/IM\
M1II
U11M61MMM\MMMI61\1
OEY
MM
V=111
N1iMM
MM
M11MI/11MMMM11M'/M
7MO\M'/EU MMM.\I
.11NNMMU111111.
qNNNNU011IINMN11.
itáMUO61MM61MMMMN
/M=MMM=M6161UMN61UMO
.11M=MUNII1111UN111=11OMNs
/61MMEMEUUUUNM61=UII
INMUMM61MMNM11MMMUMMMI
N=Ur11NNEIIMr:\NNIIOMNr\NM
MM 11\1=MMU1111MM=,
ÌMN11Im11M11MM61MI
Isalue
11MM611111M
I=NNI IM 11OMN1111UMU=MMOI IU.1NI
eu=le1111 ss1111M11s 1E1111,=N111111II111
11IM/NMN 11UUUOMII'1NMUNN111
tAl/II11
1161M
1MiU61MII4N1911MM
\UMNMIINOMUMUU11M
MM=11111MUM11111111111
1111,=UMM61U 14111UMMMI Ir,MM=UMMI11
U7NUBN, R IINON61UI1%MMUNMEI
7MM=UMUM=Mr
V11=MNN11N11MMUUMI
I'
\OOMMMMM11MMMUNMMM11M=I
IlUN11M11MNMM11N'
'IIMUUNNM=EMMNM11MU'
'11MMOMMMMUMMOMM'
NMUNOMMMMMME61M
NUMMNM61P"
'11UMEME11/-
NOMN
Potentiometer Setting
Percent Voltage 100%
Potentiometer Setting
Percent Voltage 67%
10
i11si61II.
/OU
.11M11U11
MUMNMMM11MMM=NM
IIMUMMU61NUU=111110
.IiMNNNMM=UNM=\.
ANMIINNMMEUNMMNNNIIs.
AMMMOMMEMMUNMM=MNM
AOMMMMMMMUMMMUM.
M=MMMU=UMMUIIU
IINNNMNOMMNMMMIIMMU1111
IINM=MMMMMM61NM=MUMMMNMM\
/MMUM11MM=MMMNMUMMNMM=1
.
/NNNMNNNMUNMMNNN11111111
iUNNNIIONMNIINIIMIIUUIINN=NII
IMMMU61i OMMMM611;11M61MM611,
5
MI
uus>tlst11qspI1111Aiipii¡
1111'/M=11=M /11MUM=I I'IMMNM
MI IIMMMNNU
II11=MN1111UUMMNMMUIINMMNII,
lIIN11MNNNMMMMU11MMN1111==N1
11111111111M11MMU=MMMOMM=11MI
\NM11MMMMMMMMr
INMNN11MMMMMMMN'
1NO61UU1
IIMUU
'IINUMN=MNMUMUMMUI
'1111NMMNMNNMMM/'
1MU1111U11N'
'11MNNNNNNM11NNNM1111Ur
I
I/N=MMNMI I11MMU1111
1I J=NM11UM'IMMMMI IMNUUMNN
'UMMM=61MMMUMM61M61MMMNM
1
I
INN11UMMM11M11UNONI
I11MNNIIIIIINMMNMIIMNOMIIN11N1
11MMMNMMUMMMMMNMMMMI
'MMUMMM61EM=MMUMMMF
\MINNOMMU11=MMOMOI
OMUMMUNNMNNMMNNNI
1ONNMNN11MN=N=N11U/
MMMUMNMMNr
'.111MOMMOM11."
11111MMM1161U'
NNUNIIMMMP'
Potentiometer Setting
Percent Voltage 10%
0
Potentiometer Setting
Percent Voltage 0%
OFF
Courtesy Klieg! Bros. Lighting
Fig.
1
-16. Output waveforms of
SCR
dimmer.
Note that this allows complete isolation of the control input from the
power input. Thus, the dimmer controls on the console are concerned only
with the isolated low control voltage, about 28 volts dc at 12 milliamperes.
Fig. 1 -16 shows the voltage waveforms at the output of the SCR dimmer
module for four different settings of the control potentiometer. At a set-
TELEVISION BROADCASTING CAMERA CHAINS
34
ting of 10, the output voltage is 100 percent, and the lights associated with
that control are at maximum intensity. At a setting of 5, the gate voltage
causes the SCR's to conduct later in the cycle, reducing the voltage output
to 67 percent. At a setting of zero, the firing point in each cycle is still
later, applying only 10 percent of full voltage. The control must be placed
in the off position to remove the output voltage completely.
100
80
20
0
Off
0
2
3
4
5
6
7
8
9
10
Potentiometer Settings
Courtesy Klieg! Bros. Lighting
Fig.
1
-17. Light output resulting from SCR -dimmer control.
Fig. 1 -17 gives the correlation between control setting and percentage
of light output. Note that, for example, a setting of 5, which applies 67
percent of full voltage to a lamp, results in approximately 34 percent of
maximum light output.
Models of these SCR dimmers are avaliable in 3, 6, 10, 12, and 15 kW
capacities. Nondim units are available in 3, 6, and 10 kW sizes. A built -in
device permits the nondim units to be set at any predetermined "on /off"
position.
EXERCISES
Q1 -1.
Give the nominal value of lighting, in watts /square foot, normally
planned for (A) monochrome and (B) color when incandescent
lighting is used.
STUDIO LIGHTING
Q1 -2.
Q1 -3.
Q1 -4.
Q1 -5.
35
Give the nominal value of lighting, in watts /square foot, normally
planned for (A) monochrome and (B) color when 100 -percent
quartz lighting is used.
You have the lamp of Fig. 1 -2B to be operated in the flood position.
How many foot -candles would you expect at a throw distance of
15 feet (A) on axis and (B) 10 feet off axis?
You have the same lamp as above, operated in the flood position.
How many foot- candles would you expect at a throw distance of 60
feet (A) on axis and (B) 5 feet off axis?
Approximately how many amperes does a 5000 -watt lamp draw at
a voltage of 115 volts?
CHAPTER
2
The System Concept
Fig. 2 -1 illustrates the wide variance in fundamental television camerachain systems. Because of the predominance of color telecasting, color systems will be emphasized in this training.
2 -1. NTSC AND FCC COLOR STANDARDS IN PRACTICAL FORM
The FCC standards are based on the work of the National Television
Systems Committee (NTSC) as originally filed with the FCC on July 21,
1953. We will often use the term "NTSC color" because the specific details
of the system are far more inclusive than those spelled out in the FCC
standards.
Since most colors can be duplicated by mixing correct amounts of three
properly selected primary colors, it follows that a color TV system can
be based on the transmission and reception of images in the three primary
colors. The first step is accomplished in the television camera. The camera
generates three different signals from the information contained in the
image of the scene. There are three general types of color cameras, as
follows:
1.
Three cameras are operated from a single set of controls so that the
view televised by each is identical. In front of each camera lens is
placed a red, green, or blue filter. While the view imaged by all
three cameras is identical, the light reaching the light- sensitive plate
of each camera contains only the components passed by its respective
filter. The three camera heads operate as one camera, and this camera
produces a signal corresponding to the image in each of the primary
colors. The brightness, or luminance, information is a function of the
combination, and therefore the three images must be accurately registered to obtain a reasonably sharp picture, whether reproduced in
monochrome or color.
36
THE SYSTEM CONCEPT
37
Sync
Generator
Viewfinder
Processing
Equipment
Camera Head
and
Power Supply
Interconnection Link
May Have One, Two, Three,
or Four Channels.
May or May Not Contain
All Setup Controls.
In Mobile Field Applications,
May or May Not Contain
RF
Output
May Be Up to 84- Conductor
Control
Cable, Single Triax Cable,
Panel
or
RF
Link (Field Application).
For Color,
Includes
Encoding Process.
Sending and Receiving Gear.
Fig. 2 -1. Components of camera -chain systems.
2. Other cameras use a similar system, except that four pickup tubes are
involved: three for the primary colors, and the fourth for luminance
only. In this type of camera, only the luminance channel must have
wide bandwidth; the three color channels can be relatively narrow band. With this type of camera, since the color channels carry very
little brightness information, the effect of misregistration of images
is slight on monochrome receivers.
3. A "convertible" system makes use of only two channels in the
camera. One tube is used for a full- bandwidth luminance signal,
and the second tube is used in the alternate red /blue channel. The
fields are sequenced mechanically through a rotating red and blue
filter wheel synchronized to the vertical rate of the main synchronizing generator. The green signal is obtained (before encoding)
through subtractive matrixing of the red /blue and luminance signals. Special processing is required to convert the sequential color
to the NTSC color signal.
In any of the basic types of camera mentioned above, we will be concerned with three signals, Y for luminance, and I and Q for chrominance.
Caution: You will find a rather common application of the letter M ( for
"monochrome") to designate luminance information. In this book, the
letter Y will be used for luminance, because in NTSC -FCC color specifications, the voltage of the composite color signal ( which obviously contains
both luminance information and chrominance information) is designated
by the symbol EM.'
'It is imperative that the reader have a color training background equivalent
to that of Chapter 2 in Television Broadcasting: Equipment, Systems, and
Operating Fundamentals, by Harold E. Ennes (Indianapolis: Howard W. Sams
& Co., Inc., 1971).
38
TELEVISION BROADCASTING CAMERA CHAINS
The color picture signal has the following composition:
EM= E,r' +[EQ sin (cot +33 °)
+EÌ cos (cot +33 °))
(Eq. 2-1.)
where,
EQ'=0.41 (ER' -EY') +0.48 (ER-E )
EÌ = -0.27 (ER' E'T ) + 0.74 (ER' EY' )
=0.30 ER' +0.59 EG' +0.11 ER
-
E
( Eq.
-
2-2.)
(
Eq. 2-3.)
(
Eq. 2-4.)
The phase reference in Equation 2 -1 is the phase of the color burst + 180 °.
The burst corresponds to amplitude modulation of a continuous sine wave.
Equations 2 -1 through 2 -4 have been developed in your previous training. Equation 2 -5, which is in terms of the color- difference signals only (as
demodulated by "narrow- band" color receivers), holds only for frequencies
below 500 kHz (the region in which both Q and I are double sideband) :
EM
=
E + { 1.14
L
(EB
1.78
- Er') sincot + (ER - E) coscot] }
(
Eq. 2 -5.)
This equation is as given in the NTSC -FCC standards; note that you can
more clearly visualize this as:
EM
= Ey + 0.493
(ER
- Ey) sincot + 0.877 (ER - Ey) coscot
The I and Q channel bandwidths shown by Fig.
FCC as follows:
2 -2
are specified by the
I- Channel Bandwidth: Less than 2 dB down at 1.3 MHz
At least 20 dB down at 3.6 MHz
Q- Channel Bandwidth: Less than 2 db down at 400 kHz
Less than 6 dB down at 500 kHz
At least 6 db down at 600 kHz
Color Subcarrier
3.579545 MHz
455z
tH
2
Sound Carrier
Luminance Signal
El. cos
(cut +3301
/
EQ
1
/
sin(cot
2
Chrominance
Signal
N./
+3301/ /
//
\
/
\
I
1
\\
\
3
a
Video Frequency (MHz)
Fig. 2 -2. Frequency distribution of color signal.
THE SYSTEM CONCEPT
39
The prime signs in Equations 2 -1 through 2 -5 indicate gamma -corrected
signals. In the discussion that follows, we will assume all system signals to
be gamma corrected, and designate them as follows:
EY
= corrected luminance -signal voltage
= corrected red -signal voltage
EG = corrected green -signal voltage
ER = corrected blue -signal voltage
ER
Caution: The method of gamma correction is not presently fixed by the
FCC. In the three- camera color system, ER, EG, and ER are all gamma corrected to make up the composite color signal. In the four -camera system,
the luminance channel is gamma corrected, but the individual color channels may or may not be corrected in the same way. In this section, we
assume that matrixing, encoding, and all following functions will be performed on properly corrected signals from the camera.
Fig. 2 -3 is also a review, but we want to take a slightly different approach so that your thinking remains flexible. The luminance part of the
complementary colors (yellow, magenta, and cyan) is formed as follows:
Yellow:
EY-EB= 1.00-0.11 = 0.89
Magenta:
EY-EG= 1.00-0.59=0.41
Cyan:
EY
-ER= 1.00 - 0.30 = 0.70
Be sure you can visualize this from your previous training. Thus, the
luminance (only) part of the signal is as shown in Fig. 2 -3A for 100 -percent bars.
NOTE: Section 2 -2 considers 75- percent color bars as normally used.
However, the 100 -percent bars are used for certain test procedures.
The values of I and Q for the primary colors and their complements
are reviewed in Table 2 -1. For example, the values for yellow are I = 0.32
and Q = -0.31.
EYELLOW
=\%I2 +Q2
= \/(0.32)2+ (- 0.31)2
= \/0.1024 + 0.0961
= \/0.1985
= 0.446, or simply 0.45
as shown in Fig. 2 -3B.
40
TELEVISION BROADCASTING CAMERA CHAINS
°'
d
cd
g
c
>
c.T,
i
c
1.00
--.1
0.89
0.70
(A) Luminance only.
0.59
0.41
0.30
1
0.11
0.00
3
°
c.
c
zó
c°
c.+
0. 63
0 59
1
0.59
I
l
0.63
m
1
1
0.45
0.45
(B) Chrominance only.
-0.45
-0.59
0.63
-0.59
77
0, 45
c
v
E
¢
m
m
140
118
120100
100
100
89
80
70
59
60
Color Sync
40
Burst
55
41
45
30
20
20
20
11
0
-20
-40
-18
20
40
Sync Pulse
_`Notiz
(C) Composite levels.
20
-33
-33
Fig. 2 -3. Composition of 100 -percent color bars.
40
41
THE SYSTEM CONCEPT
You should be able to develop the remaining chroma amplitudes of Fig.
2 -3B by the same reasoning. Remember that the subcarrier goes to zero
for white and black (and all shades of gray) .
Fig. 2 -3C represents the composite signal ( for maximum saturated
chroma) on the IRE scale. (NOTE: The bar between green and red may
be called "purple" or "magenta "; both are the same). Fig. 2 -3C is for
100 -percent color bars without inserted blanking pedestal.
The amplitude of the chroma signal interprets the degree of saturation
of the particular hue. The fact that the yellow signal is 0.45 times the
amplitude of the white luminance signal tells you this color is a fully
saturated yellow; it contains a maximum unit of red and a maximum unit
of green, with zero blue.
Table 2 -1. Color System Relationships for Primaries and Complements
Transmitted
Color
Green
Yellow
Red
EG
ER
En
ET
1
0
0
0.59
1
0
0.89
0
0.3
1
0
1
G
-Y
R
0.41
-0.59
0.11
-0.3
-Y
B
-Y
Q
I
-0.59 -0.525
-0.28
0.11
-0.89 -0.31
+0.32
0.7
-0.3
+0.60
+0.21
I
Magenta
0
1
1
0.41
-0.41
0.59
0.59
+0.525
+0.28
Blue
0
0
1
0.11
-0.11
-0.11
0.89
+0.31
-0.32
1
0.7
0.3
-0.7
0.3
Cyan
1
0
-0.21
-0.60
Hue is determined by the phase angle of the signal with respect to a
specified reference. (This is also a review, but, again, we want to take a
slightly different tack so that you can "see" this system in its many facets.
In this case, "familiarity breeds contentment. ") See Fig. 2 -4A. Suppose we
have a carrier that starts with the phase of the R
vector, and we pass
this carrier through a 90° delay line to obtain a carrier in quadrature (in
phase with the B
vector) Now we can say that B
lags R
by
90 °, or that R
leads B
by 90 °. But recall that vectors "in action"
rotate counterclockwise, and the starting point (0° or 360 °) is generally
taken at the B
axis on the right.
Now study Fig. 2 -4B. It will be noted that when ER E, is transmitted
alone, it produces colors from bluish -red through white to bluish -green,
which are located along its axis. When transmitted alone, signal ER El
produces colors from purple through white to greenish -yellow. From the
color triangle, it can be seen that varying amounts of both signals, when
transmitted together, produce hue and varying saturations of any color located within the triangle. For example, Area 1 is enclosed by axes designated
(ER EY) and
(EB
EY) and represents the colors that can
be produced by the presence of these two signals.
-Y
-Y
-Y
-Y
-Y
.
-Y
-Y
-
-
-
-
-
-
42
TELEVISION BROADCASTING CAMERA CHAINS
Green
R-Y
180°
Greenish Yellow
`Advance
240°
Bluish
-1B
-Y)
B
-Y
Green
Red
Bluish
Red
Blue
/
Purple
0°
-1R YI
(A) R
360°
-Y and B -Y axes.
(B) Colors for R
-Y and B - Y.
Burst
360°
(C)
I and Q axes.
Fig. 2 -4. Color phase
(D) Colors for I and
Q.
angles.
You have learned previously that it is desirable to produce a "wideband"
signal for the orange -cyan color regions. If color detail is transmitted by
ER
Ey alone (EB
Ey being zero) , the colors reproduced will be bluish red and bluish -green instead of an orange -cyan mixture. So we set up a
as
new pair of axes by starting with a vector advanced 33° from R
shown in Fig. 2 -4C. This now becomes the "in-phase" vector (I) . The
quadrature component (Q) is advanced 33° from B Y. The reference
-
-
-Y
-
43
THE SYSTEM CONCEPT
- -
burst remains along the
(B Y) axis. The new color gamut is now as
shown on the color triangle of Fig. 2 -4D. The respective bandwidth pro portionments are as shown by Fig. 2 -2. So the wideband I signal lies along
the orange -cyan axis.
Now note that when R -Y was rotated to form the I axis, it was reduced in amplitude ( Fig. 2 -5A) For example, the red vector in terms of
R -Y and B -Y is shown in Fig. 2 -5B (see Table 2 -1). Since R -Y was
reduced by a factor of 0.877, the red vector as a result of I -Q modulation
is as shown in Fig. 2 -5C. This is to say that the original 0.7 of R -Y is
now 0.7 (0.877) , or 0.614, along the R -Y axis. The original
(B Y)
amplitude of -0.3 is now 0.3 (0.493) , or -0.14, along the
(B Y)
axis. This is the relationship you will see for the red vector on the station
vectorscope as a result of I -Q modulation. We will cover this in detail later.
From trigonometric tables (or your slides rule) , cos 33° = 0.839 and cos
57° = 0.544. If you use these values you arrive at the stated proportions of
R -Y and B -Y for I and Q as given in Fig. 2 -5A.
We have developed the chroma amplitudes for saturated primary and
complementary colors in terms of I and Q. Now let us review the development of the chrominance- signal phase angle.
The graphic method for development of a specific phase angle (yellow)
is shown in Fig. 2 -5D. From previous training (tabulated by Table 2 -1),
the I and Q components of yellow are I = 0.32 and Q = 0.31. The vector
sum is 0.45 at an angle leading the +I axis by 44.1 °. The angle, then, with
the NTSC zero axis (B Y) is 123° + 44.1°, or 167.1°.
To solve by trigonometric rather than graphic means:
.
- - -
-
-
-
cotangent O =
adjacent side
opposite side
0.32
-0.31
1.032
From trigonometric tables (or your slide rule) , cos 33° = 0.839 and cos
that, since the cotangent of the unknown angle is 1.032, you find the angle
whose cotangent is this value. Without worrying about plus or minus
values of the number, by visualizing the quadrant in which the I and Q
sum must fall, you know whether the angle leads or lags the I or Q axis.
Obviously, you can use any trigonometric function-sine, cosine, tangent,
or cotangent-depending on which is most convenient in your computations.
A good way to remember the signal polarities associated with various
colors is through the use of the color triangle. We have noted that an I,
Q, or color- difference signal may have a negative polarity for some colors,
and for other colors any one of these signals may have a positive polarity.
By studying the color triangles of Fig. 2 -6, you will find it easier to remember which colors produce an I, Q, or color -difference signal that is
negative, and which colors provide a positive signal.
On the color triangle of Fig. 2 -6A, the polarity of the I signal for each
color is given. The colors that fall to the right of the Q axis are repre-
44
TELEVISION BROADCASTING CAMERA CHAINS
RY
B
Y
0.877 IR -Y) cos 33° - 0.493 IB -Y) cos 57°
0.74 IR -Y) - 0.27 IB-Y)
I
0.877 IR -Y) cos 57°
Q
0.48IR -YI
.
.
0.493(13-Y) cos 33°
0.411B-YI
(A) Rotation of R -Y to form
I.
0.877 IR-Y)
R-Y
Red
0.7
0.614
-0
0.493
BY
3
(B) Red in terms of R -Y
and B
- Y.
IB-YI
-0.14
(C) Red obtained from I-Q
modulation.
(D) Development of
Yello
yellow vector.
Burst 180°
Fig. 2 -5. Color vectors.
45
THE SYSTEM CONCEPT
Q
Green
Axis
Yellow
I
________________________________________
______Q
4 +4
Axis
Orange
----4444+++++ + ++
-- - -'--4++ ++ +4 +4+4 +4
4444++
Q ++ ++
4444+++4 ++
Cyan
4+++
-I
Axis
+444u
}4++4
+4
+4++4+}
t+ ++++4+4
+
++}
+
Red
Magenta
+++444+
4+4 +++
Blue
(B)
(A) I signal.
Q
signal.
--IR-YI-----
4+4
+1R-YI w++
+:1B-Y1++
4+...
+4+.
+4
(C) I and
Q signals.
(D)
Color
(E) Reference chart.
-Y and B -Y signals.
R
R
-Y
B
-Y
I
Q
Green
-
-
-
-
Yellow
+
-
+
-
Red
+
-
+
+
Magenta
+
+
+
+
Blue
-
+
-
+
Cyan
-
+
-
-
Fig. 2 -6. Polarities of color signals.
46
TELEVISION BROADCASTING CAMERA CHAINS
sented by a positive I signal, and the colors to the left of the Q axis are
represented by a negative I signal. For instance, when blue, cyan, or green
is transmitted, the polarity of the I signal is negative. When magenta, red,
or yellow is transmitted, the I signal is positive in polarity. Fig. 2 -6B shows
the polarity of the Q signal for each color. As can be seen, the colors that
lie above the I axis are represented by a negative Q signal, and those
lying below the axis produce a positive Q signal. The polarity of the Q
signal is negative when cyan, green, or yellow (above the axis) is transmitted; when blue, magenta, or red (below the axis) is transmitted, the Q
signal is positive in polarity.
A composite drawing of the triangles in Figs. 2 -6A and 2 -6B is shown
in Fig. 2 -6C. Notice that the I and Q signals for colors which lie in the
upper left -hand section of the triangle are negative and that the signals
representing colors in the lower right-hand section are positive. Colors
lying in the other two sections produce I and Q signals that are opposite
in polarity. For instance, the Q signal is positive for blue, but the I signal
is negative.
The key for determining the correct polarity for each of these signals is
in knowing the location of the colors on the triangle and in remembering
the negative and positive areas shown in Figs. 2 -6A and 2 -6B. With this
knowledge, the polarity of each signal for any color can be determined
easily.
Fig. 2 -6D shows the polarities of the R
-Y
-Y
and B
signals for each
color on the color triangle. This drawing can be used to determine the
polarities of the R
and B
signals in the same manner that Fig.
2 -6C can be used to determine the polarities of the I and Q signals. Fig.
2 -6E lists the various signal polarities in tabular form.
The FCC states that:
"The [angles) of the subcarrier measured with respect to the burst
phase, when reproducing saturated primaries and their complements at 75
percent of full amplitude,2 shall be within ±10 °, and [the) amplitudes
shall be within ±*20 percent of the values specified.... The ratios of the
measured amplitudes of the subcarrier to the luminance signal for the
same saturated primaries and their complements shall fall between the
limits of 0.8 and 1.2 of the values specified for their ratios. Closer tolerances may prove to be practicable and desirable with advance in the art."
The relative tolerances specified are shown by Fig. 2 -7. Note, however,
that the FCC specifies "at 75 percent of full amplitude." A visual transmitter is never checked for FCC specifications at "full amplitude" for
chroma signals, because this results in overshoots of about 33 percent, as
you have found. But also note that the FCC specifies "saturated primaries
and their complements." Observe in Fig. 2 -7 that chroma amplitudes are
-Y
-Y
2Observe that color bars are saturated even when transmitted at 75 percent
of full amplitude. This is further developed in following Section 2 -2.
47
THE SYSTEM CONCEPT
ta
.),
o
$o ó ó c
n1
o
^,g bo
^1
.3O
o
hb'
o
o
S
IS71 o
162.10
16710
V
172,10
177.1°
Burst
180°
20
Ró
30
40
50
Ñ
60
70
00
357.1°
352.10
347.1 °
It0.10
33710
o
.y
o
0
o
`Is
FCC Tolerance Large Box
Operating Tolerance Small Box
Fig. 2 -7. Tolerances for color signals.
shown at their maximum saturated levels. We will cover the practical
details of how to use this information as we progress.
The FCC fixes the frequency response of the color TV transmitter as
follows: "For monochrome transmission only, the overall attenuation characteristics of the transmitter, measured in the antenna transmission line
after the vestigial sideband filter (if used) , shall not be greater than the
following amounts below the ideal demodulated curve. .
2 dB
at 0.5 MHz
2 dB
at 1.25MHz
3
dB at 2.0 MHz
6dBat3.0MHz
12 dB at 3.5
MHz
"The curve shall be substantially smooth between these specified points,
exclusive of the region from 0.75 to 1.25 MHz...."
For color transmission, the above standard applies except as modified by
the following: "A sine wave of 3.58 MHz introduced at those terminals of
48
TELEVISION BROADCASTING CAMERA CHAINS
the transmitter which are normally fed the composite color picture signal
shall produce a radiated signal having an amplitude (as measured with
a diode on the rf transmission line supplying power to the antenna) which
is down 6 -± 2 dB with respect to a signal produced by a sine wave of 200
kHz. In addition, between the modulating frequencies of 2.1 and 4.1
MHz, the amplitude of the radiated signal shall not vary by more than
±2 dB from its value at 3.58 MHz...."
Note here that the FCC refers to the response as measured by the "ideal
detector" curve. This is treated further in the exercises at the conclusion of
this chapter.
Envelope -delay tolerances set by the FCC are as follows:
"A sine wave, introduced at those terminals of the transmitter which
are normally fed the composite color picture signal, shall produce a radiated signal having an envelope delay, relative to the average envelope
delay between 0.05 and 0.20 MHz, of zero microseconds up to a frequency of 3.0 MHz; and then linearly decreasing to 4.18 MHz so as to
be equal to -0.17 microsecond at 3.58 MHz. The tolerance on the envelope delay shall be 1-0.05 microsecond at 3.58 MHz. The tolerance shall
increase linearly to ±0.1 microsecond down to 2.1 MHz, and remain at
±0.1 microsecond down to 0.2 MHz. (Tolerances for the interval of 0.0
to 0.2 MHz are not specified at the present time.) The tolerance shall also
increase linearly to ±0.1 microsecond at 4.18 MHz."
This requirement is illustrated graphically in Fig. 2 -8. The operational
techniques for meeting this specification for the transmitter must be left
to more advanced study. However, the reader must be familiar with this
aspect now, since it is necessary to consider the overall system (studio,
transmitter, and receiver) in learning to recognize "color distortions."
Another FCC requirement is that Ey, Et, EQ, and the components of
these signals match each other in time to 0.05 microsecond. This requirement must be met by inserting proper delays in the Y and I channels
Tolerance Limits
csi
T-,
+0.1
f
g
o
\
\\
1
a
0.1
d -0.2 -Tolerance
not Specified
a
for the Present
0. 3
N.
W
0. 4
o
I.0
2.0
3,0
Radiated Sideband Frequency in MHz
Fig. 2-8. Envelope -delay tolerances.
4.0
49
THE SYSTEM CONCEPT
(both transmitter end and receiver end) to match the delay in the Q
channel brought about by the narrow bandwidth in this channel.
Radiation of the transmitter more than 3 MHz outside the channel
must be at least 60 dB below the visual transmitter power. The voltage
of the upper sideband must not be greater than -20 dB for a modulating
frequency of 4.75 MHz or greater. The voltage of the lower sideband must
not be greater than -20 dB for a modulation frequency of 1.25 MHz or
greater. For color, it must not be greater than -42 dB for a modulating
frequency of 3.58 MHz (color subcarrier).
It does not necessarily follow that a transmitter that meets the sideband
requirements when broadcasting monochrome will do so when radiating
a color signal. This is because the chrominance subcarrier, when modulated on the picture carrier, may appear in the lower adjacent channel at
a substantially higher level than the vestigial -sideband filter is designed to
handle. An additional notch filter is required in that case; it provides
attenuation at a frequency 2.33 MHz below the channel edge
(3.58 1.25 = 2.33 MHz ).
-
o
I
I
I
I
1
10
i
I
I
I
I
20
ó
E
30
1
1
1
1
1/
i
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
!
I
\1/
-1j\/
i
I
/
1
\
40
/
/
/
I
I
I
I
-42 dB
I
I
I
I
50
-3
!
I
-2.33
2
-1
0
0.5
1,
25
2
3
4
5.45 5.75
Megahertz (Reference Lower Band Edge)
Fig. 2-9. Visual sideband attenuation characteristic.
The specification for visual -sideband attenuation is shown in Fig. 2 -9.
This characteristic must be measured with the color subcarrier present.
The FCC waveforms for horizontal, vertical, and color sync signals are
illustrated in Fig. 2 -10. It is common practice for the FCC to state all
times in terms of H, where H is one television line, or 63.5 microseconds.
In practice, times in microseconds usually are much more useful; these are
given in Fig. 2 -11 (horizontal pulses and color sync burst) and Table 2 -2
(vertical pulses). With the aid of this information, you can use an oscilloscope to "standardize" your sync generator.
TELEVISION BROADCASTING CAMERA CHAINS
50
ä
ó
s
m
LL
r>
M
611
I
(A) Vertical interval. (See expanded details on page 52.)
Fig. 2-10. Standards for transmitted
THE SYSTEM CONCEPT
51
NOTES
1.
H
2. V
= time
-
from start of one line to start of next line.
time from start of one field to start of next field.
3. Leading edge and
trailing edge of vertical blanking each should
be complete in less
than 0.1H.
trailing slopes of horizontal blanking must be steep enough to preserve
minimum and maximum values of x + y and z under all conditions of picture
content.
4. Leading and
*5. Dimensions marked with asterisk indicate that tolerances given are permitted only
for long time variations and not for successive cycles.
6. Equalizing -pulse area shall be between 0.45 and 0.5 of area of
pulse.
a
horizontal -sync
follows each horizontal pulse, but is omitted following the equalizing
pulses and during the broad vertical pulses.
7. Color burst
8. Color bursts to be
omitted during monochrome transmission.
9. The burst frequency shall
be
±10
Hz
with
a
be 3.579545 MHz. The tolerance on the frequency shall
maximum rate of change of frequency not to exceed 1/10 Hz per
second.
10. The
horizontal- scanning frequency shall be 2/455 times the burst frequency.
for the burst determine the times of starting and stopping
the burst, but not its phase. The color burst consists of amplitude modulation of a
continuous sine wave.
11. The dimensions specified
12. Dimension
P represents the peak excursion of the luminance
signal from blanking
level, but does not include the chrominance signal. Dimension S is the sync amplitude above blanking level. Dimension C is the peak carrier amplitude.
13. For monochrome
transmission only, the duration of the horizontal -sync pulse between 10 percent points is specified as 0.08H ± 0.01H, the period from the leading
edge of sync to the 10 percent point on the trailing edge of horizontal blanking
is specified as 0.14H minimum; and the duration of vertical blanking is specified as
0.05V, +0.03V and -0. All other dimensions remain the same.
synchronizing waveforms.
52
TELEVISION BROADCASTING CAMERA CHAINS
1
.2pp
$
ó*
,
2
(B) Expanded details.
Fig. 2-10. Standards for transmitted synchronizing waveforms. (Cont'd.)
THE SYSTEM CONCEPT
53
The FCC further states: "The color picture signal shall correspond to
a luminance component transmitted as amplitude modulation of the pic-
ture carrier and a simultaneous pair of chrominance components transmitted as the amplitude- modulation sidebands of a pair of suppressed sub carriers in quadrature." Also, "The chrominance subcarrier frequency shall
be 3.579545 MHz ±10 Hz with a maximum rate of change not to exceed
1 /10 Hz per second."
11.1µs
0.56µs
1.59µs
2.24{E-1-1
4.76µs
960s
0.90P
1
0.9 to 1.15
0.1P
0.1S
Breezeway
Nominal
Microseconds
Tolerance
Microseconds
+0.3
Blanking
11.1
-0.6
Sync
4.76
Front Porch
1.59
-0.32
Back Porch
4.76
+0.96
-0.61
Sync to Burst
0.56
-0.17
Burst
2.24
Blanking to Burst'
6.91
Sync & Burst
7.56
-0.49
Sync & Back Porch
9.54
±0.32
±0.32
+0.13
+0.08
+0.27
0
+0.08
-0.17
+0.38
1.
Blanking to burst tolerances apply only to signal before addition of sync.
Fig. 2 -11. Time intervals for horizontal -sync pulse.
TELEVISION BROADCASTING CAMERA CHAINS
54
Table 2 -2. Vertical -Pulse Widths
Minimum
Nominal
Maximum
(µs)
(µs)
(As)
2.4
4.5
1250
2.54
5.08
1333
Pulsa
Equalizing
Vert Serration
Vert Blanking
2.0
3.81
167
1
NOTES
Vertical-sync pulses are not specified. See Detail A -A, Fig. 2 -10B. The width of the vertical sync pulse is set by the tolerance on the width of the vertical serration. An interval of
63.5 microseconds (H) must exist from the leading edge of the last (leading) equalizing pulse
to the trailing edge of the first serration.
2. The width of the equalizing pulse must be 0.45 to 0.5 of the horizontal sync width used.
3. Vertical blanking in terms of H is from 18.375 to 21 lines. Although 21 lines is shown as
"maximum" in the above chart, this is the width of vertical blanking maintained by the networks. It allows vertical- interval test signals to be inserted (usually on lines 18 and 19 of
vertical blanking) with a suitable "guard band" of blanking lines before the start of active
line scan.
4. Horizontal and vertical blanking must be of proper ratio to establish the 4:3 aspect ratio.
1.
The FCC describes the luminance modulation as follows: "A decrease
in initial light intensity shall cause an increase in radiated power (negative
transmission) ." Further, "The blanking level shall be transmitted at
75 -± 2.5 percent of the peak carrier level. The reference white level of the
luminance signal shall be 12.5 ± 2.5 percent of the peak carrier level... .
-
120
110
100
104
White Clip
- -- -30.0
00.0%
12.5%
15.0
90
80
70
-
60
40
Burst
30
20
10
7.5
22
61.25
1810
63.75
45
4
-10
-20
-30
-40
iF 62. 5%
J
68.75
71.85/
72.50
18
22
Sync Pulse
77. 50
-
Fig. 2 -12. Video -signal amplitudes.
70.3%
75.0%
/*-- 87. 5%
86. 25/
88. 75
100%
Note: Tolerance value specified at blanking level applies to
sync amplitude only. The variation of blanking with respect
to the tolerance of setup shown is assumed to be zero.
The reference black level shall be separated from the blanking level by the
setup interval, which shall be 7.5 ± 2.5 percent of the video range from
blanking level to the reference white level."
The above specification is illustrated in Fig. 2 -12. The studio scale is
shown on the left, and the corresponding transmitter scale (percent modulation) is shown on the right.
THE SYSTEM CONCEPT
55
2 -2. EVOLUTION OF THE COLOR -BAR SIGNAL
Remember that the nominal input level to the encoding system from the
camera is 0.714 volt. This is 100 IEEE units when the scope is calibrated
for 1 volt over an IEEE graticule of 140 units. The output level of tube type color -bar generators usually is changed to the 75- percent amplitude by
switching the internal output impedance from 75 ohms to 75 percent of
this value, or 56 ohms. Thus, a 220 -ohm resistor switched across the output
gives 56 ohms, with a resulting 25-percent reduction in output level. In
solid -state generators, the output level sometimes is set by switching zener
diodes with the proper reference voltage to set the clipper levels.
But there is another consideration. The camera outputs have the setup
level established at a nominal 7.5 percent of white video. The color -bar
output of Fig. 2 -3 does not have blanking inserted. To convert to the
standard transmission system, consider the signal in terms of IEEE units
for a 1 -volt (peak to peak) signal in 140 IEEE units. The setup is 7.5
percent from 0 IEEE units toward peak white. Sync is added and extends
to -40 IEEE units. Burst is added on the blanking base (0 IEEE units)
and extends between +20 IEEE units and -20 IEEE units (peak -to-peak) .
If maximum level (occurs in yellow and cyan) is to be no more than
100 IEEE units (reference white peak) , we have the following
considerations:
See Table 2 -3. The chroma amplitude for 100- percent yellow bars is
0.447 (0.894 peak to peak) , which, added to the yellow luminance of
0.89, gives a peak amplitude of 1.34, for an overshoot of 34 IEEE units
above reference white. This chroma is now reduced to 75 percent, or
(0.75) (0.447) = 0.3352. Since this level, added to the luminance level,
must not exceed 1 (or 100 IEEE units), then 1.000 0.3352 = 0.6648,
and the new luminance level for yellow at 75 percent of full amplitude is
0.664 (Fig. 2 -13) . The white bar at 1.00 is then reduced to 0.746 by this
process, or to approximately 75 percent of full amplitude. All other parts
of the signal are reduced by the same proportion. (The "voltage" scale in
Fig. 2 -13 must be taken as proportional to the IEEE scale.)
As yet, we have not added the 7.5- percent setup level. When this is
done, the video of the previous signal is reduced in peak -to -peak value
by this same amount. (The blanking pedestal contains no video information. Peak picture black is now at 7.5 IEEE units instead of 0 IEEE units.)
So now for yellow ( for example) to have the same "brightness" as before, its amplitude must be reduced toward white by the same amount it
was reduced toward black by the 7.5- percent pedestal. You have found
already that the difference between the 75- percent yellow luminance and
peak white is (essentially) 0.336 (1 0.664 = 0.336) . To move this the
appropriate amount toward white, the 0.336 is reduced by 7.5 percent of
white: (0.336) (92.5 %) = (0.336) (0.925) = 0.310. This is the new
difference (to allow for setup) between peak white and yellow luminance.
-
-
TELEVISION BROADCASTING CAMERA CHAINS
56
Table 2 -3. Color -Bar Signal Chart
Subcarrier
Signal
Y
Amplitude
Phase
R
G
B
R
1
0
0
0.30
0.635
103.4°
RG
1
1
0
0.89
0.447
167.1°
G
0
1
0
0.59
0.593
240.8°
GB
0
1
1
0.70
0.635
283.4°
B
0
0
1
0.11
0.447
347.1°
BR
1
0
1
0.41
0.593
1
1
1
0.1571
0
1
0.1571
0.4135
0.5371
0
1
0.2711
0.4265
33°
(Yellow)
(Cyan)
60.83°
(Magenta)
RGB
1
0
-
(White)
(R
B- Y
-Y = 0)
Q
= 0)
(I
-Y
-Y = 0)
R
(B
I
--Y /=
90°
Y)
(G
0.367°
1
0
0.3371
0.3371
0.5848
90.03°
1
0.4056
0
0.5393
0.4865
123°
1
0.7317
0
0.7317
0.4313
146.38°
-
-
-
-
0.8429
1
0
0.8429
0.4135
179.6°
0.4629
1
0
0.7289
0.4265
213°
0
1
0.6629
0.6629
0.5848
270.03°
Q=0
(G
-
0)
Color
0.20
180°
Burst
--Y-=
Y)
(B
(R
0)
-Q
(I
=
0)
--Y-=
Q=0
- -Y
- =/
Y)
(R
(B
0)
I
(G
(G
Y)
90°
0)
0
0.5944
1
0.4607
0.4865
303°
0
0.2683
1
0.2683
0.4313
326.4°
THE SYSTEM CONCEPT
57
1.00
1.0-
0 992
0.880
Yellow
0.8-
0.746
Cyan
0.746
Gray
0.693
0.664
Green
0.60.522
Magenta
0.440
Red
0.417
0.40.305
-
r
>°
0.2
-
0.328
0.223
Burst
Blue
0.082
0-0.2
-0.4
-
-
0.052
L
00
Black
-0.134
-0.246
-0.253
Fig. 2 -13. Color bars at reduced amplitude.
-
Therefore the final new luminance value for yellow is 1 0.310 = 0.690,
or 69 IEEE units. See Fig 2 -14. All other signals, except for sync and burst,
are reduced ( toward white) in like manner.
Fig 2 -15 illustrates the newly developed 75 percent bar signal. The
peak -to-peak chroma values are designated at the top of each bar in IEEE
values. Bear in mind that although the bars are reduced in value to 75
percent of NTSC developed bars, they still represent fully saturated signals. For any color to be desaturated, some amount of all three primaries
would have to be present.
2 -3.
DEFINING AND RECOGNIZING "DISTORTIONS"
IN NTSC COLOR
In the consideration of color-transmission "distortions," it is necessary
to consider the transmission and reception processes together; in fact, it
is impossible to consider one without the other. For example, consider the
matter of gamma correction. The pickup device must actually be made
nonlinear in gray scale to match the "average" color picture tube. The sys-
TELEVISION BROADCASTING CAMERA CHAINS
58
100
100-
100
100
Reference
White
89
i
80-
77
77
72
9
60-
56
'B-
36
38
/1
2
CO
20
20-
15
--
.5
12
Reference
Black
-
01,1
-20
-16
-
-16
20
40-
Fig. 2 -14. Color bars with setup added.
100
100
100
89
CL
80
CL
-
77
77
CL
CL
72
69
60
40
20
-
56
48
Tolerance +2.5 IEEE Units
on pk-pk Chrominance Levels
46
36
38
U
28
-
C
CL
15
12
C
7
0-5
-20
-
The duration of each of the bars is
1/7 of the scan between blankings.
Tolerance «10%
-40
Fig. 2 -15. 75- percent color -bar signal.
-16
-16
THE SYSTEM CONCEPT
59
tern following the pickup device must be made as strictly linear as possible, so that all of the amplitude- versus -brightness correction can be made
in the pickup device, and complicating factors are not introduced after
this correction.
Gamma correction is a camera -head function, and as such will be
covered in Chapter 6. The point to be emphasized here is that the system
handling video information can be carrying a nonlinear function ( from the
pickup device) and that it is normal for this function to appear nonlinear.
Nonlinearity of a certain kind is required (in the case of gamma correction) because of picture-tube characteristics, or (in other cases which
we will point out) because of receiver or monitor design. In the cases
just cited, you should understand that these nonlinearities are not
"distortions "!
The Basic NTSC Color Problem
The number one problem is phase sensitivity. For accurate reproduction
of a hue, the resultant color phasor must be held within a tolerance of
±5 °. Hue error becomes noticeable at ±-10° from the proper position
relative to burst. The transmission and reception process, to maintain such
a tight tolerance, must operate with a signal time -base accuracy of a few
nanoseconds.
Color information (hue and saturation) is affected by various types
of amplitude and phase shifts in transmitting and receiving equipment.
In addition, hue shifts can occur as a result of multipath transmission
errors. Burst and subcarrier -sideband phase shift caused by variations in
topography over different transmission paths can require the viewer to
readjust his receiver with every station change. Although this can be
annoying to the viewer, it is not a factor under the control of the station
operator. However, it is necessary to adjust the station color gear to
precise operating parameters ( and this is possible!) and to minimize, as
much as possible within the limitations of the equipment, any amplitude
or phase distortion. When you receive complaints from viewers, you rest
easy when you can prove the performance of your system to qualified
technical agents.
In the case of multipath-reflection errors, if you are transmitting an
accurately adjusted color picture, the viewer can readjust his hue control
from the setting for a different station and still get a good color picture.
Far worse is the situation in which there is a differential phase error. In
this case, different colors can be shifted by different angles, depending on
luminance. The viewer then can only attempt to strike a "happy medium"
with his hue control.
Burst Phase Error
Remember this: The color monitor or receiver will "insist" that the
burst phase not be in error. This results from the fact that the subcarrier-
60
TELEVISION BROADCASTING CAMERA CHAINS
sideband information is synchronously demodulated with the burst as a
reference. The angles clockwise from burst for various colors are as in
Table 2 -4.
Fig. 2 -16A shows the proper vectors for burst, red, and magenta. Now
we assume we have the burst -phase error (0) shown in Fig. 2 -16B. The
color receiver requires the burst phase shown in Fig. 2 -16A; to visualize
this, put an imaginary pin at the center dot, and rotate the vectors as in
Fig. 2 -16C to place the burst at the correct phase. The other vectors become R + O and M + 0, which represent a red shifted toward yellow and
a magenta shifted toward red.
Table 2 -4. Color Phase With Burst
Color
Burst
Yellow
Red
Magenta
Blue
Cyan
Green
Phase Angle (Degrees)
0
12.9
76.6
119.2
192.9
256.6
299.2
The effect of burst -phase error is to rotate all hues in the direction opposite to the burst -phase error. If the burst error in Fig. 2 -16B had been
counterclockwise, it would have been necessary to rotate the vectors
clockwise to return the burst to the proper position. Red would then tend
to go toward magenta, magenta would tend to go toward blue, etc., all
around the color gamut.
You can visualize this most conveniently by using the color triangle as
in Fig 2 -16D. Note carefully that this triangle has been turned around
from the position normally presented, to fit it into the NTSC phase diagram. The center of the circle is at illuminant C, where the color vectors
collapse to zero value. Place the imaginary pin for the circle here; the
color triangle must remain fixed in the position shown. If the burst slips
clockwise by some angle, all reproduced colors shift counterclockwise by
the same angle (and vice versa for counterclockwise slip of burst phase)
Note again the interdependence of the encoding and decoding processes.
The hue control at the receiver is an operational adjustment. It has, in
most modern receivers, a range of at least ±70 °, whereas the tolerance
is only ±10° for the overall transmission system. This tells you that any
receiver in normal operating condition should be able to have its hue control adjusted to obtain proper colors from your station. Your responsibility
at the sending end is to assure that the burst phase is as nearly correct as
possible so that (theoretically) receivers need not be readjusted for proper
color reproduction from different stations. Remember that if you do
.
THE SYSTEM CONCEPT
61
Burst
0°
(A) Correct vectors.
(B) Burst phase error.
(C) Effect at receiver.
Circle
#
Ú
Pinned
Y
'
Here
I.
Pr'"
i
Triangle Fixed
C
(D)
Effect on colors.
Fig. 2 -16. Effect of burst phase error.
transmit a burst phase error, the receiver exhibits a "locked phase error"
so that colors reproduced are not exactly the same as in the original scene.
Since the viewer does not have a direct comparison, he is not aware of this
as long as he can obtain good flesh tones.
Quadrature Distortion
Quadrature distortion results from cross talk between the I and Q video
information. It has more possible causes than most of the other types of
color distortion. It does not involve an operational control at the receiver,
but the receiver can cause this effect if the circuits, especially the quadrature- transformer adjustment, are faulty.
62
TELEVISION BROADCASTING CAMERA CHAINS
A symptom of quadrature distortion is color displacement; in severe
cases, a girl's red lips can be in the middle of one cheek. In the more usual
case, there is color fringing ( not caused by camera misregistration or a
misconverged picture tube) at the edges of color transitions.
The most obvious type of quadrature distortion occurs when I and Q
are not phased exactly 90° in the encoder. Fig. 2 -17 illustrates the case in
which Q lags I by more than 90 °. For simplicity, only the basic colors in
the first and third quadrants are plotted. Now remember the polarities of
I and Q in each quadrant. These are:
Quadrant 1 is bounded by +Q and +I
Quadrant 2 is bounded by +I and -Q
Quadrant 3 is bounded by -Q and -I
Quadrant 4 is bounded by -I and +Q
(same polarity of I and Q)
(opposite polarities of I and Q)
(same polarity of I and Q)
(opposite polarities of I and Q)
.
.
.
.
The relative values of I and Q for the colors plotted are as developed in
Table 2 -1. Note that colors with large amounts of Q are affected more than
others. For example, red and cyan have relatively small amplitude and phase
errors; green and magenta have larger amplitude and phase errors. By
adjusting the hue control on the monitor or receiver, you cannot get a good
red and cyan simultaneously with a good green (green will be yellowish)
or magenta (magenta will be bluish)
Now if you will take the trouble to plot yellow and blue ( second and
fourth quadrants) , you will find these increased in amplitude. For a Q lag
greater than 90 °, colors in the first and third quadrants are reduced in
amplitude; those in the second and fourth quadrants are increased in
amplitude. In each quadrant, the phase error is in the direction of the Q
phase error.
This is emphasized further by Fig. 2 -18, in which Q lags I by less than
90 °. For simplicity, only magenta is shown in quadrant 1. Note that it is
now increased in amplitude, whereas yellow and blue are reduced in amplitude. In either case, the phase error is in the direction of the Q error.
Since the receiver separates signals with a 90° relationship, I will crosstalk into Q and vice versa.
Before going further, be sure to grasp the fundamentals of chrominancesignal transmission and reception for NTSC color. See Fig. 2 -19 and the
following analysis:
Fig. 2 -19A represents the signal as transmitted. I and Q are double side band in the region shown. The upper sideband of the wideband I
chrominance is cut off at the transmitter to achieve a 20 -dB roll -off at the
sound carrier frequency. A portion of the lower sideband of the I
chroma constitutes single -sideband information; no Q chroma exists there.
The outputs of the I and Q demodulators are equal in the double -sideband region (Fig. 2 -19B) .
.
THE SYSTEM CONCEPT
63
Over the single -sideband region, the voltage output of the I demodulator is one-half that which occurs in the double -sideband region (Fig.
2 -19C). Note also that a one -half -I voltage, shifted in phase by 90 °,
appears at the Q- demodulator output. This is EI at its single -sideband
frequencies of about 0.6 to 1.5 MHz.
The output of the I demodulator is boosted by 6 dB above 0.5 MHz to
recover the gain lost in the single -sideband region (Fig. 2 -19D) . The Q
demodulator is limited in bandwidth to 0.5 MHz.
The filtering and relative gain action of Fig. 2 -19D results in voltages
EI and EQ free of crosstalk (Fig. 2 -19E). This assumes, of course, that I
and Q are actually being transmitted in the proper quadrature relationship.
"Narrow- band" color receivers demodulate on the R
and B
axes.
These receivers use the same bandwidth for all chrominance compo-
-Y
-Y
Red
Amolitude Error
Magenta
+I
0.6
/
0.28
//
/
r
Amplitude and
Phase Error
\
\
0,525
02\/
Burst
\\
/
/
5
\
o, z8
/
Amplitude a nd
Phase Erro r
0.6
Green
--
Amplitude Error
Cyan
Error
Correct
Fig. 2 -17. Q lagging
I
more than 90 °.
+Q
\\
\\
Error
Q+B
64
TELEVISION BROADCASTING CAMERA CHAINS
nents. With all chrominance channels of the same bandwidth, delay equalization is unnecessary, and no crosstalk occurs in a properly aligned receiver
of this type. Again, this assumes that the I and Q chroma signals are being
transmitted in the proper quadrature relationship. Also, there are other
causes of lack of quadrature than misadjustment of the Q lag.
I
ncreased
Amplitude,,
\
Magenta
+(Q -BI
i
0.32
0.525
0.31
Yellow
Reduced
Amplitude
Reduced Amplitude
/
Blue
0.31
0.32
-
--
Error
Correct
Fig. 2 -18. Q logging
I
less than 90 °.
To investigate other causes of quadrature distortion, study Fig. 2 -20.
Fig. 2 -20A shows the usual double -sideband representation of a modulated
carrier. If, for example, the highest modulating frequency is 0.5 MHz,
upper sidebands (fL) extend 0.5 MHz above f0, and lower sidebands
(fL) extend 0.5 MHz below fc. This can be represented by equal- amplitude phasors ( vectors) rotating in opposite directions as in Fig. 2 -20B.
The resultant amplitude is the vector sum of for and fL added to the carrier vector so that the resultant always lies along carrier line YO.
In Fig. 2 -20C, the amplitude- frequency response exhibits a rapid roll off above the carrier frequency. Thus for is severely attenuated (Fig.
2 -20D). The resultant vector no longer lies along line YO, but contains
a quadrature component, as shown in Fig. 2 -20E. If fc is the color sub carrier frequency, this type of sloping response will result in crosstalk in
both the I and Q detected signals in the video- frequency range of 0 to
0.6 MHz. It makes no difference whether I -Q or color- difference demodulation is used; the result is the same as crosstalk among all the colors.
65
THE SYSTEM CONCEPT
Single-Sideband Region
2.1 MHz
0 MHz
I
Demodulator
Q
Double-Sideband Region
3.1 MHz 3.58 MHz 0.1 MHz
Demodulator
Notes
Output From
Double- Sideband
Region
0.5 MHz
E112L
E
I12/90°
Output From
Single -Sideband
Region
0.5 MHz
1.5 MHz
Filters
0.5 MHz
1.5 MHz
0.5 MHz
Net Output
1.5 MHz
Fig. 2 -19. Response of
I
and
Q
demodulators.
Fig. 2 -20F shows this as applied to transmitter envelope response. A
rapid rolloff too close to the color subcarrier frequency of 3.58 MHz will
result both in denaturation of colors and in quadrature crosstalk. Fig.
2 -20G shows rapid variations in response around 3.58 MHz. Although the
FCC allows a ±2 dB variation, the Rules further state that this variation
must be substantially smooth. Sudden dips or peaks must be avoided for
good color transmission.
(Envelope -delay distortion at the transmitter is a major contributing
factor in color misregistration. However, a study of this subject is more
appropriate to a text on television system maintenance.)
66
TELEVISION BROADCASTING CAMERA CHAINS
fU
fL
J
--L_
(A) Double -sideband signal.
fc
-
fL
tU
,- ____-___
J
Carrier
(B) Double- sideband vectors.
fc
(C) Modified response.
I
L
L
N
Quadratures
I
Component
(D) Modified vectors.
JI
3.58 MHz
(E) Quadrature component.
3.58 MHz
4.1 MHz
Chroma Sidebands
Chroma Sidebands
(F) Excessive rolloff.
(G) Uneven response curve.
Fig. 2 -20. Causes of quadrature distortion.
67
THE SYSTEM CONCEPT
Effect of Carrier Unbalance
We know that in a doubly balanced modulator, the carrier is suppressed so that only the sidebands remain. If this suppression is not perfect, the carrier appears in the output, and a condition known as carrier
unbalance exists. Under this circumstance, the carrier adds itself vectorially
to all vectors present in the encoder output. To visualize this, study Fig.
2 -2IA, which represents a carrier unbalance in the positive direction of
the I modulator. Since the unbalance occurs in the I modulator, a new
line, parallel to the I axis, is drawn from the proper color vector to the
new vector representing the amount of carrier present. Since the unbalance
is in the positive direction, the new vector is toward the +I axis. The
resultant colors are shifted toward the orange axis of the +I vector, as
well as being changed in amplitude.
I
Improper
Red
Drawn Parallel
To
I
Vector
Proper Red
Improper Yellow
(A) Positive I carrier unbalance.
Improper Blue
Proper Yello
Proper Blue
-Q
Improper Green
Proper Green
Yellow
0 qs
I
lluminant C
Proper
Yellow
C
0.35
Blue
(B) Proper cancellation for white.
Improper
C
-Blue
(C) White shifted toward yellow.
Fig. 2 -21. Effect of carrier unbalance.
Now see Fig. 2 -21B. Recall that primary and complementary colors have
equal amplitudes but are opposite in phase. If both yellow and blue have
the same amplitude, the result of their vector addition is illuminant C,
or white. This is the proper complementary relationship. But note from
the vectors of Fig. 2 -2IA that the blue amplitude has been reduced and
the yellow amplitude has been increased. You see the result in Fig. 2 -21C:
white or gray areas become colored because of incomplete cancellation of
the subcarrier. Remember that "white" or "gray" can occur only during an
interval of zero subcarrier.
68
TELEVISION BROADCASTING CAMERA CHAINS
Carrier unbalance shifts all hues (as well as whites and grays) in the
direction of unbalance. A positive I unbalance shifts toward orange; a
negative I unbalance shifts toward cyan. A positive Q unbalance shifts
toward yellow -green; a negative Q unbalance shifts toward purple.
Effect of Video Unbalance
Recall that double balancing of the modulators means that both the
carrier and the modulating video are balanced out, leaving only the side bands of the subcarrier frequency. If I and Q video suppression is not
complete, the condition is known as video unbalance.
See Fig. 2 -22A. The outputs of the I and Q channels for the indicated
color bars are shown. Vector addition of the I and Q signals results in
the amplitudes shown in Fig. 2 -3. Fig. 2 -22B shows the result of a ±Q
video unbalance. Note that the axis for all colors with plus values of Q is
shifted in the positive direction. However, the actual peak -to -peak values
Y
C
G
M
0.6
0.6a
0.32
I
R
0.26 0.28
0.32
only-
0
Dc
Axis
(A) Normal I and Q.
0.52 0.52
0.31
0.31
-0.21
Q
0,21-
only-
M
l
Axis
)Shift
(B) Positive
Fig. 2 -22. Effect of video unbalance.
Q unbalance.
THE SYSTEM CONCEPT
69
of these colors remain the same. The net result is that the unwanted video
signal is added to the luminance signal after the chroma signal is combined
with the luminance signal, and the gray scale of the picture is distorted.
Note that the effect of a positive Q video unbalance is to brighten reds,
blues, and purples, and to darken yellows, greens, and cyans.
For a negative video unbalance, the colors with negative amounts of Q
would be shifted upward. In this case, reds, blues, and purples would be
darkened, and yellows, greens, and cyans would be brightened.
Effect of Chroma Gains and Gain Ratio
The transmission paths from encoder input to receiver matrix must
maintain a constant ratio for Y, I, and Q. A variation of gain in any one
of the paths results in loss of color fidelity.
The noncomposite luminance level for a color -bar pattern (Fig. 2 -3A)
is 0.7 volt to peak white. When chrominance gain is correct (Fig. 2 -3B)
and of the proper I -to -Q gain ratio, and chrominance is added to the
luminance signal of Fig. 2 -3A, we have the following condition (see
Fig. 2 -3C)
:
1.
Bars
1
2. Bars 5
3. Green
and 2 overshoot by 33 percent.
and 6 undershoot by 33 percent.
(bar 3) just touches black level.
The above assumes that 100 -percent bars are used, and that no blanking
(pedestal) is inserted in the signal.
Now see Fig. 2 -23. This is the same presentation as Fig. 2 -3B except
that the values of I and Q are shown for each color bar. Suppose that the
ratio of I gain to Q gain is not correct. As you would expect, a deficiency
of I gain would reduce the saturation of colors in the orange -cyan gamut,
leaving greens and purples practically unaffected. Conversely, a deficiency
of Q gain would reduce saturation of greens and purples without practical
difference in the orange -cyan region. By noting the relative I and Q levels
making up each color as in Fig. 2 -23, you can understand how the pattern
of Fig. 2 -3C would show these deficiencies on a scope:
I is high relative to Q, bar 2 will be higher than bar 1, and bar
will be lower (greater undershoot) than bar 6.
2. If Q gain is high relative to the I gain, bar 1 will be higher than bar
2, and bar 6 will have greater undershoot in the black region than
will bar 5.
1.
If
5
Effect of Differential Gain
Differential gain means that the gain of the 3.58 -MHz chroma information is not constant with brightness level. This results in a change of saturation sensation with brightness.
70
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 2 -24A represents a stair -step signal with a superimposed 3.58 -MHz
sine wave. Fig. 2 -24B shows the same signal observed through a high -pass
filter to eliminate the low- frequency steps; a system with strictly linear
amplitude response will result in sine waves of equal amplitudes for each
step, as shown.
0.63
0. 63
0.59
0.59
0.45
0.45
C
I+0.32 I--0.6
Q
-0.31
Q
M
G
-0. 28
I
-0.21
Q
-
I
+0. 28
-0.52 Q-+0.52
R
I+0.6
Q+0.21
-0.45
I--0.32
Q
+0. 31
-0.45
-0.59
-0.63
Fig. 2 -23.
I
-0.59
-0.63
and Q chroma amplitude ratios.
Fig. 2 -24C represents the type of nonlinearity in which black regions
are compressed and white regions stretched. Normally, this condition will
be apparent also on the steps, as shown, and would be evident when the
signal is passed through a low -pass filter to observe the steps only. However, it is possible to have linear low- frequency amplitude response and
nonlinear high -frequency amplitude response. This is why the use of
low -and high -pass filters is convenient in observing test signals of this type.
Fig. 2 -24D illustrates the opposite type of nonlinearity, and the same
conditions apply.
The transfer curves of Figs. 2 -24C and 2 -24D are unusual. Generally,
you will find that amplitude nonlinearity occurs either at one end or the
other, or at both ends, with a relatively linear middle response. This tells
you that those colors near the white or black extremes normally will be
most susceptible to saturation changes, particularly with highly saturated
colors.
Effect of Differential Phase
Fig. 2 -25A shows the same signal as that pictured in Fig. 2 -24A. Although the sine waves may look the same on each step, a phase displace-
71
THE SYSTEM CONCEPT
White
-
(A) Signal to be amplified.
Step
Light Gray
Dark Gray
I
/
(B) Linear amplification.
Linear
Response Curve
Step
1
Step 2
Step 3
(C) Black compressed,
white stretched.
/
Nonlinear
Transfer Curve
(D) White compressed,
black stretched.
Nonlinear
Transfer Curve
Fig. 2 -24. Effects of
differential gain.
TELEVISION BROADCASTING CAMERA CHAINS
72
ment can occur with brightness level, as shown by Fig. 2 -25B. This error
is termed differential phase.
Remember that the subcarrier phase carries hue information. A low brightness yellow should be the same as a high- brightness yellow. This is
not to say that the two yellows would appear the same on the receiver. But
the point is, the observed color should be yellow and not ( for example)
green or red as the brightness of the yellow component changes.
'
1-Hi
Phase Displacement
With Luminance Level
123
(A) Signal to be amplified.
(B) Sine -wave output components.
Fig. 2 -25. Differential phase.
In practice, the effect of differential phase is judged best in the yellow
and blue areas ( the two extremes of the luminance scale) A system introducing as much as 10° of differential phase can result in a monitor or
receiver adjustment that gives proper reproduction of a high -luminance
hue such as saturated yellow, or a low- luminance hue such as saturated
blue, but not both simultaneously. One or the other will be off-color.
When this defect is accompanied by more than 10- percent differential
gain (as often occurs) , the error becomes quite noticeable.
.
2 -4.
DIGITAL CONCEPTS
The rate of change of technology in broadcasting requires technical
personnel to spend a greater proportion of time in acquiring new
knowledge to solve problems. Continuing education therefore is no longer
incidental to the job, but an essential part of it.
One example is the rapid increase in the application of digital circuitry
to broadcast equipment. Synchronizing generators, video switchers, and
electronic character generators commonly employ such circuitry. Recently,
camera chains using digital control, either wholly or partially, have been
introduced. In such systems, a single triaxial cable replaces the bulky
85- conductor camera cable. In field applications, a simple rf link can be
used to control a remotely located camera from a base station.
THE SYSTEM CONCEPT
73
This section briefly reviews the common symbols and terminology of
the basic logic functions.3 Integrated -circuit (IC) "chips" are normally
used, and "schematics" are usually just flow diagrams which the user must
know how to interpret.
The digital concept recognizes only two numbers, or conditions:
1 (one) and 0 (zero) . In positive logic, a 1 (high level) represents the
true or more positive level, and a 0 (low level) represents the false or
less positive level. This kind of logic is used most often. Negative logic
(sometimes used) means that the voltage level assigned to logic 1 is negative with respect to the voltage level assigned to logic O.
See Fig. 2 -26. In row 1 is the symbol for a noninverting amplifier. Thus,
if we have A on the input, we should have A on the output. Sometimes
the output is identified by X, as shown, to distinguish between input and
output signals. If the input is 1, the output is 1, and if the input is 0,
the output is O. Truth tables are input /output tables that show input conditions and the resultant output conditions.
Truth Table
Symbol
Row
A
Terminology
A
A
X
1
1
0
0
A
X
1
0
0
1
A
X
0
1
1
0
1
A
X
A
Á
2
A
X
A
A
3
A
A
A
o
Noninverting
Amplifier
Inverting Amplifier Output (X)
is Complement of Input
(Phase Inversion).
Significant input is high W.
Same as 12), but
input
significant
is low level (0).
X
Input or output signal.
Not A; complement of A; false if A is true.
Low 10) is the significant (reference/ state.
Fig. 2 -26. Common logic symbols (amplifiers).
°The reader should have basic background at least equivalent to that contained in Harold E. Ennes, Workshop in Solid State (Indianapolis: Howard
W. Sams & Co., Inc., 1970).
TELEVISION BROADCASTING CAMERA CHAINS
74
In row 2 is the symbol for an inverting amplifier. A small circle at the
input or output of a symbol indicates that 0, or the low state, is the significant state. If just the input or output has the small circle, the amplifier
is inverting. Thus, for the symbol shown in row 2, we know that if the
input is A, the output is inverted A, which is called not A. The bar over
the letter ( for example, as in A) means "not A," inverted A, or the
complement of A. All three terms are synonymous and simply mean that
the input signal is inverted. Another way of saying this (see the truth
table) is that if A is 1, then X is 0 ( the significant state) and if A is 0,
then X is 1.
Row 3 illustrates just the opposite of row 2. The action should be apparent from inspection.
Digital circuitry is made up of amplifiers, gates, and flip -flops. A 1 or 0
is either on
condition is analogous to an output controlled by a switch
(closed) or off (open) The digital stage is either fully conducting or
,
-it
.
fully cut off.
The AND gate is a basic logic circuit. It has two or more inputs and
one output. The output will be 1 only when all inputs are 1 simultaneously.
If one (or more) of the input signals is 0, the necessary condition for a
1 output is not fulfilled, and the output is O.
The NAND gate is functionally equivalent to an AND gate followed by
an inverter. Thus, the NAND gate produces a 0 at the output when all the
input signals are 1 simultaneously.
The OR gate is another basic logic circuit. Like the AND gate, it has two
or more inputs and a single output. The output is 1 when one or more of
the inputs are at the 1 state. Thus, if any signal input is at the 1 state,
the output is 1.
The NOR gate is functionally equivalent to an OR gate followed by an
inverter. Thus, the NOR gate produces a 1 output only when all the input
signals are 0 simultaneously. A 1 applied to any input results in a 0
output from the NOR gate.
The basic AND, NAND, OR, and NOR gates are reviewed in Fig. 2 -27.
Consider the AND gate in row 1. Since no circle is shown, we know that
the significant output is 1 (high level) when the inputs are 1 simultaneously. This may be expressed as X = AB, which is read "X equals A
and B." Note from the truth table that the output is 1 only when both A
and B are 1.
Also note in row 1 that the OR gate, as symbolized, has the same truth
table. We know from the symbology that the significant output is 0 ( low
level) when one or both of the inputs are O. This may be expressed as
X = A + B, which is read "X equals A or B." X is 1 only when both inputs
are 1 simultaneously.
The reader should follow through the remainder of Fig. 2 -27 and be
sure he understands the truth table for each pair of AND ( NAND) and OR
(NOR) gates.
THE SYSTEM CONCEPT
Row
AND INONDI
\
A
1
B
A
2
3
B
T
cl
d
)
)
\
1
1
0
0
0
1
0
0
0
0
A
B
X
1
1
0
1
0
0
0
1
1
0
0
0
A
B
X
0
).
A
X
J
cl
1
0
1
0
0
0
0
A
B
X
1
1
0
1
0
0
0
1
0
0
0
1
A
B
X
1
0
1
0
1
1
0
0
0
A
B
X
X
^----\
g
)
1
1
1
0
0
0
0
1
A
8
X
X
B
X
X
A
)
B
g
0
B
0
1
A
B
X
A
1
1
0
A
g
1
0
1
g
0
1
1
0
0
1
Fig. 2 -27. Common logic symbols (gates).
X
X
:::»
))x
A
B
).
\//
A
0
:::»
) >
)
/
1/
/
X
p
7
A
B
0
MORI
X
X
///
8
1
X
B
B
B
1
X
a
6
OR
A
C
g
A
uth Table
B
B
A
4
75
4
)
/
X
X
X
76
TELEVISION BROADCASTING CAMERA CHAINS
Another basic block of digital circuitry is the flip -flop. This is a bistable
multivibrator which remains in its most recent state until an input causes
it to change states. The input trigger pulse usually is differentiated and
then applied to a diode to polarize the pulse so that only the positive going or negative -going edge causes the bistable circuit to respond. Sometimes a dc shift rather than a pulse is used.
For example, note the toggle flip-flop of Fig. 2 -28A. Every time a negative -going transition occurs at input T, the bistable changes state. Recall
that a multivibrator can have two output signals with a 180° phase relationship. Thus, if a trigger arrives when Q = 0 and Q = 1, then Q
changes to 1 and Q changes to O.
With the toggle flip -flop, there is no predetermined state for the two
outputs when the circuit is first turned on. Thus, the state of the outputs
after a trigger pulse is applied cannot be predicted unless the present state
is known. The circuit simply changes states each time a negative -going
transition is applied. The set -reset bistable, or RS flip -flop, (Fig. 2 -28B)
overcomes this problem. The set -reset circuit has two inputs and the usual
two complementary outputs. As indicated by the truth table, a 1 input to
the set (S) terminal makes the Q output 1 and the Q output O. A 1 input
at the reset (R) terminal reverses the state: The Q output becomes 0
and the Q output becomes 1. Zero signals on both inputs do not change
the state. If both inputs should receive simultaneous signals (l's) , the next
state cannot be predicted. Thus, simultaneous inputs are commonly termed
not allowed or forbidden combinations. This simply says that the device
cannnot be in both states simultaneously. The RS flip -flop cannot be used
in logic situations which include the possibility of simultaneous set and
reset inputs.
The action of the RS flip -flop is best understood by going momentarily
to Fig. 2 -29. Two NAND gates cross-connected as shown form a flip-flop.
When power is applied, opposite states will exist; we will assume arbitrarily these are a 1 output for B and a 0 output for A. The 0 output of
A at the input of gate B becomes a 1 at the output, and the 1 output of
gate B at the input of gate A becomes a 0 at the output ( phase inversion
of NAND gate) .
Now assume a 1 appears at the set (S) terminal. This 1 becomes a 0
at the output of gate B and drives the output of gate A to 1. Thus, a set
input has set the significant output to the normal 1 and the previous 1
output to O. A 1 input to R will now reset the device to the previous
state. Follow this action again from the truth table for Fig. 2 -28B. Note
that the behavior of the circuit is predictable for three of the four possible
input conditions.
Flip -flops may be clocked or unclocked. In the unclocked flip-flop just
discussed, the outputs respond to the inputs as the inputs change. In the
clocked flip -flop, a clock input must exist at the time the inputs change
for the outputs to respond.
THE SYSTEM CONCEPT
77
-Q
Z Input responds
to negative -going
Given Present
State After
State
Trigger Pulse
Q
Q
Q
Q
0
1
1
0
1
0
0
1
signal only.
(A) Toggle.
Input
Output
Q
R
S
0
0
0
1
1
0
1
0
0
1
1
I
Not Allowed
S
Q
0
Not Allowed
0
I
0
1
1
0
1
0
1
1
I
Q
No Change
(B) Set -reset.
Input
R
0
- Input responds to positive-going signal only.
Output
I
No Change
(C) Clocked set- reset.
Q
CP
K
-
Input
Output
J
K
0
0
0
1
0
1
1
0
1
0
1
1
Q
No
l
Q
Change
Comp ement
- Input responds to negative -going signal only.
(D) Clocked JK.
Fig. 2 -28. Common logic symbols
(flip -flops).
Fig. 2 -28C illustrates a clocked flip -flop drawn for negative logic. The
symbol at the clock -pulse (C0) input indicates that the clock input responds only to a positive -going transition. The small circles at the R and S
inputs indicate logic inversion at those points. Interpret this to indicate
that a false level is inverted to become a true level within the block. Note
that this circuit is the complement of that of Fig. 2 -28B.
TELEVISION BROADCASTING CAMERA CHAINS
78
Arbitrary
Normal
Output
Output
I.
0
1
Fig. 2 -29. Cross -connected
NAND gates.
o
One of the most popular logic units is the JK flip -flop shown in Fig.
There are no ambiguous states. When a 1 is applied to the J
input, the Q output is 1 and the Q output is O. (In a clocked flip -flop,
the clock pulse must be present.) When a 1 is applied to the K input, the
Q output flips to 0, and the Q output flips to 1. When l's are applied to
both the J and K inputs, the flip -flop switches to its complement state.
Sometimes two or more J and K inputs exist. One J and one K input may
be tied together for use as a clock input.
Special forms of gates are used in certain logic functions; special symbols are used to represent these functions. Fig. 2 -30A indicates two
NAND gates with outputs paralleled. The symbol indicating that this circuit
actually performs as an OR circuit is shown at the junction of outputs fl
and f2. This is termed a wired OR, or sometimes a phantom OR. The truth
statement for this circuit is: If fl is true OR f2 is true, the output is true.
A circuit that produces a true output only when the input states are
not identical is termed an EXCLUSIVE OR gate (Fig. 2 -30B) . Note from the
2 -28D.
A
B
X
o
o
o
o
X
o
o
D
(B) EXCLUSIVE OR.
(A) Wired (phantom) OR.
R
*5V
A
B
7]
/
X
A
B
o
o
X
o
o
o
o
(C) EXCLUSIVE OR complement.
(D) Discrete RTL EXCLUSIVE OR.
Fig. 2 -30. Special gates for logic circuitry.
THE SYSTEM CONCEPT
79
truth table that a 1 output is produced only if just one input is 0. If both
inputs are of like polarity (either 0's or l's) , the output is 0.
The complement of this function is shown in Fig 2 -30C. A 0 output
is obtained only if the inputs are of opposite polarity. To make this clear,
study the discrete resistor -transistor logic (RTL) circuit illustrated in
Fig. 2 -30D. If both inputs are 0's or l's, both transistors have zero- biased
base-emitter junctions, and neither can conduct. Under this condition, output X sees the full value of supply voltage (high level) , because there is
no current through RL. Since the A input is tied to the emitter of Q1 and
the base of Q2, while the B input is tied to the emitter of Q2 and base of
Q1, unlike polarity (opposite logic levels) will turn one of the transistors
on. The resultant voltage drop across RL sends output X to a low level.
Note again that this is the complement of the logic function in Fig. 2 -30B,
as indicated by the small circle at the output of the logic symbol shown in
Fig. 2 -30C.
A digital system contains numerous switching devices that have 0 or
1 outputs. Since operation is based on two states, the binary numbering
system (based on the number 2) is a "natural" for this application.
Binary and decimal numbers are reviewed in Table 2 -5. Only two symbols, 0 and 1, are used in the binary system. Note that when decimal 2 is
reached, we move the binary 1 one place to the left to indicate we have
counted to two one time. At the count of three, we use binary 11 to indicate one two plus one one, or 3. At the count of four, we are again out
of symbols, so we write 100 which indicates one four plus no twos plus
no ones. At the count of five, we write 101, which indicates one four plus
no twos plus one one. We continue until at the count of seven we write
Position Value
of Each Symbol
Decimal
Binary
1
X 23
8
1000
1
X 22
4
100
0X21
0
0
X 20
1
1
1
Total
13
Comments
Any number times 0isO.
Any number to the ()power is equal to
1.
Binary number is composed strictly of zeros and ones.
1101
1101
1x23
j
1x22
1
0x21
1x20
I
0
Total
=
13 in
Decimal Form
Fig. 2 -31. Conversion of binary 1101 to decimal form.
TELEVISION BROADCASTING CAMERA CHAINS
80
Table 2 -5. Binary and Decimal Numbers
Binary
Decimal
0
0
1
1
10
2
3
11
4
100
5
101
6
110
7
111
8
1000
9
10
1001
1010
11
1011
12
13
1100
14
15
16
17
18
19
1110
1101
1111
10000
10001
10010
10011
20
10100
21
10101
22
23
24
25
26
27
28
29
30
10110
10111
-
11000
11001
11010
11011
11100
11101
11110
31
11111
32
100000
111. Again we are out of symbols in all columns, so we write 1000, which
indicates one eight plus no fours plus no twos plus no ones. The conversion of one example (binary 1101) to the equivalent decimal number
is reviewed in Fig. 2 -31.
The term bit means binary digit. The term character refers to a group
of bits. The term word refers to the total number of bits required for a
particular system. A word may be defined either by the total number of
bits or the total number of characters. For example, a certain system may
use a 192 -bit word. A character may consist of ( for example) a group
of four bits. Thus, this system uses a 48- character word.
Digital logic circuitry is easy to troubleshoot if the technician has a
little experience and familiarity with a particular system. Fig. 2 -32 is a
simplified schematic diagram of a small portion of video logic circuitry in
81
THE SYSTEM CONCEPT
which raw sync is inserted at the input of IC4. Assume that sync pulses
exist at the output of gate IC4 but not at the output of IC3. This does not
necessarily mean that integrated circuit IC3 is faulty. Note that the symbols all indicate NAND gates. Pin 12 of IC3 must have +5 volts dc for
the sync pulses to pass. It is obvious that for this to occur transistor Q1
must be cut off. Note that the hold -off bias for this stage is determined by
a video clamp level, which may or may not be an internal adjustment.
Checking the dc levels back from pin 12 of IC3 is necessary to determine
the cause of the trouble. If pin 12 is receiving +5 volts when the sync
output is lost, the IC3 chip probably is at fault. When a chip is definitely
determined to be faulty, the entire chip is replaced. The important point
to remember is that the input conditions to a chip must be correct for the
proper output condition to exist.
+5 V
Ql
To
Other Logic Circuitry
+V
5
+5V®. sv
12
13
Bias Set by
Video Clamp Level
Sync In
110.
D.
Fig. 2 -32. Example of troubleshooting in logic circuit.
2 -5. A
DIGITALLY CONTROLLED COLOR CAMERA
A portable 3-Plumbicon color camera employing digital control is illustrated in Fig. 2 -33. The camera was developed by CBS Laboratories,
and, under agreement with CBS, it is being manufactured and marketed
worldwide by Philips as the PCP -90 "Minicam." This camera, with back
pack, produces an encoded signal intended for direct broadcasting. All signal processing is accomplished in the back pack so that separate red, blue,
and green signals need not be sent to the base station. This reduces the
possibility of noise pickup and cuts down on color errors caused by
multipath effects.
The camera transmits signals to its base station (Fig. 2 -33B) on the
microwave frequency of 2 GHz (7 GHz and 13 GHz are optional) from
an omnidirectional antenna on the back pack.
Remote -control signals to the camera are carried on a frequency of 30
MHz ( for cable) or 950 MHz ( for rf) , with a 450 -kHz subcarrier for
the command signal and a low- frequency interphone carrier. The command
system permits radio control of all functions from a base station located
as far as 10 miles away, depending on the transmission path. The camera
can be linked to its base station by a triaxial cable if terrain features
82
TELEVISION BROADCASTING CAMERA CHAINS
interfere with wireless communication. However, cable losses limit the
distance between camera and station to one mile, unless repeaters are used.
For on- the-spot recording with a portable video recorder at the camera
location, a local control box plugged into the back pack allows the
operator to perform all functions of the digital command system.
A three -inch picture tube is used in the viewfinder along with a simple
magnifying lens and a polarized filter to enhance contrast under outdoor
Receiving
Antenna
transmitting
Antenna
Visible
Viewfinder
..
Not
Lens
Assembly
Camera
Head
Iris Position
Local Control
(A) Camera and back pack.
Microwave Receiver
Remote Control Signals
Microwave Video Link
Back Pack
VHF
Transmitter
Base Station
crf
Alternate Coaxial Cable
(B) System block diagram.
Courtesy Philips Broadcast Equipment Corp.
Fig. 2 -33. PCP -90
digitally controlled portable color camera.
83
THE SYSTEM CONCEPT
viewing conditions. A waveform is presented at the left of the picture
display to enable the cameraman to adjust pedestal and iris locally when
this is desirable. There is also a pulse to indicate field strength of the signal received at the base station.
The PCP-90 is available with either a Canon 6 -to-1, f/2.8 zoom lens
or an Angenieux 10 -to -1, f /1.8 zoom lens. The camera head weighs 181/2
pounds, and the back pack weighs 12 pounds (32 pounds with battery)
The top section of the back pack (Fig. 2 -33A) contains the uhf-vhf
data receiver and microwave transmitter (in the wireless version of Fig.
2 -33B) The main center section houses video processing (Chapter 6) ,
including the NTSC encoder (Chapter 9) , sync circuits, and command
control circuits. At the bottom is the battery pack for wireless use, or the
cable power converter.
For the digital command system, frequency modulation of a 950 -MHz
carrier is used. An audio tone at 5.4 kHz is phase- shifted 180° to identify
a single bit of information. The resultant significant sidebands occur between 2.7 kHz and 8.1 kHz on the 450 -kHz command subcarrier. This
permits adding a 250 -to -2500 Hz interphone voice channel to the main
carrier without excessive cross talk. For cable operations, amplitude modulation of a 30 -MHz carrier is used to transmit the video.
An example of a 24 -bit command word is shown in Fig. 2 -34A. Since
up to six cameras can be controlled, the first three bits address the specified camera to be commanded in the immediately following word. This is
followed by 15 bits of simultaneous commands, and then by six bits of
one-at -a -time commands that allow any one of 64 individual functions to
be controlled by the base- station operator. The complete command word
for each camera is preceded by a 9 -bit code of all l's (Fig. 2 -34B) to alert
the camera decoder that an event is to take place. Also, the 24 -bit command
word is followed by a 24 -bit parity, which is the complement (interchanged l's and 0's) of the preceding command word. All of this is done
for the purpose of reliable noise immunity of control. Thus, unless the
command word is followed by the complement, nothing happens. This
virtually eliminates false commands and erroneous functions of the command system.
The complete command word may be observed to consist of 57 bits
(9 -bit start plus 24 -bit command plus 24-bit parity) The complete 57 -bit
command requires 10 milliseconds to transmit (Fig. 2 -34C) At the end
of this time, the base -station sequencer steps to the address of the next
camera. Thus, the sequence time for six cameras is 60 milliseconds, allowing each camera control an interval of 10 milliseconds 16 times during
.
.
.
.
each second.
A genlock module (horizontal, vertical, and color -subcarrier lock) is
used to compare local sync with sync from each remote camera. Digital
commands are transmitted to bring all cameras into exact synchronization
with each other and with the local sync generator. This permits fades, lap
84
TELEVISION BROADCASTING CAMERA CHAINS
dissolves, and special effects between any combination of remote and local
signals.
It is quite natural that the principles of this design be extended into
the area of studio operations. Such is the case with the Philips PC -100
illustrated in Fig. 2 -35. The camera and studio control unit are shown; a
portable control unit also is available.
The camera head is contained in a magnesium casting. Hinged covers
are provided for access to interior components. The integral lens mounting
includes internal mechanical drive shafts that automatically couple to
internal shafts in the camera body. Available zoom lenses include fully
servo -controlled lenses, as well as lenses with manually controlled focus
and zoom and servo -controlled iris. Range extenders for these lenses are
available. A remotely controlled, motor- driven filter wheel and two slide
filters are provided between the lens and color beam splitter for color and
neutral- density correction.
The color beam splitter is a prism block. Linear matrixing is included
to allow the use of a more efficient beam -splitter prism. A one -inch
separate-mesh Plumbicon tube is used; this tube provides a minimum of
Gen lock
Registration Trim
Field Strength
Beams
Master Pedestal
Encoder Setup
Iris
Gain
Tally Light
Pedestals
011
101100010101101
3 Bits
15 Bits
Camera Address Simultaneous Commands
00101
6 Bits
One-at -a -Time Commands
24 -Bit Command Word
(A) 24 -bit word.
100010011101010010011010
111111111
9 -Bit
24-Bit
24-Bit Parity
Start
Command Word
(Complement of Command Word)
Complete Command Code for Camera
1
(B) Complete command.
ms-¡-
1--- is
ou ms
I
1
2
3
4
5
6
(C) S :x- camera sequence.
Fig. 2 -34. Camera command code.
1
etc.
85
THE SYSTEM CONCEPT
(A) Camera and studio control unit.
To Camera
On
-Air
Encoded Video
Monitor Video
(B) Signals multiplexed on
triaxial cable.
-
External Viewfinder
Camera Power
Intercom
--
To CCU
-
Program Sound
X60 Digital and 60 Analog
Control Functions
Courtesy Philips Broadcast Equipment Corp.
Fig. 2 -35. PC -100
digitally controlled studio color camera.
86
TELEVISION BROADCASTING CAMERA CHAINS
35- percent modulation at 400 lines. Low target capacitance contributes to
a 50 -dB signal -to -noise ratio through the preamplifier. An anti -comet -tail
gun increases the dynamic range of the tubes so that high lights of 20 to
30 foot-candles on the surface of the tubes can be handled without blooming. This gun effectively imparts a light -level saturation "knee" to the
transfer curve.
The deflection -coil assembly is shielded by a spun Mumetal housing to
eliminate the influences of external magnetic fields on the registration
accuracy. This assembly is mounted in machined castings. The complete
yoke assemblies are fixed into position in the RGB casting and factory
aligned by fixing the yoke shell to the front housing. The yokes can be
removed for servicing and replaced with only the normal line -up procedure. The "spider" casting provides long -term optical stability. Yoke
rotation and focus are accomplished with separate knurled shafts.
Since the video-processing circuitry is in the camera head, the effects of
temperature variations on a long camera cable do not influence the
camera performance. With the exception of some presets, the camera does
not contain any setup controls, so that the complete line -up of the camera
chain can be carried out by the camera -control -unit (CCU) operator.
Information is transferred between the camera and the CCU by a triaxial cable; this cable weighs one -tenth as much as standard color cable.
Transmission is accomplished by multiplexing three channels of information through the cable (Fig. 2 -36). These channels include a video channel, for sending encoded video from the camera to the control location; a
monitor channel, for sending monitor signals from various points in the
video- processing chain to the control location; and a telecommand channel,
for transmitting all control, registration, and setup signals from the CCU
to the camera. In addition to these three channels, 100 -volt dc power is
supplied to the camera through the triaxial cable. The maximum cable
length is one mile (more with the addition of repeaters) .
The electronic view finder contains a 7 -inch rectangular picture tube.
It is tiltable, rotatable, and removable. Any combination of either RGB
minus G or Y and external video signals can be selected for display. An
electronic zoom indicator is superimposed on the top of the picture. The
"on -air" tally light can be seen from all angles.
The camera control unit is separated into three major subassemblies;
the monitor unit, the registration and operating panel, and the electronics
unit. These subassemblies are linked by cables in the rear of the console
and can be accommodated either in a standard 19 -inch rack or in two
19 -inch transit cases.
The electronics unit consists of a 7- inch -high rack that houses a two level card bin. The card bin is mounted in a retractable, tiltable drawer for
ease of servicing. Coaxial and multipin connectors are mounted on the
rear panel. The circuitry includes the cable -drive demodulators, audio
modulators, the analog-to- digital (A /D) converters, and the power
87
THE SYSTEM CONCEPT
E
ó
j ó
I
II
5 a
8
-TT$
t.
pV 27
g
Courtesy Philips Broadcast Equipment Corp.
Fig. 2 -36. Simplified block diagram of the PC -100 system.
88
TELEVISION BROADCASTING CAMERA CHAINS
supply. Final processing of the encoded video signal occurs in the electronics unit with the addition of studio sync and timing.
The registration and operating panels are mounted in a drawer assembly
that can be placed in the cabinet at various positions. An overlay panel
is provided to cover the registration controls after setup of the camera.
To aid in camera setup, switches and associated lamps are interlocked to
reduce the number of manual operations required. For example, a
REGISTRATION push button is provided to switch the matrix and contours
(Chapter 6) to off and the encoder input signals to the waveform monitor.
At the same time, it also switches the waveform monitor to the RGB
sequential mode and minus -green video to the picture monitor.
Other camera systems -the two -channel color camera and the more conventional cabled camera systems -are covered elsewhere,4 and these descriptions will not be duplicated here. However, details of camera cables
and other types of control and distribution cables are covered in the next
chapter. Also included in Chapter 3 is coverage of the special types of
camera -system regulated power supplies and power distribution.
EXERCISES
Q2 -1.
Q2 -2.
Q2 -3.
Q2 -4.
Q2 -5.
Q2 -6.
Q2 -7.
Q2 -8.
Q2 -9.
The camera is delivering unity green, unity blue, and zero red signals. Calculate the resulting color amplitude and phase.
The chroma signals on the black level immediately following the
burst in Fig. 2 -3C are special I and Q signals from a color -bar
generator. They are simply special test pulses that are fed to the I
and Q inputs of the encoder. They obviously are of zero luminance;
the ac axis centers on black level. Why, then, are these signals
actually visible on the color monitor?
You may have learned that when (for example) the red chroma
was less than 0.63 of unit luminance, this meant it was not fully
saturated; some degree of "white" was present, meaning some degree of both of the other primaries was present. How can a fully
saturated red be simulated at much less encoded amplitude?
Where is the FCC requirement for the width of the horizontal
front porch called out in Fig. 2 -10?
What is the proper amplitude of the color -sync burst?
Does the color -sync burst occur following every horizontal -sync
time in the complete composite color signal?
How many cycles of color -burst signal should be present?
What are the attenuation -vs- frequency requirements for a color
transmitter?
If you apply a linear stairstep signal to the first amplifier of a
camera head, should the signal at the output of the color encoding
system be linear?
4Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971).
THE SYSTEM CONCEPT
89
Q2 -10. If you apply a linear stairstep signal to the input of an encoder,
should the signal at the output of the encoder be linear?
Q2 -11. What is the basic problem in handling NTSC color?
Q2 -12. What factor other than transmitting- system deficiencies can result
in receiver color problems?
Q2 -13. How would the following colors be reproduced in a properly adjusted monitor or receiver, if the burst in the encoder were at
-160° from zero reference on an NTSC polar diagram? (Assume
all other adjustments to be correct.)
1. Yellow
2. Cyan
3. Green
4. Magenta
5. Red
6. Blue
7.
White
Q2 -14. What could cause white to be contaminated with a certain hue?
Q2 -15. What would cause "washed-out" color in a properly adjusted
receiver?
Q2 -16. What is the major criterion in the adjustment of a receiver for
"good color "?
Q2 -17. What could cause a human face to be reproduced with unnaturally
ruddy complexion and very dark red lips? Assume a properly adjusted receiver.
Q2 -18. What would cause yellows to be "washed out," while blues are reproduced well?
CHAPTER
3
Camera Mounting, Interconnection
Facilities, and Power
Supplies
This chapter covers camera mounting facilities, cable interconnection
methods, and power supplies and power distribution for the camera chain.
3 -1.
THE CAMERA PAN AND TILT CRADLE
The camera head normally mounts directly on the panning head to allow
it to be moved vertically and horizontally by the camera operator. The panning head then is mounted on a pedestal (Fig. 3 -1) , a tripod that is either
fixed or on wheels (Fig. 3 -2) , or a crane (Fig. 3 -3) Any panning -head
mounting on wheels is termed a dolly. Thus you may have a pedestal dolly,
a tripod dolly, or a crane dolly.
The panning head includes controls that are adjustable to allow the
camera operator to make pan and tilt operations smooth and stable with
his individual touch. The panning head (or cradle) is the basic link in the
chain of camera operations and must be mastered completely before the
cameraman can do an adequate job.
The panning head normally engages the camera head by means of a
slotted screw on the mounting plate. Once in place, the camera head is
secured by tightening this screw. All lock and drag controls are then disengaged, and the panning handle is held loosely by the cameraman so that
the center -of- gravity adjustment can be made. The adjustment usually is
accomplished by means of a control on the front (or, in special cases on
the rear) of the panning head. This adjustment allows the camera to be
balanced properly (tilts neither forward nor backward) when the panning head control handle is not held.
90
.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
91
The adjustable friction (pan drag and tilt drag) controls must be set
to suit the individual touch of the operator. Just sufficient drag must be
used so that the movement of the camera does not "jiggle" or jerk the
image. At rare intervals, the production may require a whip shot calling
for a rapid pan. In this event, the drag must be light enough to allow
rapid but smooth operation.
Panning Head
Pedestal Adapter
Steering Ring
Studio Pedestal
Courtesy RCA
Fig. 3 -1. Camera mounted on studio pedestal.
A common error of new camera operators not familiar with the exact
location and use of adjustments is to confuse the drag, brake, and lock
controls. Sometimes the braking or even the locking of the panning head
is done by means of the drag controls. This practice considerably shortens
the life of the panning-cradle mechanisms.
The variety of panning heads and dollies in current use makes a detailed description of each type impractical. However, the nomenclature of
adjustments should allow the cameraman to adapt the following procedures
to his particular equipment. The information given is directly applicable
to the panning head of Fig. 3 -2.
TELEVISION BROADCASTING CAMERA CHAINS
92
v
-erinder
,
'e',ti'inder
Hood
,.e'inder
Cmlrols
F II Brake Adjust
Jilt Lock Pin
I
Tin Drag Adjust
'Forward of
nter ohne.
C
hissi
"-,'.l Drag
Drag Adjust
Control Fio
Angie Ad ]u
r
Pan
Adjustl
Brvi
ipr^; Adapter
Tripod
Adju st
m
Panning Hea','
Control Ha c ',
Hantlle
Focus Knot)
Courtesy RCA
Fig. 3 -2. Controls on panning head.
Center of Gravity Adjustment
After the camera is installed, the center of gravity must be established
for balanced operation. While firmly holding the handle, proceed as
follows:
Release the tilt -lock pins and all tilt-motion drag on the cam -head
mount.
2. Release the handle enough to note the direction in which the camera
head tends to tilt.
3. Tilt the camera toward the required center of gravity adjustment and
rotate the CG ADJUST control knob until the camera balances.
4. Apply the tilt brake with the TILT BRAKE lever, or secure with the
tilt -lock pins.
1.
Operation
The following procedures enable the operator to adjust the cam -head
mount pan- and-tilt motion while the camera is operating.
Pan Motion -The PAN BRAKE knob is located at the lower rear center
of the cam -head mount. To lock the mount in azimuth, rotate the PAN
BRAKE knob clockwise. The PAN DRAG ADJUSTMENT knob is located on
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
93
the right-hand side plate. To increase the drag in pan motion, rotate this
knob clockwise.
Tilt Motion -The TILT BRAKE lever is located in the rear of the cam head mount, below the top plate. To maintain the mount in the selected
tilt position, press the lever to the left. To lock the mount vertically,
rotate and release the spring -loaded tilt-lock knobs on both sides of the
mount, and adjust the tilt until the pilot pins engage with the detents in
the balance cams.
The TILT DRAG ADJUSTMENT knob is located just forward of the PAN
DRAG ADJUSTMENT knob. To increase the drag in tilt motion, rotate the
knob clockwise. CAUTION: Do not tighten the TILT DRAG ADJUSTMENT
until the cam roller shaft is locked and not free to rotate. Otherwise, the
cam will be damaged.
Lubrication-After every 100 hours of operation, three areas of the
cam -head mount should be coated lightly with Dow Corning Silicone
Lubricant, Compound No. 4, or equivalent. Apply the lubricant to the
operating surfaces of the balance cams, the vertical stabilizer bars ( just
under the top plate of the panning head) , and the CT ADJUST screw. The
cam operating surfaces should be kept clean at all times.
NOTE: The cameraman should always lock the head when he leaves the
camera for more than a few minutes.
3 -2.
PEDESTAL DOLLIES
The most common form of studio camera support is the pedestal dolly
(Fig. 3 -1) An example of such a dolly, the Houston Fearless Model PD -9
pedestal, is described in this section.
.
Description
This pedestal is designed to mount a standard friction head and a blackand -white or color television camera. It is designed for one -man operation.
The pedestal acts as a firm, stable mount for the camera, and it also provides mobility for dolly shots and for raising and lowering the camera
during operation. The rising column on which the camera is mounted is
motor driven so that the camera is raised or lowered by operating a small
switch attached to the head or camera control handle. The steering wheel,
which is located directly below the camera at all heights, guides the three
sets of dual wheels, one at each apex of the triangular base. The wheels are
equipped with ball bearings.
Two types of steering are available, and by operating a foot pedal on
the base of the pedestal, either may be chosen by the operator while the
camera is on the air. Synchronous steering, in which all wheels are locked
parallel and turn simultaneously, is best for tracking in a straight line. By
operating the hand wheel and observing the positioning arrow on one of
the spokes of the wheel, the operator can set the direction of the wheels
94
TELEVISION BROADCASTING CAMERA CHAINS
before the pedestal is in motion. The pedestal will then travel in the direction indicated by the arrow. In tricycle steering, only the forward wheel is
steered, and the back wheels are locked parallel. This enables the pedestal
to follow a curved course and to turn sharply in any direction.
On the corners of the triangular base are nonslip step plates on which
the cameraman can stand when the camera is in the raised position. The
base of the pedestal is of arc -welded steel. The center column is seamless
steel tubing. The trim and steering wheel are satin -chrome finish. All bearings in the equipment are packed and sealed at the factory, so no lubrication is required.
The specifications of the equipment are as follows:
Maximum Height (Not Including Friction Head)
Minimum Height (Not Including Friction Head)
Maximum Width
Minimum Width
Electrical Requirements
55 inches
36 inches
381/4 inches
341/2 inches
115 volts,
60 hertz,
5.6 amperes
Weight
365 pounds
The motor -driven column consists of a seamless steel tube that is raised
and lowered inside a larger, fixed seamless steel tube. It is held in position
by two sets of three rollers, 120° apart, which guide and align the rising
column. A rubber pad is provided at the center of the column so that the
column will settle softly when it is moved to its lowest position. The
column is driven up and down by a motor and gear box, the hoisting being
done with a flexible steel cable. Limit switches prevent overtravel at either
end of movement. The motor may be reversed instantly and is controlled
by a switch at the end of a cable. The switch is fitted with a mounting
bracket, and the mounting bracket may be attached to the control handle
of either a standard tilt head or the RCA color camera (when panning head
is not required) . The switch usually is mounted so that it may be operated
with the thumb of the left hand.
A large wheel is located just below the mount for the friction head.
Turning this wheel steers the pedestal in the direction shown by an arrow
on one of the spokes of the wheel. Operating the wheel drives a gear shaft
that extends down through the center column of the pedestal. Since this
shaft is telescoping, steering operation is available regardless of the height
of the pedestal. The wheels are steered by a chain -and -sprocket arrangement
that operates directly from the base of the telescoping shaft. The chain
drive is directly to the front wheel of the triangular base. When the arrows
on the wheel and column are in line and the STEER 3 pedal is depressed, all
of the wheels are connected by the chain. Consequently, when the steering
wheel is turned, all three wheels move in synchronism and remain parallel.
When the arrows are aligned and the STEER 1 pedal is depressed, the two
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
95
rear wheels are declutched from the chain drive and locked in their fixed
positions. The steering then operates only on the front set of dual wheels
and is of the tricycle type.
Cable guards are provided all around each pair of wheels to prevent the
wheels from running into the cables. These guards have slotted holes and
are adjustable to any height above the floor by pushing them up or down
with the foot.
The head of the rising column is equipped to mount a standard friction
head. To install the friction head, remove the lock nut from the head. Place
the head in the circular hole in the top of the column, and insert the lock
nut through the opening in the side of the column. Then attach and tighten
the nut to hold the head in place.
A clip, located on the upper portion of the soundproof motor blimp, is
provided as a retainer for the motor cable. To prevent the cable from dragging, it may be inserted into the clip after the motor -switch bracket is
mounted on the control handle of the tilt head or camera.
Operation
The large wheel just below the tripod mount is turned to direct the
wheels as desired by the operator. This may be done while the pedestal is
in any position, as the operator normally operates the panning arm with
one hand and moves and steers the pedestal with the other. The arrow on
one of the spokes of the wheel indicates the direction in which the wheels
are turned and the direction in which the pedestal may be expected to
move when it is pushed by the operator. When synchronous steering is desired, the arrow on the wheel is aligned with the arrow on the column, and
the STEER 3 pedal, on the back of the pedestal base, is depressed. Operating
the hand wheel then turns all three sets of wheels simultaneously, and the
sets of wheels remain parallel. This enables the pedestal to move in a
straight line in any direction from its resting position. When the arrows
are aligned and the STEER 1 pedal is depressed, the front wheel is steered
as the front wheel of a tricycle, and the pedestal may be moved on a curved
path determined by the setting of the steering wheel.
The cable guards are provided to push away cables as the pedestal is
moved, to prevent the cables from jamming the pedestal or rocking the
camera as it rolls over them. These cable guards are adjustable. On a level
floor, they may be set so that they are barely above floor level. On uneven
surfaces, they are set slightly higher. The cable guards are equipped with
slotted holes so that they may be moved up and down with the foot to
obtain the desired clearance.
Access to the column motor drive is obtained by releasing three buckles
on the motor blimp. This housing should be removed on a regular basis,
such as once a month, to permit inspection of the lifting cable for signs of
fraying. Whenever fraying is observed, the defective cable should be
replaced immediately.
TELEVISION BROADCASTING CAMERA CHAINS
3 -3.
THE CRANE DOLLY
The crane dolly is used for more elaborate productions and is rarely found
outside the main network operating centers. The four-wheeled truck
shown in Fig. 3 -3 is pushed by one or more trackers, or pushers, who also
may be responsible for controlling the height of the pivoted arm. The
camera is mounted at the top of the arm, and a seat behind the camera
position is provided for the cameraman.
The manual type of crane dolly is being replaced rapidly by motor -driven
mechanisms, some of which can be operated solely by the cameraman. A
recent model allows the cameraman to sit alongside the tracker; camera
operation is by remote control. The viewfinder is replaced by a video monitor, and fingertip controls allow camera-aiming, lens-zoom, and focus
adjustments to be made.
The elimination of the bulky camera cable by the techniques of digital,
multiplexed systems (Chapter 2) could drastically affect camera dollying
systems in the future. It is possible that lightweight color cameras, not inhibited by the more conventional 84-conductor cable, may be suspended
on lever arms from the ceiling and remotely controlled to pan, tilt, raise,
lower, and swing in any desired arc and at any height.
Courtesy Houston Fearless Corp.
Fig.
3
-3. Crane dolly for television camera.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
3 -4.
97
PROMPTING EQUIPMENT
This entire chapter is concerned mainly with components external to,
but used in conjunction with, the actual camera or control units. One important system external to the camera is the means used to "prompt" the
performer.
Formal speakers, newsmen, and commercial announcers "on camera"
normally require a prompter. Cue cards ( sometimes termed "idiot cards")
are often used when the required information is limited. These are large
cards, on which the information is printed in large letters, held by a person
either near the camera or in some other suitable location.
Several types of prompters are in use; one of these is illustrated in Fig.
3 -4. A special typewriter with %s -inch type is used to prepare the copy on
paper that has sprocket holes along each edge. The prompter unit normally
mounts at the top front of the camera as shown in Fig. 3 -4A. Eight lines of
copy are visible at a time, and since the lines are 22 characters long, some
twenty to twenty -five words are visible on one full frame. The sprocket
holes along each edge of the paper engage sprockets on the prompter
mechanism, and the sheet rolls upward at a speed controlled to suit the
pace of the performer.
The prompter chassis shown in Fig. 3 -4A is a one-piece aluminum structure, corner braced and welded. The copy- correction panel and spools are
gold anodized to resist marks, scratches, and feed or take -up spool paper
displacement burrs. The device is about 15 inches across the face, 11 inches
high, and 5 inches thick.
The sync control ( Fig. 3 -4B) has a removable aluminum cover. Internal
components are mounted on subchassis that are removable from the bed
plate. The bed plate is secured to a one -piece foam -rubber shock mount
attached to the bottom of the case. The inside of the case cover is acoustically lined with vinyl foam. An extendable hand control is shown lying on
top of the case.
The external casing of the sync control, prompters, and hand control are
electrically isolated from the ac supply. The cables connecting the sync
control and the prompters are shielded, and the shield provides the continuous case ground between units. A 2 -pin connector provides ac. The
ground lead provided should be secured to a solid ground.
Line -for -line sync is provided by the control unit. In this application,
"sync" means the rolls on two or more prompters, advancing in the same
direction, move together, line for line with one another, regardless of speed
or the amount of typed copy on each prompter. The dual function of drive
and sync is performed by a synchro torque system consisting of a control
transmitter coupled to 1, 2, 3, or more torque receivers as required. The
transmitter is contained within the sync control and is coupled to the respective receivers in the prompters by five wires in the 7 -core cable .The transmitter and the receiver are physically and electrically similar. Each consists
98
TELEVISION BROADCASTING CAMERA CHAINS
THAN
MORE
NOW
SYSTEMS
TY
INC
400 SYNC
USED
.
BY
FILM STUDIOS:
BUSINESS
AND
US
GOV'T AGENCIES.
TELESYNC'S FUTURES
INCLUDE:
EASY
SETUP:
(A) Prompter on camera.
(B) Prompter control unit.
Courtesy Telesync Corp.
Fig. 3 -4. Example of
a
prompter.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
99
of a single -phase ac rotor surrounded by a 3 -phase ac stator. Power to the
rotor is fed through slip rings. If the rotors of two such machines are fed
in similar phase with ac, transformer action causes a 3 -phase voltage to be
induced in the stators of the two machines. If these stators are connected
to each other (delta configuration) in the correct phase, 3 -phase currents
are established in the three connecting wires. If one rotor is turned through
90 °, the voltages induced in its stator are advanced or retarded by 90 °, and
this produces motor action in the other machine until it also has advanced
90 °. If this is done smoothly, the two machines will move exactly together.
If it is done abruptly, the second machine will, because of its inertia, tend
to overshoot the 90° point and then return. In order to reduce this inherent
action, a rotary damper is added to each receiver rotor shaft. It can be seen
from this necessarily brief description that if one rotor is driven by the
sync control transmitter, then the second rotor will revolve synchronously
and will be able to drive a prompter. If one rotor is held so that it cannot
revolve, the second rotor produces considerable resistance to any attempt to
revolve it. It is this feature that provides the lock between one or more
prompters and the sync control.
3 -5.
CAMERA -CHAIN POWER SUPPLIES
Camera -chain power supplies are normally rack-mounted units with
means of distribution to rack equipment, the camera control console, and,
through the camera cable, to the studio camera. Electronically regulated
supplies are universal.
The principle of a vacuum -tube regulator is shown in Fig. 3 -5. To review
this function:
If the load draws more current or if the ac input to the rectifier section falls, the result normally would be lower terminal voltage.
2. Resistor R1, tube V2, and gas regulator tube V3 are in series across
the rectifier filter output. Tube V3 holds the cathode of V2 at a con1.
Rl
V2
DC
Amplifier
Output From
Rectifier Filter
Voltage
0--.
R2
V1
Series Regulator
Adjust
Control
V3
Reference
Regulator
¡
Fig. 3 -5. Basic vacuum -tube regulator circuit.
Termina Voltage
(Load Connection)
loo
TELEVISION BROADCASTING CAMERA CHAINS
stant positive potential with respect to ground. The setting of R2
determines the bias on dc amplifier V2.
3. A reduction in terminal voltage results in a more negative bias on
V2, less current through V2, and hence less current through R1.
4. The decreased IR drop across R1 results in less negative bias on the
series tube (V1). This, in turn, results in lowered series resistance;
hence, it increases the terminal voltage to overcome the initial decrease in Step 1.
Most newer regulated power supplies employ transistors and (in some
cases) zener diodes. The circuit shown in Fig. 3 -6A is a common version
of the basic emitter- follower (or common -collector) circuit. The powersupply load (symbolized by a variable resistance) is placed in series with
Unregulated
Unregulated
Voltage Source
Voltage Source
Reference
Voltage
Load
(A) Stabilized against load
(B) Stabilized against load and
variations only.
source variations.
Unregulated
Voltage Source
Reference
Voltage
(C) Further stabilization by error
amplification.
Differential
Amplifier
Load
Fig. 3 -6. Basic transistor voltage regulators.
a transistor. The impedance of this transistor is controlled automatically
in such a way that it tends to compensate for impedance changes (or cur-
rent changes) in the load, thus maintaining an essentially constant voltage
across the load. This action may be explained by noting that the voltage
drop across the emitter -base junction of a transistor is usually negligible
in comparison with the supply voltage (at least over reasonable operating
ranges) , so the emitter tends to remain near the potential established by
the voltage divider in the base circuit. Since the base current is only a small
fraction of the emitter (or load) current, the base voltage is not altered
significantly by changes in load current, provided the resistors in the voltage
divider are not too large.
An alternative approach to the explanation of the regulating action of
Fig. 3 -6A is to point out that the output impedance of an emitter follower
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
101
is inherently low, and it approaches the impedance of the emitter -to -base
junction alone as the base impedance decreases to zero. The output impedance never decreases to zero, however, so the regulation never becomes
perfect with this simple circuit.
Note in passing that a pnp regulating transistor is more conveniently
placed in series with the negative side of the load, rather than the positive
side as would be the case with most vacuum -tube regulators. The transistor
itself must, of course, be capable of handling the maximum load current.
In practical transistorized power supplies, it frequently is necessary to
mount the large series regulators on radiators or other types of heat sinks
to keep the temperatures of the transistor junctions within safe limits.
While the simple circuit in Fig. 3 -6A is reasonably effective in stabilizing
the output voltage against load variations, it does not remove variations
caused by voltage changes in the unregulated source. This is because the
voltage on the base of the transistor is changed in proportion to the unregulated voltage. The circuit in Fig. 3 -6B overcomes this problem through
the use of a separate, stabilized reference voltage source at the base. Although a battery symbol is shown, the reference -voltage source in a practical
circuit could be a reference diode ( zener diode) , which is a semiconductor
diode with enough reverse bias to operate in the breakdown region. A
diode operated in this manner behaves very much like the familiar glow
tube, or gaseous voltage regulator; that is, the voltage drop across the device
is essentially independent of the current over a rather wide range. Reference
diodes are preferable to gaseous voltage -regulator tubes for most transistorized power supplies, because they operate at lower voltages ( usually
from 5 or 6 volts up to 60 volts or more) and because they are generally
superior in stability and inherent regulation.
The degree of regulation attainable with the circuit in Fig. 3 -6B is
determined by the emitter-to -base impedance of the transistor itself, which
might be of the order of a few ohms. Even better stabilization (or lower
output impedance) can be provided by the use of additional gain in the
control circuit to supplement the gain of the regulating transistor itself.
Such an approach is illustrated in simplified form in Fig. 3 -6C. The voltage
across the load may be compared with a stabilized reference voltage in a
differential amplifier, which can be designed with enough gain to make the
voltage variation at the load as small as required.
In a reverse-biased junction diode ( zener diode) , at a certain value of
reverse -bias voltage the current increases rapidly while the voltage across
the diode remains essentially constant. This breakdown voltage, which may
be any value between 2 and 60 or more volts, depends on the construction
of the diode. This characteristic is similar to that of the gas -tube regulator,
which begins conduction at a certain voltage and continues to conduct
varying amounts of current while maintaining constant voltage between the
elements. The zener diode is used in transistorized regulated power supplies
to hold an element of the transistor at a given reference voltage.
102
TELEVISION BROADCASTING CAMERA CHAINS
A basic diagram of a constant -voltage regulated power supply is shown
in Fig. 3 -7. A regulated reference voltage is obtained from a full -wave
rectifier; a bridge rectifier supplies the series -regulator transistor, which
receives at the base a feedback voltage from a comparison circuit. This
voltage is an error signal of such magnitude and polarity as to change the
conduction of the series regulator, hence changing the current through the
load resistor until the output voltage (E0) equals the voltage across the
voltage control (REO) . Note that since the series regulator is in the positive
side of the circuit, an npn transistor is used.
AC
Input
Series Regulator
Regulated DC
IEOI
Summing
Point
REO
Front-Panel
Voltage- Adjust
Control
Fig. 3 -7. Basic regulated power supply for constant voltage.
The difference between the two voltage inputs to the differential amplifier (Q1A and Q1B) is held at zero by feedback action. Thus, the voltage
across summing resistor Rs is held equal to the reference voltage. The current through Rs and the output -voltage control (REO) is termed the programming current, I. The input impedance of the differential amplifier is
high, and essentially all of the current (Ir) through Rs also passes through
REO. Because the voltage across Rs is constant, I is constant. Since REO
is variable, the output voltage is directly proportional to the resistance of
this control. Thus, the output voltage is the same as the voltage drop across
REO and will become zero if this control is reduced to zero ohms. This
variable control is sometimes in series with a fixed resistor to hold the output voltage to a given minimum value.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
103
Most reference supplies for units employing npn power transistors, as
in Fig. 3 -7, are referred to the positive output (or positive sensing, to be
described) as a circuit common. The reference auxiliaries for supplies using
pnp power transistors are referred to the negative output terminal.
The differential amplifier contains matched transistors (Q1A and Q1B)
placed in thermal proximity or contained in a single "chip." This markedly
improves the drift performance of the supply.
We can follow the feedback regulating action by assuming that the regulated dc output has momentarily increased. We will regard the positive
output terminal as "common "; then the output- voltage increase causes the
summing point to become instantaneously more negative. The resultant
decrease in current through Q1A causes its collector to become more positive; hence, a more negative voltage is applied through the inverting feedback amplifier ( -A) to the base of the series regulator. The resultant decreased conduction of the series regulator reduces the output voltage by an
amount proportional to the momentary increase, and the error voltage between the bases of Q1A and Q1B is reduced to zero.
Camera -chain regulated power supplies are never as simple as the basic
diagram of Fig. 3 -7. For even moderate power outputs, the dissipation requirement for the series- regulator circuit is such that multiple transistors
sometimes are used in parallel to provide adequate power- handling capacity.
Most recent power supplies employ some means of preregulation in the
rectifier path. The purpose of a preregulator is to allow the rectifier output
to change in coordination with the output voltage so that minimum voltage
drop occurs across the series regulator, and power dissipation is reduced
to a small value in all series regulator elements. Silicon controlled rectifiers
(SCR's) usually are used in the preregulator so that firing time can be controlled for required conduction angles.
Another feature found in modern supplies is the use of two extra wires
between the supply and the load (Fig. 3 -8) This results in optimum regulation at the load terminals rather than at the power-supply output terminals, compensating for the IR drop across the resistance of the wire. The
current through the sensing lead is so small that, in spite of the resistance
.
Power
Power
Supply
Supply
- Sensing Lead
Change in
E0
Change in E0 Not
0
Equal to Zero With
Wire
Resistance
in Cable
Change in
Change in Lead
E0
Current Because of
Drop Across Wire
Resistance in Cable
AA
Sensing Lead
(A) Sensing at supply.
(B) Remote sensing.
Fig. 3 -8. Principle of remote sensing.
O
104
TELEVISION BROADCASTING CAMERA CHAINS
of these leads, the voltage drop is negligible. This automatic arrangement
eliminates the need for an adjustable tapped-transformer switch in the
camera head to compensate for a change in cable length.
A block diagram of a power- supply module incorporating preregulation
and remote sensing is shown in Fig. 3 -9. Basic analysis by blocks is as
follows:
Bridge Rectifier: The bridge rectifier provides full -wave rectification of
the ac from the power transformer.
Switching Preregulator: The switching preregulator switches on and off
twice during each half cycle to charge the filter capacitors. It maintains
across the series regulator a small voltage drop that varies little with ac line voltage changes. Thus, it is possible to reduce the power dissipated in the
series regulator to the minimum required for adequate ac ripple filtering.
Preregulator Control: The preregulator control determines the conduction angle of the switching preregulator by comparing the full -wave rectified output from the bridge rectifier with the output from the series
regulator.
Slow Turn-On: The slow turn -on circuit causes the output voltage to increase from zero to the rated output in about one second. This gradual
build -up of dc reduces possible damaging current and voltage surges within
equipment that obtains power from this unit, as well as high current surges
within the supply itself.
Power
Supply
upply A
Power Transformer/
I
1
AC
Line
Three Secondary
Windings
r AC
tt
Power
Supply
Preregulator
Control
Voltage
Doubler
8
DC
B
Bridge
Switching
Rectifier
Preregu lator
Series
Regulator
-
Volts
m
s
Z
a
-12.5
Slow
Filter
Turn -On
Capacitors
Differest'aI
Amplifier and
Reference
V
_F-Sensing
+12.5
V
Sensing
Overload and
Overvoltage
Volts
Protection
Courtesy RCA
Fig. 3 -9. Block diagram of power -supply module.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
105
Filter Capacitors: The filter capacitors are charged twice during each half
cycle through the switching preregulator. During the time interval between
charges, the energy that has been stored is discharged through the series
regulator to the load.
Voltage Doubler: The voltage doubler provides the current source for
the control -signal input to the series regulator. The use of the voltage
doubler greatly reduces ripple on the dc output and minimizes the change
in output voltage with changes in the ac line input. The voltage doubler
also provides power to operate the switching preregulator.
Series Regulator: The series regulator filters out almost all of the ripple
that appears across the filter capacitor. It compensates for changes in line
voltage and load current; it also compensates for moderate voltage drops
across the resistances of connector contact and intrarack wiring.
Differential Amplifier: The differential amplifier and reference provide
the control signal for the series regulator. A differential amplifier rather
than a single -ended amplifier is used because of its inherent temperature
stability. Since the supply voltage must be extremely accurate, a precision
reference is used instead of a voltage-adjustment control.
Overload and Overvoltage Protection: The overload -protection circuit
prevents component damage, especially to transistors, because of an overload or short -circuited output. This circuit does not have to be reset after
an overload; the output returns to its correct voltage as soon as an overload
or short circuit is removed. The overvoltage- protection feature prevents
the series regulator and load from being damaged (as a result of excessive
power dissipation) in case the switching preregulator should develop a
short circuit.
Fig. 3 -10 illustrates the basic idea of overload protection. The value of
R8 is such that, during normal regulator operation, Q3 is saturated. This
places negligible resistance in series with the negative return. Potentiometer RE is adjusted so that it produces sufficient voltage drop to cause X1
to conduct if the load should develop a short or a specific value of overload
Overload Protection Circuit
Series
Ql Regulator
Load
Fig. 3 -10. Basic overload -protection circuit.
106
TELEVISION BROADCASTING CAMERA CHAINS
low resistance, excessive current) . Conduction of X1 reduces the bias on
Q3 so that it appears as increasing series resistance in the regulator circuit.
Relatively high voltages are required for the pickup tube or tubes in the
camera head. The well -regulated low voltages required for transistor circuitry (such as plus 12 and minus 12 volts) are used to supply a square wave oscillator. Usually, this oscillator is line -locked to one -half the horizontal scanning frequency so that any radio-frequency interference (RFI)
components that should escape the heavy filtering do not appear as beat
patterns in the picture.
A block diagram of the basic high -voltage supply is shown in Fig. 3 -11A.
This supply normally is located in the camera head. The binary counter
receives either camera horizontal -drive pulses or separated horizontal -sync
pulses from the composite camera blanking signal. Each input pulse is
differentiated to obtain a trigger for the binary stage. Thus, the square -wave
oscillator is synchronized to one-half the horizontal- scanning rate. This
prevents any ripple or transients from appearing in the active scan interval.
The square -wave oscillator is driven from cutoff to saturation. Since the
voltage supply to this stage is extremely well regulated, excellent regulation
of the generated high voltage is obtained.
The source of the high voltages is a step -up torroid transformer at the
oscillator output. All "low sides" of the voltages are tied together for a
common reference to insure tracking (Fig. 3 -11B) . The +1600 volts is
developed by doubler X1 -X2. The +800 volts is provided by half -wave
(
+12 V
-12 v
I
Blanking
15.75 kHz
High-Voltage
Binary
Divider
Osc
Filter
ifHI
1/2
-
+1600 V
+800 V
800 V
1H
(A) Block diagram.
X4
800 V
+800 V
1600 V
0.02
Osc
1!2
nn
f
High -Voltage
Common
High -Voltage
Tran former
(B) Filter and voltage divider.
Fig. 3-11. Typical camera -head high -voltage supply.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
107
rectifier X3, and the -800 volts is provided by X4. The current requirement is extremely small, allowing simple RC filtering. The slight alteration
in output voltage because of the load current is a negligible factor, since all
voltages to the pickup tube (s) track from the common reference.
3 -6.
CAMERA-CHAIN POWER DISTRIBUTION
Camera -chain power supplies normally are rack-mounted units from
which cabling extends to the camera control console and then to the camera
head. Fig. 3 -12 illustrates a specific example involving the RCA TK -11
( monochrome) camera, a WP -15 power supply, and the focus -current regulator. J2 and J3 are paralleled output receptacles on the WP -15 regulator
chassis. J2 supplies the required voltage to the focus -current regulator and
incorporates an interlock circuit. If the focus- current regulator becomes
inoperative, B+ is removed from the power-supply output by means of the
J2
WP 15
Regulator
Chassis
e
J3
9
10
11
12
O
0
0
0
9
8
10
0
0
11
12
0
0
0
NC
7
Focus -
Current
e
10
e
e
11
12
e
e
Rack
iG
J1
Equipment
Regulator
U
Terminal
2
3
Board
J
Cable
To AC Outlet
Camera
J3 Of
Camera
A
Control
8
19--9 19
9
10
11
9
12
Fig. 3 -12. Power distribution to camera control unit.
lControl
fChassis
(Console)
108
TELEVISION BROADCASTING CAMERA CHAINS
interlock relay. A customer-installed terminal board collects the outputs of
the focus-current supply and the power supply, and directs these to the
camera control console through the power cable.
Fig. 3 -13 gives the complete picture of the power distribution in the
TK -11 camera chain. The incoming power at J3 on the camera control unit
is routed to the master monitor through an individual power cable, and to
the camera through the camera cable. Some of the indicated voltages in the
power cable are from internal supplies. For example, -500 volts is taken
from the high -voltage supply in the camera head, and is fed through conductor 21 of the camera cable back to the control unit to be used for image
focus control ( photocathode potential) . It is fed back to the camera head
through conductor 15 of the camera cable.
ALTERNATE EQUIPMENT
CONNECTIONS
MUM
__
=I:;idR/nYIAMTl GdS?IJi.L\+-w
fH:T.TPI`-.1
12
-CONO. SHIELDED
POWER CABLE IMI
L
-801
WP-I5
POWER SUPPLY
RECT CHASSIS
I
MI- 6087
f -_ .-1bÌÌ 1J2
J
I
WP-I5
M
cLizi
439r
L
.26068
J2
7
'
B
REG.
I
IIII
Fig. 3-13. Interconnection diagram,
109
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
Typical voltage distribution for a modern color camera chain is shown
in simplified form in Fig. 3 -14. The main power supplies are rack mounted,
and all operating and fixed voltage distribution is by way of the control
cable from the rack to the control panel and the camera cable from the rack
to the camera head.
Fig. 3 -15 represents a portion of a typical control panel for a four -channel color camera; voltage distribution is shown for the operating controls
designated. The master white -level control is common to all channels, and
the master chroma -level control is common to the three color channels.
Individual color -channel controls are level controls to permit proper white
balance on a neutral gray -scale ( chip) chart. Distribution is then made to
the respective circuitry in the camera head.
MASTER MONITOR
MI-26136-8
PIN
HMI- 26759 -14 COAXIAL CABLE
WITH PLUGS
12
43ma
I
¿;
:
Q1llErái7CTJfíQ
-
tT7t^i
FIMM1SP=111
1XaZIFCILLQ
LI.. 71:147
7i79MR.:774.IFat=
I UH:Ti>ffI".N
.CEE
surnum- mnzrawRtf
I
II
OR
'.7XII
II.NAIM
f7WT UU.4..
COAXIAL (OPTIONAL) ..
RG-59,'u
/
PICTURE OUTPUT-
COAXIAL
NO
BLAcx
S
REMOTE MONITOR -
RG-I
COLOR
TRACER
BODY
M
'MI
WIDE7LIIMI QM/
.Iliaillail
flEQ 41ISLAKI7>
7M
Ma
W 114E>ELM
J7
r
PIN
COLOR
TRACER
BODY
NO
- :fftlti}
r
JIORJ2
r-1
f
{I
POWER CABLE IMI
2
r'
JB
A,.
¡LI'J7
PROGRAM SOUND B
REMOTE IRIS MOTOR POWER
Aç
_J
J9
JO
STUDIO CAMERA
CONTROL
IRDICATION
ON
I1R7vHaC
,
INTV ROOM. DC V01 TAGE 1+1
ENGINEERING INTERCOM
I
MI- 26056 -A
Or
Ilr
3
56
B RFDTE
PROEO.T
I
J¡,
5
\
-26759 -61
WITH PLUGS
J
I
FIGOM DC VOLTAGE
1
I
VOUAGEI!]
LUFREN
STUDIO CAMERA
-
±0
TARGET
11
2r
117V AC
Mr
fN{ANG
13
J3
BEAM
OC
I
WLTAr.(
]
NM
1780V1
4y+$6
9
IT
10
RMREEVB71PLIEP!'OCVS
REGULATED FOCUS
UNREGULATED
TALLY RELAY
N6
ag
[[
IT
)
CURìENTII
1
19
31
II
I20
_12J
I
JIOORJII
;
t
;
J2
JSOR 6
r
21r
-500
.20
VOLS
HORIZONTAL DRIVING-COAXIAL
:V9
BfS54'
i23_,
iRY.ñ''
1
-A2
AND
VIEW FINDER
MI- 26016 -A
15
7
B
M1- 26011
CARA
CABLE
(MI-9001
<CV
s
o
8 -COND. SHIELDED
ROR.
:p61IN
RG-59/U
TO
S
ATION
SYNC. GENERATOR
1:
-[[TO
INTERCOM. CABLE
IMI -821
STUD
SW TCHING CONTROL
Courtesy RCA
TK -1
1
A camera system.
110
TELEVISION BROADCASTING CAMERA CHAINS
Control Room
Rack
Studio
Monitors
Processing Units,
Encoder and
Output Amplifiers
[Picture
f
Power Supply
I
Waveform,
Camera
Power Supply
Control Panel
(Operating Position)
Control
- fCable
Voltage
Module
Camera Cable
Focus
Current +12.5
-12.5
V
V
-w-
+
-
Sensing Voltage
Sensing Voltage
-
-+
High Voltage
+12.5 V
-12.5 V
+ DC Ref
Voltage
r +11.0V
- DC Ref Voltage
rw-11.0V
Distribution to Each Module
Within Camera Head and Viewfinder
Fig. 3 -14. Typical
voltage distribution in color camera chain.
As shown in Fig. 3 -14, the camera head normally contains a voltage
module that receives the regulated voltage and serves as a main distribution
point for all other modules in the camera. Fig. 3 -16 illustrates basic reference- voltage generators in the voltage module for supplying other modules
in the camera. Note that in Fig. 3 -16A the regulated +12.5 volts is divided
to + 11.8 volts at the base of transistor Ql. Since the transistor is silicon
White-Balance Controls
12,5V
Master
White Level
From Rack
Master
Chroma
Green
Red
}
Power Supply
Throug h
Control Cable
Gnd
Luminance
Channel
Gnd
To
Luminance Channel
To
Green Channel
a
Channel
e
To Red
To
Blue Channel
T
Return to Rack Through Control Cable,
Then to Camera Head Through Camera Cable.
Other controls normally are
Master Black Level
Black Balance Controls
Registration (Horiz and Vert Centering Controls, Each Channel)
Lens Iris Control
Fig. 3 -15. Portion of control panel for four -tube color camera.
Blue
111
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
+11.8
V From
-11.8 V From Voltage Divider
Voltage Divider
Regulated
-12.5 V
Regulated
+12.5V
R1
Q1
Ql
Silicon
Silicon
NPN
PNP
+11.2V Ref
2200
R2
1%
-112V
To
Ref To
Other Modules
2200Q
Other Modules
in Camera Head
1%
in Camera Head
(A) Positive voltage.
(B) Negative voltage.
Fig. 3 -16. Basic reference -voltage generators.
npn, the emitter is at 11.8 -0.6, or + 11.2, volts for reference -voltage distribution. The circuit for the negative reference-voltage generator (Fig.
3 -16B) is identical, except that a silicon pnp transistor must be used.
Circuitry in all other individual modules is decoupled from both the
main power supply and the camera -head voltage module by decoupler
circuits. Basic negative -voltage decoupler circuits are shown in Fig. 3 -17.
( Positive decouplers are identical except that npn transistors are used)
The circuit of Fig. 3 -17A employs a regulating zener diode for the base voltage reference. The circuit of Fig. 3 -17B uses the common reference voltage supply from the camera -head voltage module.
.
Regulated
-12.5 V
Silicon
O
E)
10.6 V
Zener
-10 Volts to Module Circuits
(A) Reference from zener diode.
Regulated
-12.5
V
Ql
Silicon
-10.6 Volts to Module Circuits
I
-11.2 V
Reference
(B) Reference from generator.
Fig.
3 -17.
Basic module decoupler circuits.
TELEVISION BROADCASTING CAMERA CHAINS
11 2
3 -7.
POWER- SUPPLY MAINTENANCE
Power supplies, in addition to the overvoltage and overcurrent protection circuitry described previously, often include thermal relays and fuses.
Thermal -Relay Shutdown: The thermal relay opens the input circuit
only when the power-supply output current exceeds the current rating
specified for the operating ambient temperature. When the temperature
decreases to normal, the thermostat will reset automatically. If this occurs
often, forced-air cooling may be required.
Shutdown from Blown Fuse: Fatigue failure can occur as a result of
mechanical vibrations combined with thermally induced stresses that
weaken the fuse metal. Many times, fuse failures can be caused by a temporary condition, and replacing the blown fuse will make the fuse -protected
circuitry operative again. Never replace a fuse with the unit turned on
( power applied) The resulting temporary loose connection before solid
contact is made may open the fuse again through no fault of the equipment.
Always inspect fuse holders for tightness and cleanliness. Never substitute
a fuse with a rating higher or lower than the original rating.
When a power supply shuts down from causes other than a thermal relay
or blown fuse, it is necessary to determine whether the fault is internal to
the supply or in the load. Suitable dummy loads should be made up and
kept available so that this problem can be solved readily. Fig. 3 -18 illustrates
the use of a dummy load for such checks. The plus and minus sensing
terminals should be connected to their respective outputs for internal sensing. Automotive -type lamps, such as the type 1073 (1.8 amperes at 12.8
volts) , serve as excellent substitute loads. As many as needed should be
paralleled to approach the maximum or nominal load of the supply. For
example, if the nominal load is 6 amperes, three such lamps in parallel are
quite suitable.
A typical power supply of this kind might have a nominal load of 6
amperes, an overload -protection limit of 10 amperes, and an overvoltageprotection limit of 16.5 volts. With the dummy load substituted for the
normal load, an internal fault will cause the lamps to burn brighter than
normal for an instant, and then the supply will shut down. The higher.
Barrier
S
or Plug
+
+
rip
Automobile Lamps,
Paralleled as Necessary
12 -Volt
Out
+
Out
+ Sens
Sens
Common
- Sens
i
-
Out
- Sens
- Out
Substitute Load
(A) Single supply.
(B) Dual supply.
Fig. 3 -18. Dummy loads for power supply.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
113
than -normal voltage is caused by lack of regulation in the supply. If the
lamps light at the normal output voltage and remain lighted, the trouble
is obviously in the load. In the case of the dual power supply of Fig. 3 -18B,
usually only one of the supplies or loads will be faulty. Thus, if the trouble
is internal, one bank of lights will glow brightly and then go out. If all
lamps remain lighted, the trouble is isolated to the load.
In the case of a 280 -volt supply, a power resistor of the proper value to
bring the power- supply output current near maximum normally is used as
a dummy load. For example, if the specifications call for a maximum load
current of 1.5 amperes, two 500 -ohm, 200 -watt resistors connected in parallel should be used. Such dummy loads should always be available and ready
to connect to the barrier strip or receptacle at the power- supply output.
Older tube -type regulated supplies without modern protective circuitry
do not shut down automatically in case of trouble. The output voltage
simply wavers in value around nominal voltage. Such a supply sometimes
can be out of regulation with almost no noticeable shift on an external
voltmeter. This condition is best checked with an oscilloscope set on ac
input at high sensitivity. The change in dc on the coupling capacitor in the
scope will show up as a "bouncing" line on the free -running scope trace,
usually with a large ripple component.
In the case of intermittent or unusual camera problems which seem to
"chase themselves back" to some kind of power- supply problem, use of the
scope is mandatory. However, certain precautions must be observed to make
such tests valid. The importance of proper connection of load and monitoring leads to the power- supply output terminals cannot be overemphasized,
since the most common errors associated with the measurement of power supply performance result from improper connection to the output terminals. Failure to connect the monitoring instrument to the proper points
will result in measurement of the characteristics not of the power supply,
but of the power supply plus the resistance of the leads between its output
terminals and the point of connection. Even connecting the load by means
of clip leads to the power- supply terminals and then connecting the monitoring instrument by means of Clip leads fastened to the load clip leads can
result in a serious measurement error. Remember that the power supply
being measured probably has an output impedance of less than 1 milliohm,
and the contact resistance between clip leads and power- supply terminals
will, in most cases, be considerably greater than the specified output impedance of the power supply.
All measuring instruments (oscilloscope, ac voltmeter, differential or
digital voltmeter) must be connected directly by separate pairs of leads to
the monitoring points. This is necessary in order to avoid the mutual
coupling effects that may occur between measuring instruments unless all
are returned to the low- impedance terminals of the power supply. Twisted
pairs (in some cases shielded cable will be necessary) should be used to
avoid pickup on the measuring leads.
114
TELEVISION BROADCASTING CAMERA CHAINS
Care must be taken that the measurements are not unduly influenced by
the presence of pickup on the measuring leads or by power -line frequency
components introduced by ground -loop paths. Two quick checks should be
made to see if the measurement setup is free of extraneous signals:
1.
Turn off the power supply and observe whether any signal is observable on the face of the CRT.
2. Instead of connecting the oscilloscope leads separately to the positive
and negative sensing terminals of the supply, connect both leads to
either the positive or the negative sensing terminal, whichever is
grounded to the chassis.
Signals observable on the face of the CRT as a result of either of these tests
are indicative of shortcomings in the measurement setup.
In measuring the input voltage, it is important that the ac voltmeter be
connected as closely as possible to the input ac terminals of the power
supply so that its indication will be a valid measurement of the power supply input, without any error introduced by the IR drop present in the
leads connecting the power- supply input to the ac line -voltage source.
Use an autotransformer of adequate current rating. If this precaution is
not followed, the input ac waveform presented to the power supply may be
severely distorted, and the rectifying and regulating circuits within the
power supply may be caused to operate improperly.
A regulated power supply beginning to "slip" in performance usually
has an increased amount of ripple over that specified by the manufacturer
as the maximum. Fig. 3 -19A shows an incorrect method of measuring peakto -peak ripple. Note that a continuous ground loop exists from the third
wire of the input power cord of the power supply to the third wire of the
input power cord of the oscilloscope. This path is through the grounded
power- supply case, the wire between the negative output terminal of the
power supply and the scope, and the grounded scope case. Any ground current circulating in this loop as a result of the difference in potential (EG)
between the two ground points causes an IR drop that is in series with the
scope input. This IR drop, normally having a 60 -Hz (line -frequency) fundamental, plus any pickup on the unshielded leads interconnecting the
power supply and scope, appears on the face of the CRT. The magnitude
of this resulting noise signal can easily be much greater than the true ripple
developed between the output terminals of the power supply, and can completely invalidate the measurement.
The same ground-current and pickup problems can exist if an rms voltmeter is substituted in place of the oscilloscope. However, the oscilloscope
display, unlike the meter reading, tells the observer immediately whether
the fundamental period of the signal displayed is 8.3 milliseconds (1/120
Hz) or 16.7 milliseconds (1/60 Hz). Since the fundamental ripple frequency present at the output of a supply is 120 Hz (as a result of full wave rectification) , an oscilloscope display showing a 120 -Hz fundamental
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
115
component is indicative of a "clean" measurement setup, whereas the
presence of a 60 -Hz fundamental usually means that an improved setup
will result in a more accurate (and lower) measured value of ripple.
External Scope
Chassis
Power - Supply
Chassis
AC Hot
AC Hot
AC Common
AC Common
Gnd
Gnd
RG
IG,
-NW
EG
(A) Incorrect method.
Power-Supply
Chassis
AC
External Scope
Chassis
AC Hot
AC Common
Gnd
.R:]
Gnd Broken by
3 -to -2
Adapter
Shield Connected
to Scope Gnd Only
(B) Single -ended scope.
Power-Supply
Rack -Mounted
Chas is
Scope Chassis
Shield Connected
to Scope Gnd Only
(C) Differential scope.
Fig. 3 -19. Ripple measurements with oscilloscope.
Fig. 3 -19B shows a correct method for using a single -ended scope to
measure the output ripple of a constant -voltage power supply. The ground loop path is broken with a 3 -to -2 adapter in series with the ac line plug
of the power supply. Notice, however, that the power-supply case still is
connected to ground through the power- supply output terminals, the leads
connecting these terminals to the scope terminals, the scope case, and the
third wire of the scope power cord.
116
TELEVISION BROADCASTING CAMERA CHAINS
Either a twisted pair or (preferably) a shielded two -wire cable should
be used to connect the output terminals of the power supply to the verticalinput terminals of the scope. When a twisted pair is used, care must be
taken that one of the two wires is connected both to the grounded terminal
of the power supply and the grounded input terminal of the oscilloscope.
When shielded two -wire cable is used, it is essential for the shield to be
connected to ground at one end only so that no ground current can exist
in this shield and induce a noise signal in the shielded leads.
To verify that the oscilloscope is not displaying ripple that is induced in
the leads or picked up from the grounds, the plus scope lead should be
touched to the minus scope lead at the power-supply terminals. The ripple
value obtained when the leads are in this position should be subtracted
from the actual ripple measurement.
In most cases, the single -ended scope method of Fig. 3 -19B will be
adequate to eliminate extraneous components of ripple and noise so that
a satisfactory measurement may be obtained. However, in more stubborn
cases, or in measurement situations in which it is essential that both the
power-supply case and the oscilloscope case be connected to ground (e.g. if
both are rack-mounted) , it may be necessary to use a differential scope
with floating input, as shown in Fig. 3 -19C. If desired, two single -conductor
shielded cables may be substituted for the shielded two -wire cable with
equal success. Because of its common -mode rejection, a differential oscilloscope displays only the difference in signal between its two vertical -input
terminals, thus ignoring the effects of any common -mode signal introduced because of the ac difference in potential between the power- supply
case and scope case. Before a differential -input scope is used in this manner, however, it is imperative that the common -mode-rejection capability
of the scope be verified by shorting together its two input leads at the
power supply and observing the trace on the CRT. If this trace is a straight
line, the scope is properly ignoring any common -mode signal present. If
the trace is not a straight line, the scope is not rejecting the ground signal
and must be realigned in accordance with the manufacturer's instructions
until proper common -mode rejection is attained.
The complete hookup for checking a regulated power supply for comparison to manufacturer's specifications is shown in Fig. 3 -20. Most modern 12.5 -volt regulated supplies maintain the rated output voltage within
plus or minus 1 percent up to maximum rated load with an ac line input
of from 90 to 130 volts. The most severe test is at maximum output current with minimum line -voltage input. The ripple voltage is normally
around 5 millivolts peak-to -peak minimum for a 12.5 -volt supply. Note
that with the variable load, the overcurrent-protection circuitry can be
checked conveniently in this same setup.
The peak-to -peak ripple voltage and waveshape may be measured readily
with the oscilloscope, but some manufacturers give the ripple specification
in terms of rms voltage. Fig. 3 -21 shows the relationship between the peak-
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
+
117
Out
Scope
+ Sens
Variable
Autotransformer
- Sens
Out
AC Line
Reversed for
Grounded-Positive
Supplies
Fig.
3 -20.
Setup for checking power -supply performance.
to -peak and rms values of three common waveforms. The output ripple of
a dc power supply usually does not approximate the sine wave of Fig.
3 -21A; in many cases the output ripple has a waveshape that closely approximates the sawtooth of Fig. 3 -21B. In this case, the rms ripple is
1/3.464 of the peak-to -peak value displayed on the oscilloscope. The
square wave ( Fig. 3 -21C) is included because this waveshape has the
highest possible ratio of rms to peak -to-peak values. Thus, the rms ripple
and noise present at the output terminals of a power supply cannot be
greater than one -half the peak-to -peak value measured on the oscilloscope.
In most cases, the ripple waveshape is such that the rms value is between
one-third and one-fourth of the peak -to -peak value.
When a high - frequency spike measurement is being made, an instrument of sufficient bandwidth must be used; an oscilloscope with a bandwidth of 20 MHz or more is adequate. Measuring noise with an instrument that has insufficient bandwidth may conceal high- frequency spikes
that are detrimental to the load. The test setups illustrated in Figs. 3 -19A
and 3 -19B generally are not acceptable for measuring spikes; a differential
oscilloscope is necessary. Furthermore, the measurement concept of Fig.
3 -19C must be modified if accurate spike measurement is to be achieved.
2.828 Erms
Epk-pk
Epk
pk
Epk-pk
1
Erms
2.828
'
0.3535Epk-pk
(A) Sine wave.
Epk-pk 3.464 Erms
Epk-pk
Epk pk
rms
3, 464
(B) Sawtooth wave.
f
Epk-pk -
2
Erms
Epk-pk
Erms
0.5 Epk-pk
(C) Square wave.
Fig.
3
-21. Conversion of peak -to -peak to rms.
0.288
E
pk-pk
118
TELEVISION BROADCASTING CAMERA CHAINS
The Hewlett- Packard Company suggests the following procedure for
checking their regulated supplies for noise spikes:
1.
2.
3.
4.
5.
As shown in Fig. 3 -22, two coaxial cables must be substituted for the
shielded two-wire cable.
Impedance- matching resistors must be included to eliminate standing waves and cable ringing, and capacitors must be used to block
the dc current path.
The lengths of the test leads outside the coaxial cables are critical and
must be kept as short as possible; the blocking capacitor and the
impedance-matching resistor should be connected directly from the
inner conductor of the cable to the power-supply terminals.
Notice that the shields at the power- supply ends of the two coaxial
cables are not connected to the power- supply ground, since such a
connection would give rise to a ground- current path through the
cable shield, resulting in an erroneous measurement.
The measured noise -spike values must be doubled, since the impedance-matching resistors constitute a 2 -to -1 attenuator.
500 Termination
Differential
Oscilloscope
Power Supply
T
Connector
504 Termination
Fig. 3 -22. Arrangement for measuring noise spikes.
The circuit of Fig. 3 -22 also can be used for the normal measurement of
low- frequency ripple and noise. Simply remove the four terminating resistors and the blocking capacitors, and substitute a higher -gain plug-in
preamplifier in place of the wide -band plug -in module required for spike
measurements. Notice that with these changes, Fig. 3 -22 becomes a two cable version of Fig. 3 -19C.
It may happen that a camera chain loses transistors in certain modules
on a rather consistent basis. Sometimes this trouble is attributable to the
power supply. Modern regulated supplies have a slow turn -on and also a
certain time constant for turn -off to prevent excessive transients from
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
119
damaging delicate transistors. Transistors actually can be damaged upon
turn -off of the power supply; this damage obviously does not show up
until the equipment is turned on again.
The checking for on and off transients must be done with an oscilloscope, but many pitfalls exist in such measurement. All power- supply output terminals have a small amount of inductance, and the load can be
either capacitive or inductive. The best that can be done without laboratory type equipment and setups is to compare the on and off transients with
those of another camera chain that has not exhibited a problem from this
cause. When such a check does reveal a power supply with excessive on
or off transients, all filter time constants and all "overshoot" or transient
protection circuitry should be checked thoroughly.
The preventive maintenance of regulated power supplies is extremely
important to overall stability. The following four tests enable the maintenance engineer to keep a running check on the condition of his regulated
supplies, particularly the older tube -type variety:
Test 1. Determine the voltage output range at fixed load. Use a fixed
load that will draw at least two- thirds of the maximum rated load current.
Rotate the voltage-adjust control to its extremes, and record the minimum
and maximum voltages. For example, the normal available range of a
280 -volt regulated supply might be from 270 to 300 volts, at a given load
current. Failure to reach the normal maximum voltage is usually the result
of a weak dc-amplifier tube or voltage-adjust tube (or transistor).
Test 2. In vacuum -tube regulators, check the currents in the series regulator tubes for balance. Most regulated supplies incorporate a meter selector switch on the panel for measuring individual regulator -tube currents. Table 3 -1 shows the application of such readings. Notice that the
total load in this example is 1014 milliamperes; therefore, the ideal average for each of the six tube sections is 1014/6, or 169 mA. Since maximum tube life and stability can be expected when these currents are balanced within -±10 percent (20 percent total variation) , a record of individual currents is kept, and it is compared to the minimum and maximum
Table 3 -1. Tabulated Data for Test 2
Current (mA)
Tube
VIA
168
V1B
160
180
182
164
160
V2A
V2B
V3A
V3B
Operating Data
Total
=
1014 mA
Average/Section = 169 mA
Lowest Desirable = 152.1 mA
Highest Desirable = 185.9 mA
120
TELEVISION BROADCASTING CAMERA CHAINS
values that should occur for the given load. This indicates the need for a
tube change before trouble occurs, barring any sudden failure.
Test 3. Measure the input -voltage regulation (voltage output with fixed
load and varying input ac line voltage) . The setup for this and the following test is shown in Fig. 3 -23. Adjust the power supply to be tested
to 0.5 volt above the reference supply, and connect a voltmeter between
the two outputs to measure this voltage difference. By means of the variable autotransformer, make measurements at the reference line voltage
(usually 117 volts) and over a specified range, such as 100 to 130 volts ac
input. Table 3 -2 shows the data recorded at station WTAE -TV for an
RCA WP -15B supply. Note the excellent input voltage regulation of this
supply with a fixed load.
Difference Voltmeter
AC
AC Line
Variable
Autotransformer
Input
_y
B+ Reference
Power Supply B+
To Be Tested
Power Supply
rElectronic
-
Load
Select for
5000
200W } Suitable Load
Fig. 3 -23. Test setup for checking power -supply regulation.
Test 4. Measure the output -voltage regulation under varying loads (fixed
ac line voltage with varying load current) . In this test, the variable auto transformer is adjusted for an ac input of 117 volts, and the electronic
load on the supply under test is varied over a specified range. Table 3 -3
shows the corresponding data for the WP -15B of Test 3. Note that the
results indicate low internal power- supply resistance and good regulation
under varying loads.
Table 3-2. Tabulated Data for Test
Line
Volts
117 (Ref)
100
105
110
115
120
125
130
Output
Volts
280.5
280.94
280.75
280.62
280.5
280.42
280.37
280.31
3
Actual
Volts
Variation
+
+
+
--
0 (Ref)
0.44
0.25
0.12
0
0.08
0.13
0.19
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
121
Table 3 -3. Tabulated Data for Test 4
Load
Current
Output
Volts
(mA)
1000 (Ref)
400
600
800
1200
1400
1500
280.54
280.52
280.5
280.52
280.54
280.56
280.51
Actual
Volts
Variation
-+
-
0 (Ref)
0.02
0.04
0.02
0
0.02
0.03
NOTE: For stability in video levels, the associated power supply should
have very low internal resistance, theoretically zero. (This is never attained in practice.) The internal dc output resistance may be found as
follows:
AVo
Ro
AI,
where,
Ro is the dc output resistance,
AVo is the change in output voltage,
AIL is the change in output current.
For example, if the output voltage changes 0.1 volt with a load -current
change of 1000 mA (1 ampere)
:
Ro
=Oil = 0.1 ohm
The condition of power- supply filters and general regulation efficiency
should be checked several times yearly by observing the ripple content of
the output voltage on an oscilloscope. An increase in ripple content indicates the need for filter replacement or better regulation efficiency before
deterioration reaches troublesome proportions. Typical commercial power
supplies have a maximum of 2 millivolts peak-to -peak ripple content on
a 280-volt regulated output. The ripple on an unregulated voltage may be
as high as 2 volts (peak -to -peak) on a 400 -volt dc output.
The operating condition of a zener diode (or any other type of diode)
is most reliably checked with an oscilloscope and the simple associated
circuitry shown in Fig. 3 -24. The upper trace in Fig. 3 -25 illustrates a
typical curve obtained by this method. If desired, the forward trace may
be eliminated by means of an added silicon diode, as shown by the dash
lines in Fig. 3 -24. In this case, a curve such as the lower trace illustrated in
Fig. 3 -25 results.
Fig. 3 -26 is an interpretation of the trace in Fig. 3 -25. As the variable ac
voltage (Fig. 3 -24) is increased from zero, the voltage is traced horizon-
TELEVISION BROADCASTING CAMERA CHAINS
122
If Desired to Eliminate
Forward Trace
12 -14 Volts AC
(Variable)
Fig.
3
-24. Test setup for checking
semiconductor diodes.
tally (A -B) along the scope graticule, which may be calibrated in volts/
centimeter. When the zener breakdown voltage is reached, the horizontal
trace should remain the same length as the current curve increases. If desired, the vertical scale (B -C) may be calibrated in milliamperes /centimeter. Care should be taken not to exceed the maximum current specification ( wattage = voltage applied across the diode times the diode current) of the zener diode being tested.
Fig. 3 -25. CRO traces showing diode
operation in test circuit.
The same test circuit should be employed to check a regular diode at
the operating potential encountered in the circuit in which it is used. Some
diodes (such as the common 1N34) have a natural hysteresis loop, as
shown by the upper trace in Fig. 3 -27. This loop should remain stable
without jitter or erratic "looping" as the voltage is varied around the normal operating level. Other diodes (such as the Type 1N279) do not reveal
Volts
8
A
Current
C
Fig.
3 -26. Interpretation of
diode trace.
Fig. 3 -27. Traces with and without
hysteresis.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
123
a loop (lower trace in Fig. 3 -27) . There should be no instability of trace
as the voltage is varied around the normal operating level.
3 -8.
INTERCONNECTING FACILITIES
There are two basic types of camera interconnections: (A) Camera cable
between camera and rack or control units, with multiconductor cable between rack equipment and remote -control panels, and (B) Rf links, employing either air link or triaxial -cable link (Chapter 2) . In the second
method, multiconductor cable may be employed between base- station rack
equipment and remote -control panels.
Representative types of camera cable are shown in Fig. 3 -28. Monochrome camera cable usually consists of three coaxial conductors and 21
(Fig. 3 -28A) or 25 (Fig. 3 -28B) single conductors. In the typical application of a 24- conductor cable in Fig. 3 -13, note that some conductors are
paralleled for proper current capacity, and the three coaxial elements normally provide drive-pulse feed to the camera and carry the video output
current from the camera to the control unit. The surge impedance of the
coaxial portions of camera cable is 50 ohms.
For assembly purposes, conductors in a camera cable are divided into
groups. For the 24- or 28-conductor cable, the single wires are divided
into three groups. Together with the three coaxial conductors, these groups
are assembled around a waterproof jute core. The entire assembly is then
taped and covered with a woven shield over which are placed a cotton
braid and a neoprene outer jacket. Color cameras may require cables with
82 (or more) conductors, as shown in Fig. 3 -28C.
(A) 24 conductors.
(B) 28 conductors.
(C) 82 conductors.
Courtesy Belden Corporation
Fig.
3
-28. Examples of camera cables.
124
TELEVISION BROADCASTING CAMERA CHAINS
Courtesy RCA
Fig. 3 -29. Cable and connectors for RCA TK -42 color camera.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
=
2.25 Max
2.05
1. v5
1.85
0.80 Approx
11,_._
AÍ
o. zsol
I
....
i\`,
Dimensions In Inches
Coax
No. 14
No.
10
Strip 0.25
No. 22
(A) Male end of cable.
Solder Hole
Sweat shield to pin and sleeve, applying solder through hole.
(B) Male end of coax.
No.
22 Leads,
3.00 Max
Cut to Fit
2.7
2.6
2.35
Dimensions In Inches
Coax No. 14
No.
10
No. 22
Strip 0.25
(C) Female end of cable.
Shield
Sweat shield to pin and sleeve, applying solder
through hole.
(D) Female end of coax.
Fig. 3 -30. Preparation of cable ends.
Strip 0.18
125
TELEVISION BROADCASTING CAMERA CHAINS
126
Table
Pin
1
2
3
4
3 -4.
Conductors in TK -42 Camera Cable
Color
Pin
Wire
74
28
73
72
22
22
22
Wht- Blu -Orn
22
Wht -Blu -Brn
44
45
46
47
48
49
24
53
54
35
60
22
22
22
22
22
14
22
Wht- Yel -Vio
Wht -Blu
22
Tinned Copper
22
Wht- Yel -Orn
Wht- Yel -Yel
Wht- Yel -Blu
Wht -Vio
Wht- Blk -Blk
Wht-Grn -Red
Wht-Yel -Grn
Wht-Gry
Wht- Grn -Brn
Wht-Yel -Gry
Grn
Wht- Grn -Blk
Orn
Wht-Brn -Red
Wht -Blu -Red
22
Tinned Copper
6
71
22
7
42
40
39
22
22
22
Wht- Blu -Blk
Wht- Red -Vio
Wht- Red -Grn
Wht- Red -Yel
Wht- Blu -Blu
Wht- Blu -Yel
Wht -Blu -Grn
Wht- Grn -Gry
Wht- Red -Gry
Wht- Red -Blu
Wht- Brn -Brn
Wht- Red -Orn
Wht -Blu -Gry
Wht- Blu -Vio
22
22
Wht- Orn -Blk
Quad
Drain
22
Tinned Copper
21
22
Wht- Blk -Yel
18
Tinned Copper
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
Wht- Brn -Blk
Wht-Blk -Brn
Wht- Blk-Orn
Wht- Brn -Orn
Wht-Orn -Brn
Wht -Blk -Gry
Wht-Blk -Red
Wht- Blk-Blu
Wht- Orn -Orn
Wht -Orn -Red
Wht- Red -Red
Wht -Orn -Grn
Wht- Orn -Yel
Wht -Orn -Gry
Wht- Red -Brn
Wht- Orn -Blu
Wht- Orn -Vio
9
10
77
11
14
75
76
70
43
15
41
16
27
38
79
78
30
44
12
13
17
18
19
20
21
22
23
24
25
26
27
28
29
Color
AWG
Shield
Drain
8
AWG
Wire
Core
Drain
26
18
30
20
29
45
25
31
19
32
33
34
35
36
23
47
37
38
48
52
46
37
49
39
36
40
50
41
51
42
43
55
22
22
22
22
22
22
22
22
22
22
22
Brn
50
51
Quad
Drain
56
Wht- Blk -Vio
Wht- Yel -Blk
Wht-Yel -Brn
Blu
62
63
64
65
66
67
68
34
62
32
33
66
65
69
68
69
31
70
67
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
71
7
10
Grn
72
73
74
75
76
77
78
79
80
NC
2
14
Wht
1
14
Blk
3
14
Orn
4
14
5
14
Blu
Brn
6
10
Red
10
81
11
Coax
Coax
Coax
Coax
Coax
Coax
52
53
54
55
56
57
58
59
60
61
82
83
84
85
57
59
15
17
64
58
16
63
61
Yel
Wht- Grn -Yel
Wht- Grn -Orn
Wht -Grn -Vio
Wht- Grn -Blu
Red
Wht- Grn -Grn
NC
9
12
8
13
Wht -Red
Wht-Orn
Wht-Brn
Wht-Yel
Wht-Blk
Wht-Grn
Wht -Yel -Red
Wht -Blk -Grn
Courtesy RCA
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
Table
3 -5.
1
2
3
4
5
6
7
8
9
10
11
12
13
Conductor Functions in TK -42 Camera Cable
Function
Pin
Spare
Tally
Spare
Spare
Shield Drain
Spare
Gamma
Mono Blk. Level Relay
Lens Cap
Spare
Test Switch
White Level Cont
Spare
15
Aperture
Hor. Advance
16
Red Hor. Cent.
17
Blue Vert. Cent
+12.5V Sense
Black Level Cont
Spare
14
18
19
20
21
22
23
24
25
26
27
28
29
30
Power Switch
Quad Drain
Interphone Prod Coil
Core Drain
V.F. Relay (M- Blanker)
Spare
Interphone- Common
Sensitivity Sw
Spare
+250
V
31
Interphone -Eng. Coil
32
33
34
35
36
37
38
39
40
50/60 Hz Sel.
41
42
43
127
Spare
Spare
V.F. Relay K2
Green Selector
V.F. Relay K5
Blue Selector
Red Selector
Spare
R.B.G.M. Sel. Corn
M Selector
A.C. Sen.
Pin
Function
44
45
46
47
48
49
50
A.C. Sen.
Spare
Brg. Cap
Polarity Sw
Green White Level
V.F. Relay (Color Blanker)
Quad Drain
Blue White Level
Blue Black Level
Red White Level
Spare
Interphone -Cue Lo.
Blue Hor Cent.
Red Black Bal.
Interphone -Cue Hi
-12.5V Sense
Time Const. Relay
Green Hor Cent.
Red Vert. Cent.
Spare
Spare
Relay Gnd. Return
Green Vert. Cent.
Sensitivity Cont.
D.C. Gnd Sense
Spare
Mono White Level
12.5V
Spare
A.C. Reg.
A.C. Reg.
A.C. Unreg.
A.C. Unreg.
Gnd.
Spare
12.5V
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
+
-
Red Video
Mono Video
Blue Video
V.F. Video
Green Video
Camera Sync
Courtesy RCA
TELEVISION BROADCASTING CAMERA CHAINS
128
Snap-In
Pin Contact
Clamp
Clamp
Set
End Bell
Screw
Ferrule
Screw
Grommet
Shell With Snap -In
Monobloc Insulator
6
B
'I
1
1
oI
(A) Receptacle.
Barrel With Snap-In
Monobloc Insulator
Coupling
Nut
Set
Grommet
Ferrule
Screw
End Bell
Clamp
Screw
Snap-In
Clamp
Socket Contact
6
(B) Plug.
Contact
Inserted
(C) Contact insertion.
Insertion Tool
Courtesy ITT Cannon Electric
Fig. 3 -31. Connectors for camera cable.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
129
To acquaint the reader with a typical color-camera cable installation,
Figs. 3 -29 and 3 -30 and Tables 3 -4 and 3 -5 are included. At the top of
Fig. 3 -29, Section B -B shows a typical grouping of conductors by shielding
isolation and drain -wire combinations. The numbers in Section B -B are
"wire numbers" only, not connector pin numbers. Note that at the center
are four No. 14 AWG, tinned-copper, vinyl -plastie insulated, color -coded
conductors, along with three No. 22 AWG conductors (Table 3 -4). An
aluminum Mylar shield is wrapped around this group and a standard,
tinned -copper, No. 18 AWG core drain wire. Two quad groupings with
associated drain wires are contained in the cable, and the overall shield
drain wire is connected to pin 5 of the connector and in turn to frame
ground.
Fig. 3 -32. AMP -LEAF connector for
printed circuit boards.
Courtesy AMP Incorporated
Note that pins 72 and 78 of the connector have no wires attached, but
the pins still must be installed. The largest wires used are No. 10 AWG
( pins 71 and 79 ) ; these wires are used to carry the + 12.5 - and -12.5 -volt
power- supply current to the camera (Table 3 -5) Note from the correlation of Tables 3 -4 and 3 -5 that the core shielded group is employed for
ac, and the shielded quad groups are used primarily for interphone
(Chapter 8) .
Since the rack -mounted end of the camera cable receives power, the
receptacle at the rack is female, and the male end of the camera cable goes
to this receptacle. The camera head normally has a male receptacle to
receive the female end of the camera cable. Fig. 3 -30 shows the preliminary conductor preparation for each end.
.
130
TELEVISION BROADCASTING CAMERA CHAINS
Camera cables and multiconductor control cables normally are terminated in a plug or receptacle of the general type shown in Fig. 3 -31. There
are many individual types of connectors, and the camera technician should
obtain the correct assembly procedures for every type of connector he
may become concerned with in assembly or disassembly of cables. ITT
Cannon Electric, AMP Incorporated, and Amphenol all supply instruction
sheets for their respective connectors; these sheets specify proper crimping
tools to use and give all other necessary information. Sometimes such
sheets are included in the instruction books for camera chains.
Figs. 3 -31A and 3 -31B give the general nomenclature of plug and
receptacle parts. After the pin or socket has been crimped properly to the
conductor, it usually is inserted with a special tool, as illustrated in Fig.
3 -31C. It is most important that exact instructions be followed for the
particular assembly or disassembly involved. Special pin and socket extraction tools usually are available for disassembly.
(A) Studio termination.
(B) Patch rack for cameras.
(C) Patch rack for control units.
Courtesy WBBM -TV
Fig.
3
-33. Camera -cable connections at WBBM-TV.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
áw c..C.
IírhiWN......
. A c 1. ...«.
131
Cut end of cable even.
Remove vinyl jacket 1 1/8 ".
Bare 5/8" of center conductor.
Trim braided shield.
Slide coupling ring on cable.
Tin exposed center conductor and braid.
Screw plug sub- assembly on cable.
Solder assembly to braid through solder holes.
Use enough heat to create bond of braid to shell.
.
..,. i
i
Solder center conductor to contact.
final assembly, screw coupling ring on plug
subassembly.
For
'GIIIIVIIII
Courtesy Amphenol Corporation
Fig. 3 -34. Assembly of UHF
connector.
Cut end of cable even.
Remove vinyl jacket 3/4 ".
Slide coupling ring and adapter on cable.
Fan braid
POrtti
1-3/8
5/8-I
-re
^'^f
-Ui
ilíllI:II
l
1--..1,~n11111u,1
I
slightly and fold back
as
shown.
Position adapter to dimension shown.
Press braid down over body of adapter and
trim to 318". Bare 5/8" of conductor.
Tin exposed center conductor.
Screw plug subassembly on adapter.
Solder braid to shell through solder holes.
Use enough heat to create bond of braid
to shell. Solder conductor to contact,
final assembly, screw coupling ring on
plug subassembly.
For
Courtesy Amphenol Corporation
Fig.
3
-35. Assembly of UHF connector with adapter.
132
TELEVISION BROADCASTING CAMERA CHAINS
Individual modules of modern color camera chains normally employ a
quick connect-disconnect arrangement (Fig. 3 -32) for both rack -mounted
and camera-head boards. The particular arrangement shown is the AMP LEAF connector. The boards are slotted to mate with keying plugs that are
positioned in the connector housing. Thus, accidental insertion of a contact board other than in the specified position is prevented.
Since camera -control units and rack-mounted gear are normally in the
control room, which is some distance from the tightly enclosed studios,
the camera cable seldom connects directly from camera head to rack or
Trim jacket 19/64" for RC -58/U, 5/16" for
21/64" for RG -71 /U.
Fray shield and strip inner dielectric
Tin center conductor.
*1110)Dmi
Nut
1
RG -59/U
or
/8 ".
Taper braid and slide nut, washer, gasket, and clamp
over braid.
Clamp is inserted so that its inner shoulder fits
squarely against end of cable jacket.
Washer Gasket Clamp
With clamp in place, comb out braid, fold back smooth
as shown, trim 3132" from end.
3'32
Slip contact in place, butt against dielectric and solder.
Remove excess solder from outside of contact.
Be sure cable dielectric is not heated excessively and
swollen so as to prevent dielectric from entering into
connector body.
Female Contact
ra
I
IIIIIIIIII1
Ilr
Push assembly into body as far as it will go.
Slide nut into body and screw in place with wrench
until tight.
For this operation, hold cable and shell rigid and
rotate nut.
e
kam.--WW
Male Contact
Courtesy Amphenol Corporation
Fig. 3 -36. Assembly of BNC connector.
CAMERA MOUNTING, INTERCONNECTIONS, AND POWER SUPPLIES
133
control unit. The camera-head cable normally is connected to a receptacle
on a wall in the studio, as shown in Fig. 3 -33A. (Note how the outlets in
Fig. 3 -33A are tilted downward so that the cable reaches the floor with a
minimum of strain.) A patch rack sometimes is used, as shown installed
at WBBM -TV in Fig. 3 -33B. By the arrangement illustrated, seven cameras
can be made available to any one of fourteen outlets in four different studios. Fig. 3 -33C shows the patch rack for control units assigned to studio
distribution of the cameras. Note the portion of the camera-cable patching
rack on the left.
The preparation of miniature coaxial cable for plug -in module connectors or camera -cable connectors is a special technique for the particular
type of connector, as mentioned. However, the distribution cable (nominal surge impedance of 75 ohms) is usually RG /59U or the larger
RG /11U. Fig. 3 -34 shows the technique for assembling RG /11U cable
to the UHF -type (83 -1SP) connector. Fig. 3 -35 includes the adapter used
with the UHF plug for the smaller RG /59U cable. Fig. 3 -36 illustrates
the BNC connector assembly as normally used for RG /59U cable.
EXERCISES
Q3 -1.
Q3 -2.
Q3 -3.
Q3 -4.
Q3 -5.
Q3 -6.
Upon what does the camera head normally mount?
Name the basic controls found on a camera panning head, and describe the basic function of each.
What are "synced prompters "?
What is the purpose of the preregulator section of a regulated power
supply?
What is remote sensing, and why is it used in modern regulated
power- supply systems?
Why are oscillators in camera high -voltage supplies usually synchronized to one -half the horizontal- scanning frequency?
CHAPTER
4
The Camera Pickup Tube,
Yoke Assembly,
and Optics
In this chapter, consideration will be given to the more advanced characteristics of the pickup tube, yoke, and optical assemblies associated with
television cameras. To understand this material, the reader should have a
fundamental background in these subjects.'
We will consider the following characteristics of the image orthicon,
vidicon, and lead -oxide tubes:
1.
Sensitivity
2. Spectral response
3. Dark current
4. Gamma ( transfer characteristics)
5. Resolution
6. Signal -to -noise ratio
7. Lag
8. Dependence on target voltage
9. Temperature dependence
10. High-light handling
11. Operational techniques
4 -1. THE IMAGE ORTHICON
Fig. 4 -1 illustrates the three basic types of image- orthicon ( I.O.) tubes.
Fig. 4 -lA shows the 3 -inch non -field -mesh tube. In the 3 -inch field -mesh
type (Fig. 4 -1B), two electrodes, the field mesh and suppressor grid, are
'Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sains & Co., Inc., 1971),
Chapters 1 and 4.
134
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
Photocathode
Accelerator
Grid 6
Target
Decelerator Grid
Grid 5
135
Horizontal and
Vertical
Deflecting Coils
4
Camera
Focusing
Coil
Grid
3
Alignment
Electron
Gun
Coil
Five-Stage
Lens
Multiplier
/X
Scanning
/
Return Beam
Beam
REP
Grid 2 and
Televised
Scene
Image Section
Dynode
1
to
5
Multiplier Section
Scanning Section
(A) Non-field-mesh type.
Photocathode
Accelerator
Annular
Target
Base
Grid 6
Camera
Lens
Alignment
Focusing
Coil
Horizontal and
Vertical
Deflecting Coils
Dynodes
2
Four-Stage
Coil
Multiplier
Diheptal
Base
,
Scanning Beam
[lì
Return Beam
`.
\
Field-Correcting
Decelerator
Grid 5
Televised
Scene
Grid
Mesh
4
Electron
Grid 2 and
3
Dynode
Gun
1
Multiplier Section
Scanning Section
Image Section
Grid
Suppressor
Grid
(B) 3-inch field-mesh type.
Focusing Coil
Accelerator
Target
Grid 6
Pancake"
Focusing
Coil
Horizontal &
Decelerator
Vertical Deflection
Grid 5
Coils
Alignment
Grid
Coil
3
Five- Stage
Electron
Multiplier
Electron
Gun
Televised
Scene
Camera Lens
:11110111111111.0""'"
Grid
Photocathode
Field Mesh
Envelope Termina
I
4
Scanning Section
Image Section
(
Internal
Conductive Coating
Grid
2
and Dynode
1
Multiplier Section
(2) 41/2-inch type.
Courtesy RCA
Fig. 4-1. Basic construction of image orthicon.
136
TELEVISION BROADCASTING CAMERA CHAINS
added. In this tube, the additional electrodes normally are operated at a
fixed potential. The 41/2-inch tube (Fig. 4 -1C) always incorporates a field
mesh that is provided with a variable bias potential. Note also the addition of the "pancake" (magnifier) coil at the front of the tube. This magnifier coil is connected in series aiding with the main focusing coil, and
provides about 120 gausses in the plane of the photocathode. The useful
area on the photocathode is identical to that of the 3 -inch I.O. (1.6 -inch, or
40.7- millimeter, diagonal) . The pancake coil magnifies the image 1.5 times
(to a 2.4-inch, or 61- millimeter, diagonal) to fill the target.
The image orthicon consists of three primary sections: the image section, the scanning section, and the electron- multiplier section. A basic
review of these functions is provided in the following subsections.
Image Section
The image section contains a semitransparent photocathode on the inside of the faceplate, a grid to provide an electrostatic accelerating field, and
a target that consists of a thin glass disc with a fine -mesh screen very
closely spaced to it on the photocathode side. Focusing is accomplished
by means of a magnetic field produced by an external coil, and by varying
the photocathode voltage.
Light from the scene being televised is picked up by an optical lens
system and focused on the photocathode, which emits electrons from each
illuminated area in proportion to the intensity of the light striking the
area. The streams of electrons are focused on the target by the magnetic
and accelerating fields.
On striking the target, the electrons cause secondary electrons to be emitted from the glass. The secondary electrons thus emitted are collected by
the adjacent mesh screen, which is held at a definite potential of about 2
volts with respect to target cutoff voltage. Therefore, the potential of the
glass disc is limited for all values of light, and stable operation is achieved.
Emission of the secondary electrons leaves on the photocathode side of
the glass a pattern of positive charges that corresponds to the pattern of
light from the scene being televised. Because of the thinness of the glass,
the charges set up a similar potential pattern on the opposite, or scanned,
side of the glass.
Scanning Section
The side of the glass away from the photocathode is scanned by a low velocity electron beam produced by the electron gun in the scanning section. This gun contains a thermionic cathode, a control grid (grid 1) , and
an accelerating grid (grid 2 ) . The beam is focused at the target by the
magnetic field of an external focusing coil and by the electrostatic field produced by grid 4.
Grid 5 serves to adjust the shape of the decelerating field between grid 4
and the target in order to obtain uniform landing of electrons over the
137
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
entire target area. The electrons stop their forward motion at the surface
of the glass and are turned back and focused into a five -stage signal multiplier, except when they approach the positively charged portions of the
pattern on the glass. When this condition occurs, electrons are deposited
from the scanning beam in quantities sufficient to neutralize the charge
pattern on the glass. Such deposition leaves the glass with a negative charge
on the scanned side and a positive charge on the photocathode side. These
charges neutralize each other by conduction through the glass in less than
the time of one frame (Fig. 4 -2)
.
Mesh
+2 Volts
Glass
Direction of Scan
I
Stored Image Charge
(Immediately Before Scan)
Net Charge - 0
o
During Scan
-- ------_
--__-_ -
-
Scanning
'
'.\(\
Beam
Neutralizing Charge
Dy Scanning Beam)
(Deposited
After Scan
Positive charges move through glass
to neutralize electrons deposited by
beam in less than time of one frame.
Fig. 4 -2. Target charge and discharge cycle of image orthicon.
Alignment of the beam from the gun is accomplished by a transverse
magnetic field produced by an external coil located at the gun end of the
focusing coil. Deflection of the beam is accomplished by transverse magnetic fields produced by external deflecting coils.
The electrons turned back at the target form the return beam, which has
been amplitude modulated by absorption of electrons at the target in
accord with the charge pattern. (The more positive areas of the charge
pattern correspond to the high lights of the televised scene.)
One of the basic problems with the image orthicon is shown in Fig. 4 -3.
When the photocathode is excited by a very bright object in the scene, it
emits a very large number of photoelectrons. These electrons are attracted
toward the target because the target screen is at about +2 volts (above
cutoff) and the photocathode is at a much higher negative potential. When
the photoelectrons strike glass target B, secondary electrons are emitted
138
TELEVISION BROADCASTING CAMERA CHAINS
and attracted to target mesh A, where they normally are passed to ground
through the target -potential supply. In extreme high -light areas, the target
glass rapidly assumes a potential higher than that of the mesh, and collection of additional electrons on the mesh is prevented. These free electrons
return to the target in areas adjacent to the bright spot, there nullifying
the positive charges in areas C. Thus, the scanning beam is repelled in
these areas, and a large return beam results, Since this represents black, a
black halo surrounds the excessive high light in the image. We will see
later in this discussion how the combined effects of target area, target
spacing (capacitance), and the field mesh can minimize this effect.
Multiplier Section
The return beam is directed to the first dynode of a five -stage electrostatically focused multiplier. This multiplier utilizes the phenomenon of
secondary emission to amplify signals composed of electron beams. The
electrons in the beam impinging on the first dynode surface produce many
other electrons, the number depending on the energy of the impinging
electrons. These secondary electrons are then directed to the second dynode
and release more new electrons. Grid 3 facilitates a more complete collection by dynode 2 of the secondary electrons from dynode 1. The multiplying process is repeated in each successive stage, with an ever -increasing
stream of electrons until those emitted from dynode 5 are collected by the
anode and become the current in the output circuit. The multiplier section
amplifies the modulated beam between 500 and 1000 times.
Target Mesh
+2 Volts
A
(;)
o
i
f
ti
Very Bright Spot
- -t - -
-Og:
-
o
Fig. 4-3. Formation of halo.
Scanning
Large Amount of Return Beam
Because of Negative Charge
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
139
The signal -to -noise ratio of the output signal from the image orthicon
multiplier raises the output signal sufficiently
above the noise level of the video- amplifier stages that these stages contribute a negligible amount of noise to the final video signal. The signal -tonoise ratio of the video signal, therefore, is determined only by the random
variations of the modulated electron beam.
is high. The gain of the
Anode
47k
I
4 7k
Dynode
Focusing Coil
Signal Output
lo Video Amplifier
.1250V
II000¡
0330
Alignment
Focusing -Coll
Current
Coils
0 03
0.05
000
2W
-4
Regulator
lip firs
5___
V
2000
2W
100k
0.01
220k
100k
Dynode
Grid
3
PC
200k
51k
0.001
2
200k
150k
1501
Grid
270k
Dynode
IW
4
75k
Type 5820
Dynode
3
250k
rid
1
5
11W
v
and Grid 2
Heat
200k
0.001
T
IW
IW
GrldI
Grid6
250k
510k
..u01
560k
500k
310k
Ilk
1W
Ptgtocatnode
To Blan
ing-
-150 V
1W
-5t10 V
Voltage Supply
Courtesy RCA
Fig. 4 -4. Typical voltage- divider and control circuits for 3 -inch image orthicon.
It can be seen that when the beam moves from a less -positive portion of
the target to a more -positive portion, the signal -output voltage across the
load resistor (RL in Fig. 4 -4) changes in the positive direction. Hence,
for high lights in the scene, the grid of the first video- amplifier stage swings
in the positive direction.
Fig. 4 -4 is a typical schematic diagram of the control circuitry associated with a 3 -inch non -field -mesh tube such as the Type 5820. The dynode supply is almost universal, as can be noted from the operating potentials in Table 4 -1. This table illustrates the similarities and differences between four representative types of I.O.'s.
140
TELEVISION BROADCASTING CAMERA CHAINS
oN.
Ra
Ñ Ç
O
I
I
O
n
co
N.
00,
0
0
O
O
Ov
O
.O0
Q
WI
+
f7
h
N
+
8
CO
CO
i
h
W1
N
Ó
^
N
.-
a0
cV
.-
m
O
0
h
.0
I
N
O
C
h
N.
r
00
.0
I
I
O
O
M
O
O
T
O
o
1N1
N.
n
N
o
O
Ñ
f
N
N.
Ñ
a,
n
LL
ë
o
Ñ
o
O
CO
p
c
Ñ
°<
N
m
U
O
.0
N
'0
O
I
.00
I
O
I
O
<
Ñ
I
ro
X
2
á
^
Lo
+
O
v.
0
F..,3
O
O
N
^
.-
d
N
O
O
,
,+)
V
O
.O
Co
O
8
N
O
D
N
M
i
v
N
O
Lc)
,)
1...
{r)
O
o
Ln
i
t)
h
O
V
I
I
0
0
cl
O
+
O
N
I
^
2
O
CO
O
Q
O
0
0
O
^
Y
01
O
8
8
Ó
S
Ñ
V
N
N
Si
it
(
ñ
v)
I
I
O
>
V
a
ó
o
O
aD
U,
at
3
,°
ar
O
w
0
c
.=....
>
ö
>
ÿ
Ñ
°
.0"
á
n
a.
o
Q
m
"0
0--
0
It)
ó
>
o
..
>
c
42
Q
>
V
")
N
U
D
7
Ú
Ú
a,
rn
2
ó
d
f
L
7 7 7
Ú
á
ä
>
u
>
dO
ÿ
U
W
É
ÿ
'5
..
v >
á ó
u
Q
-a á
ó
C
«
Ó
Y0_
O
o
úO
>.
>.
a,
-0
-0
c
c
o
m
..
>
a
C
a,
c
a,
-o
o
c
>.
C
>
vO va,o
W
C
C
]
W
V
U
E
'c
ú
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
141
Basic Problems
Table 4 -2 shows still more similarities and differences between representative types of image orthicons. A "close" target -to-mesh spacing is
about 0.001 inch, a "medium" spacing is about 0.002 inch, and a "wide"
spacing is about 0.003 inch. The closer the spacing and the greater the
target area, the more the capacitance is.
In designing an image orthicon for television, there are certain "tradeoffs" in performance. The main drawbacks in earlier I.O.'s may be listed
as follows:
1.
Noise
2. Black halo around high lights
3. Untrue edge transitions on vertical image lines
4. Limited and poor gray -scale reproduction
5. Target sticking (retention of image over many
frames)
There are several approaches to help minimize such drawbacks, and hence
there are differences among tubes in target-to -mesh spacing, target material, and target size.
For example, the signal -to -noise ratio of an I.O. varies in direct relationship to the stored charge at the target. That is, as the target charge increases (greater capacitance) , the signal -to -noise ratio increases. There are
three possible ways to increase target charge:
Reduce the spacing between the target and target mesh. This reduction must not be carried to the point at which microphonics (picture
noise caused by vibration of elements under camera motion) are induced, or the onset of burn -in (picture retention) occurs.
2. Increase the target -mesh potential. This must not be carried to the
point at which spurious effects are induced by deflecting the scanning
beam away from its proper point of target incidence, causing loss of
resolution.
3. Increase the overall area of the target and target mesh, as in the 41/2 inch image orthicon as compared with the 3 -inch tube. Note from
Table 4 -2 that this approach results in the only effective increase in
signal -to -noise ratio.
1.
Increasing the target capacitance (by closer spacing between target and
target mesh) results in a somewhat longer linear transfer curve. It also increases the amplitude of the high -light signal current, helping to reduce the
black -halo effect. Higher capacitance also decreases the sensitivity of the
tube. Note from Table 4 -2 that, as the target capacitance increases, more
light is required on the photocathode in order to reach the knee of the
transfer curve.
Poor edge transitions are caused by poor beam landing. Since the beam
approaches the target at near zero velocity, it may be attracted from its
normal point of incidence to a more positive area, resulting in the convergence of highly positive (white) areas into darker areas. By incorporat-
142
TELEVISION BROADCASTING CAMERA CHAINS
«
'5
to >
I
I
I
I
I
I
1
co
N
I
el
h
N
C
cp
ó
o
I
Q
>
;
I
{'1
q°o
H
N K
I
>
ó
O
=
c
w-)
o
M
ó
>V
m
«ó
A
>
á
" 3 IE?,';
O
C
W
Ç.8
d
O
Ó7.E
a.
I
m
h
in
el
co
co
C
I
co
o
m
ó
r
p
O
N
°
o
O
Ó
I.
.o
Ó
o
.0
N
8
m
N
o
.o
Ó
m
E
É`c
OC
m
K
.-.
>
j-
C
4
Q
«
N
c
C
H
_
0
d
O
W
C
m
Ñ
d
Ñ
0O
40
C
C
áá
É
L
N
O
O
:
C
O
0p
C
.5
ú
a«
W W
C _
_W
L
N
L
o
Ó
q
o
o
O
O
In
Ci
pl
Lc)
q
o
o
o
o
O
Ó
wYd
C)
a
`c
E b
o
o.
d
W
C«
:
C
ó6
O°
C
C
a
Y
Y
C
Ñ Y
Q
°
d
O
a
O
C
0. 2 Y
Ñ Y Q
Ú
Ñ
2
7
C
ñ
T..
d
C
O
O. C
Y
Y
d
n
Q
Q
y L
Y
t«
ú
h W W
C
2'
Ä3
ó
a
m
î
O
w w
q Ç
" « o V
y C
E
¡
=
Ii
v
a
N
d
#
.
N
.O
ó
Ó ó
O O
p
0
Ñ
g
Ó
O
Ñ
g 0
O
U
N
ó
C
<
Ñ
oo
Ó Ó
0
áO cÑ
G
Si
ä
Ó
o
O
Ú
VÿL
j
_Y
W
W
('
W
W
()
°1«
11«C
W N
Ñ
_
L>A
vj q
M
Ó.
O
co
-ó
V
O
l'.
m«
ó
8
ó
N
N
?
Ñ
O
g
o_
d
CC
O
h
m
m
m
m
o
o
7,
h
ÿ
O
Ñ
-oÉC
co
m
OO
«L
U
L. r_
d
6:
Ñ
d
X
Ç.
Ñ
r
N
z
ú
O
C
Ñ
ú
c
f7
^
>-
Vc
M..
f")
úc
ch
N
C
N
ÿ
"
>
V
C
V\V
c
«o
Q 2
\
.Q
N
O
>
V
C
V
W Ñ
-f
N A O
-o
q O. N
« O C
n W
IUI
;
143
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
ing a field- correcting mesh (Figs. 4 -1B and 4 -1C), which straightens and
stiffens the electrostatic field on the scanned side of the target, much better
beam landing results. This eliminates the 'overemphasized" outlines seen
from non -field -mesh tubes.
There are two types of target glass (Table 4 -2) standard glass and
electronically conducting glass. Conduction in standard glass is ionic, and
this type of target is subject to burn -in and sticking. Electronically conducting glass targets have higher stability, greater resistance to burn -in,
and less granular effect. With this target material, no orbiter is required.
Bialkali photocathodes (Table 4 -2) are characterized by long -term
stability in sensitivity and resolution.
,
Resolution
It can be shown" that each megahertz of bandwidth corresponds to
80 TV lines (approximately) of resolution. Thus, an amplifier with 10MHz bandwidth would allow a picture resolution of 800 TV lines. But the
-O
Scanned Element
Scanning
Beam
Transient Response
Fig. 4-5. Aperture distortion.
co
pickup tube is limited in its amplitude response at high frequencies (thin
vertical lines in the image) by the aperture effect (Fig. 4 -5) The scanning beam is round and is of finite size. Note from Fig. 4 -5 how the tube
output voltage is proportional to the cross -sectional area of the scanning
beam; a somewhat sinusoidal transition is produced when the ideal transition would be a square wave.
.
2Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971) ,
Section
3
-3.
144
TELEVISION BROADCASTING CAMERA CHAINS
Curve B of Fig. 4 -6 is typical of the Type 5820 I.O. The increased amplitude response at low line numbers (below approximately 250 lines)
results from signal redistribution effects on the target when the lens is
operated one or two stops above the knee of the transfer curve. The field mesh 3 -inch tube and the 41/2-inch tube exhibit this characteristic to a
much lesser extent because of the improved beam landing and lowered
spurious response. However, many practicing engineers are at a loss to
explain why the image appears "sharper" from a non -field -mesh tube ( such
as the 5820) than it does from a field -mesh type ( such as the 7293) , even
though they can "see" the same horizontal resolution on the test pattern
from either tube.
With a properly set up Type 5820 I.O., the "sharpness" and "snappiness"
resulting from the overemphasized outlines of the picture elements cannot
be denied. However, it is this very spurious response characteristic that is
hard to control on "glints" from sequins on a dress at the studio, or when
panning across unexpected light sources on remotes. We are all familiar
with the black halos that result, sometimes giving the appearance of sending the entire scene into an unlighted coal mine. The bad-landing effects
of the non -field -mesh image orthicon prohibit its use in color cameras.
The curves of Fig. 4 -6 show the typical uncompensated response of the
three pickup tubes in most common use today. These curves represent the
condition in which the camera is focused on a square-wave test pattern
and the signal is measured with an amplifier of 10 -MHz bandwidth.
140
\
/
120
//
a°f
100
ä
°
Test Pattern: Square Wave
High Lights One Stop Above Knee
\
Target Volts:
\A
80
\B
cc
W
ce
m
n
-Curve B: Non-Field Mesh
Curve C: 4 -112
á
C
\\ \`C\
Curve A: Field Mesh
40
Volts Above Cutoff
\ `\
U
C
2
`,
Inch Tube
20
o
100
200
300
400
500
600
700
800
TV Line Number
Courtesy RCA
Fig. 4 -6. Uncompensated response of image orthicons.
145
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
The "aperture effect" is similar to passing the signal through a low-pass
filter with no phase distortion. Thus, aperture correction is ideally made
with a device that produces a rise in high- frequency response that corresponds to the slope of the aperture loss, without introducing phase shifts.
This is normally achieved by using an "open -ended delay line." The input
capacitance of the preamplifier stage does produce some phase shift. Thus,
we will always find some means of correcting the phase of the video signal.
These circuits are variously termed high -peakers or phase correction, and
there is no sharp line of demarcation between the two terminologies.
NOTE: High -peaker and phase- correction circuitry is covered in Chapter
5, since this is generally incorporated in video preamplifiers. Aperture
correction is discussed in detail in Chapter 6.
/
Note from Fig. 4 -6 the improvement in amplitude response of the 41/2 inch tube compared to the 3 -inch tube at higher line numbers such as
400 TV lines. The image size at the target of the 41/2 -inch tube is 1.5 times
Scanning Spot Same Diameter
EMIAn.
3-Inch Tube
Charge on Target
Signal Ou put
4 -1/2
Inch Tube
Scan
Scan
Charge on Target
Signal Output
Fig. 4 -7. Influence of tube size on aperture effect.
that of the 3 -inch tube (Fig. 4 -7) . Since the scanning spot can be made
just as small in the larger tube as it is in the smaller tube, the net effect is
equivalent to obtaining a smaller scanning spot per picture element. Thus,
the larger tube has better resolution and requires less aperture correction.
Since aperture correction in turn introduces noise, particularly in the dark
areas of a scene, better signal -to -noise ratio, as measured through the 10MHz amplifier, results.
Tube Setup Controls
Fig. 4 -8 illustrates setup controls and typical electrode voltages for the
41/2-inch I.O. Actual terminology used by various manufacturers for corresponding setup controls differs. For example, photocathode focus may be
termed "image focus." On the other hand, the image accelerator (G6) control is termed "image focus" on some cameras; on others, it may be called
"acc" or "G6 volts." Beam focus sometimes is termed "wall focus" or "orth
146
TELEVISION BROADCASTING CAMERA CHAINS
jale
H
H,.
p.
E a
U
A OOZt
V
00£1
A
VO
d
A OOOI
£a
A
o
008
c.,
Z0
A
009
TO
Z0
£9
A
A
06Z
06Z-OSZ
C9
A OOI
-0L
A OC
59
A
SZ-SI
ysapy 1,96.1e1
90
8
alettlapej
d
é
>
§
-
Fig. 4 -8. Controls and typical voltages for 41/2 -inch image orthicon.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
147
focus." Every operator must familiarize himself with the terminology for
his particular camera, since this may vary not only between manufacturers,
but also between systems of the same manufacturer.
The target control normally incorporates a "target set" switch that biases
the target to the specified operating voltage below cutoff. With the minimum amount of light to be used, and maximum f/ stop on the lens, the
target -voltage control is then set so that an image is just visible on the
view finder. Restoring the target -set switch to normal then automatically
sets the target at the calibrated voltage above cutoff. During operation, the
sensitivity is controlled with a knob that controls either the lens iris or a
neutral-density wheel in front of the optical system to vary the amount of
exposure.
The beam control sets the G1 bias, and therefore determines the amount
of beam current. This is adjusted to just discharge the maximum high
lights, and must be readjusted continually during initial setup of the tube
for all other operating parameters. After this, it is left alone. Too little
beam results in white clipping or a reversed -polarity picture. Excessive
beam results in loss of resolution from beam spreading, and excessive noise
in the picture.
PC (photocathode) focus and beam focus are adjusted for maximum
resolution. The voltage on G6 also affects resolution and normally is adjusted, consistent with the highest PC voltage that can be obtained, to
eliminate "S" distortion. This distortion is evident as wavy horizontal
image lines as the camera is panned.
The decelerator (G5) voltage is adjusted for minimum corner shading
and maximum corner resolution.
The multiplier -focus ( sometimes labelled MULTI FOCUS) control for
G3 is adjusted for the most uniformly shaded picture that occurs near
maximum signal output.
Note that ganged controls for dynodes D3 and D5 are shown in Fig. 4 -8.
Sometimes only one control (D3) is used, and this normally is labelled the
ORTH GAIN control. It is set to give a specified signal high -light current
into the input of the video preamplifier.
When the voltage on the field -mesh electrode is adjustable (41/2 -inch
tubes), it usually is adjusted to eliminate mesh beat interference patterns
in dark areas of the picture.
Setup techniques for pickup tubes are covered extensively elsewhere.3
The following general outline is a composite of manufacturers' tube instructions, and is presented as a review.
The setup procedure for non -field -mesh tubes is as follows: After the
tube has been inserted in its sockets and the voltages have been applied,
;For example, see: Harold E. Ennes, Televison Broadcasting: Equipment,
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co.,
Inc., 1971).
148
TELEVISION BROADCASTING CAMERA CHAINS
allow it to warm up for I/4 to I/2 hour with the camera lens capped. Uncap
the lens momentarily while adjusting the grid -1 voltage to give a small
amount of beam current. This procedure will prevent the mesh from being
electrostatically pulled into contact with the glass disc. Make certain that
the deflection circuits are functioning properly to cause the electron beam
to scan the target. Adjust the deflection circuits so that the beam overscans
the target, i.e., so that the area of the target scanned is greater than its
sensitive area. This procedure during the warm -up period is recommended
to prevent burning on the target a raster smaller than that used for on -theair operation. Note that overscanning the target results in a smaller -thannormal picture on the monitor.
With the lens still capped and the target voltage set at approximately 2
volts negative, adjust the grid -1 voltage until noise or a rough -textured
picture of dynode 1 appears on the monitor. Then adjust the alignment coil current so that the small white dynode spot does not move when the
beam -focus (grid 4) control is varied, but simply goes in and out of
focus. During alignment of the beam, and also during operation of the
tube, always keep the beam current as low as possible to give the best
picture quality and to prevent excessive noise and burning of the dynode -1
surface.
Next, uncap the lens and partially open the lens iris, and focus the
camera on a test pattern. Advance the target voltage until a reproduction
of the test pattern is just discernible on the monitor. This value of target
voltage is known as the "target cutoff voltage." The target voltage should
then be raised exactly 2 volts above the cutoff -voltage value, and the beam current control should be adjusted to give just sufficient beam current to
discharge the high lights. Then adjust the lens to produce best optical
focus, and adjust the voltages on the photocathode and grid 4 to produce
the sharpest picture.
At this point, attention should be given to the voltage controls for grids
3 and 5. Grid 5 is used to control the landing of the beam on the target
and consequently the uniformity of signal output. The grid-5 control
should be adjusted to produce the most uniform picture shading from
center to edge with the lens iris opened sufficiently to permit operation
with the high lights above the knee of the light- transfer characteristic. The
value of grid -5 voltage should be as high as possible consistent with uniform shading. Grid 3 facilitates a more complete collection by dynode 2
of the secondary electrons that are released from dynode 1. The grid -3 control should be adjusted to the position that results in the maximum signal
output.
Now, with a test pattern consisting of a straight line centered on the
face of the tube, adjust the voltage on grid 6 and the voltage on the photocathode to produce a sharply focused straight line on the monitor. Improper adjustment of the grid -6 control will cause the straight -line pattern
to be reproduced with a slight S shape.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
149
The above adjustments constitute a rough setup. Final adjustments
necessary to produce the best possible picture from the image orthicon
camera are as follows:
With the lens capped, realign the beam. Beam alignment is necessary
after each change of the grid -5 control and sometimes after each adjustment of the grid -3 control.
The proper illumination level for camera operation should be determined next. For most 3 -inch non -field -mesh tubes, adjust the target voltage
accurately to 2 volts above the target -cutoff value. Remove the lens cap
and focus the camera on a test pattern. Open the lens iris just to the point
at which the high lights of the test pattern do not rise as fast as the low
lights when viewed on a video-waveform oscilloscope.
Next, cap the lens and adjust the grid -3 voltage control so that the
video signal when viewed on a video- waveform oscilloscope has the flattest possible trace. This represents the black level of the picture.
The lens iris setting then should be noted, and the lens opened not more
than one stop beyond this point, unless extreme scene -contrast ranges
necessitate further opening of the lens.
The use of a higher value of target voltage than that recommended will
shorten the life of the tube. The target -voltage control should not be used
as an operating control to match pictures from two different cameras.
Matching of cameras should be accomplished by control of the lens -iris
openings.
Retention of a scene, sometimes called sticking picture, may be experienced if the tube is allowed to remain focused on a stationary bright scene,
or if it is focused on a bright scene before it reaches an operating temperature in the range from 35° to 45°C. Often, the retained image disappears
in a few seconds, but sometimes it may persist for long periods before it
completely disappears. A retained image generally can be removed by
focusing the tube on a clear white screen and allowing it to operate for
several hours with an illumination of about 1 foot- candle on the photocathode.
To avoid retention of a scene, it is recommended that the tube always
be allowed to warm up in the camera for 1/4 to I hour with the lens iris
closed and with a slight amount of beam current. Never allow the tube to
remain focused on a stationary bright scene, and never use more illumination than is necessary.
The setup procedure used for the field -mesh type of image orthicon differs in the following respects from that to be used for image orthicons that
do not employ field -mesh design: First, because the dynode aperture of
the field -mesh tube cannot be brought into focus, different alignment
techniques must be used. Second, the field -mesh tube may not be operated
on certain grid -4 voltage loops because of severe mesh -beat patterns that
result. To obtain optimum performance from the field -mesh tube, the following setup procedure should be followed carefully:
150
TELEVISION BROADCASTING CAMERA CHAINS
Before the proper voltages are applied to the tube, the lens should be
uncapped and the lens iris opened. This is a very important step for this
type of image orthicon. The proper voltages should then be applied, and
the grid -1 voltage should be adjusted immediately to produce a small
amount of beam current. This procedure will prevent the mesh from being electrostatically pulled into contact with the glass disc. Make certain
that the deflection circuits are functioning properly to cause the electron
beam to scan the target, and adjust the deflection circuits so that the beam
overscans the target. (The purpose of this procedure during the warm -up
period is to prevent burning on the target a raster smaller than that used
for on- the -air operation.) The lens should then be capped, and the tube
should be allowed to warm up for 1/4 to I/z hour before use or before other
adjustments are made.
Next, uncap the lens and partially open the lens iris. Increase the target
voltage until information appears on the monitor. Then adjust the beam
focus, image focus, and optical focus until detail can be discerned in the
picture. Adjust the controls for alignment -coil current until picture response is maximum. If the picture appears in negative contrast, increase
the beam current. Further adjust the alignment -coil current so that the
center of the picture does not move when the beam -focus (grid 4) control
is varied, but simply goes in and out of focus. During alignment of the
beam, and also during operation of the tube, always keep the beam current
as low as possible to give the best picture and prevent excessive noise.
Next, focus the camera on a test pattern. The distance from the camera
to the test pattern should be set so that the corners of the test -pattern
image just touch the inside of the target ring. Next, the deflection circuits
are adjusted so that the entire test pattern just fills the TV raster. The target voltage is then advanced or reduced to the point at which a reproduction of the test pattern is just discernible on the monitor. This value of
target voltage is known as the target -cutoff voltage. The target voltage
should be raised exactly 2 volts above the cutoff -voltage value, and the
beam -current control should be adjusted to give just sufficient beam current to discharge the high lights.
Now adjust the lens to produce best optical focus, and adjust the voltages on the photocathode and grid 4 to produce the sharpest picture. To
prevent mesh -beat problems, the voltage on grid 4 should be adjusted to
a value between 150 and 180 volts.
Proper adjustment for suppression of high -light flare, or "ghosts," and
for proper geometry is obtained when the grid -6 voltage is accurately set
at 73 percent of the photocathode voltage. This adjustment may be effected by positioning a small, bright spot of light on the edge of the field
to be viewed and then adjusting the grid -6 voltage so that the "ghost" seen
on the viewing monitor disappears as the image section is brought into
sharpest focus. Improper adjustment is evident when a light spot on the
right edge of the viewing monitor produces a "ghost" above the spot and
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
151
when a light spot on the left edge of the monitor produces a "ghost"
below the spot.
Following the adjustment of the grid -6 voltage, the voltage on grid 5
should be adjusted to produce the best compromise between high signal
output in the picture corners and best geometry. Best geometry is indicated
by the absence of S distortion of straight lines and by a rectangular raster.
After adjustment of the image -section voltages, the voltage of grid 3
should be set for maximum signal output. The deflection yoke should be
rotated, if necessary, so that the horizontal scanning of the camera is
parallel to the horizontal plane of the scene.
Finally, readjust the target voltage so that it is accurately set to 2 volts
above target cutoff. With the lens open, the lens iris should be opened to
one or two lens stops beyond the point at which the high lights of the
scene reach the knee of the light transfer characteristic.
NOTE: The general setup techniques for 41/2 -inch tubes are similar to
those described, except that the target voltage is 2.3 to 4 volts above cutoff, and the field -mesh potential is adjustable for optimum performance.
In color- camera operation of an image orthicon, the tube normally is
operated at no more than 1/2 stop over the knee of the transfer curve.
The return beam in the I.O. scans an area of about I/4 inch on the first
dynode around the "aperture" of the beam at the gun end of the tube. In
a non -field -mesh tube, this beam aperture is apparent as a small white dot
near the center of the picture when the lens is capped and the target and
beam are adjusted so that the texture of the first dynode can be seen. In
beam alignment of this type of tube, the grid -4 voltage is "rocked" manually by means of the beam -focus control, and the beam alignment -coil
current is adjusted so that the dot does not swirl but simply "blinks" in and
out of focus.
Most monochrome and color cameras now employ the field -mesh type
of image orthicon. The mesh defocuses the return beam so that the
texture of the first dynode does not appear in the background (dark
areas) of the picture. To make beam alignment more accurate and rapid,
a rock, or wobbulator, circuit (Fig. 4 -9) for grid 4 usually is provided in
the camera. This circuit is an Eccles- Jordan flip-flop with a pulse from the
vertical -sweep circuits applied to the emitters as shown. This pulse cuts
off the transistor that is conducting, causing its collector to go more negative. Triggering causes no change in the nonconducting transistor, since
it is merely driven further into cutoff. Collector coupling causes the
transistor that is cut off to drive the other into conduction. This state remains until the next pulse arrives to initiate the reverse action. Thus, one
output pulse occurs at the collector of Q1 for every two input pulses. This
provides a 30 -Hz square wave to grid 4 when the align switch is in the
on position. With the camera looking at an alignment test pattern, a "split
field," or two images, occurs if the beam is not properly aligned. The cur-
TELEVISION BROADCASTING CAMERA CHAINS
152
10 V pk-pk
1300
30Hz
Otto
To G4
of
I
.
O.
f---i
0. 33
\
On
Align Sw
Q2
Q1
T
30 V pk -pk
Vert Pulse Input
60 Hz
10 -Volt
Zener
10k
Diode
+15 V
Fig. 4 -9. Wobbulator circuit for beam alignment.
rents in the horizontal and vertical alignment coils are then adjusted to
superimpose the two fields.
4 -2. THE YOKE ASSEMBLY
Electrons from the electron gun in the tube are first focused into a
narrow beam, and then this beam is caused to sweep back and forth across
the image on the target with a definite time interval and sequence. Such
movement of the electron beam is called the scanning process.
The beam is focused by a coil that surrounds the outside of the deflection -coil assembly and creates a magnetic cross field that narrows the
emitted electrons into a beam of constant diameter. The cross section of
the electron beam is termed the scanning aperture, probably a carryover
from the days of revolving mechanical discs with small holes that traversed the projected area of the scene.
The beam is caused to scan the image by horizontal- and vertical- deflection coils constituting a yoke around the neck of the pickup tube. The
deflection coils are provided with sawtooth current waves that deflect the
electronic beam electromagnetically.
The entire yoke assembly is surrounded with a Mumetal wrap (magnetic
shield) with removable end caps. The pickup tube mounts within the
yoke assembly. Openings in the end caps expose the pickup-tube faceplate
and base.
The yoke assembly is mounted on a cradle and base plate that pivots out
from one side of the camera housing to facilitate maintenance, adjustments, and I.O. installation and removal ( see Fig. 4 -10) The entire assembly normally pivots on a large screw located at the connector end of the
yoke. A locking thumbscrew is provided to secure the assembly in place.
.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
(A) Normal yoke position.
153
(B) Assembly pivoted out.
(C) Faceplate coil removed.
Fig. 4 -10. Yoke assembly for 41/2-inch 1.0. in RCA TK -42 color camera.
NOTE: Older monochrome cameras require the lens turret to be removed
to replace the I.O. The yoke assembly is fixed on sliding rails to allow
optical focusing by means of the focus handle on the camera head.
A temperature -control plenum with three heating elements is mounted
on one side and end of the yoke assembly but is not a part of the yoke coil assembly. The temperature-control plenum is automatically connected
to or disconnected from the plenum of an air -circulating blower when the
yoke assembly is pivoted in or out of the operating position.
Adjusting screws are provided on the yoke assembly for optical focusing
and target -mask alignment. Note that the I.O. yoke in a color camera, once
properly positioned for correct optical focus at the correct image size, is
fixed. Optical focusing during normal operation is achieved by means of
controls that operate the zoom-lens assembly.
The yoke assembly of Fig. 4 -10 performs these functions:
1.
Horizontal deflection
2. Vertical deflection
TELEVISION BROADCASTING CAMERA CHAINS
154
3.
Beam alignment
4. Focusing
5. I.O.
temperature control
Horizontal- and vertical- deflection coils produce transverse magnetic fields
that deflect the I.O. beam in a scan pattern determined by external deflection circuits. Alignment coils produce magnetic fields to effect alignment of
the beam. A focus coil provides a magnetic field for the image orthicon. A
temperature- control system circulates warm or cool air through the yoke
assembly as required to maintain the I.O. temperature at approximately
370 C.
From Vert
Scan Gen
From Horiz
Scan Gen
Horiz Coil
Cl
Rl
Horiz Centering
Horiz Coil
A
160 pF
C2
1800
R2
B
160 pF
1800
Vert Centering
Fig. 4-11. Schematic diagram of monochrome deflection yoke.
Two center- tapped, parallel-connected coils are employed in the yoke
assembly for horizontal deflection (Fig. 4 -11) . The center tap and one
side of each coil are connected to RC damping networks (C1 -RI and
C2 -R2) on a terminal board located within the coil -assembly shield. The
coils are driven by the output transformer of the horizontal- deflection circuit. Connections are made through pins of the yoke connector.
Two parallel- connected coils are employed in the yoke assembly for
vertical deflection (Fig. 4 -11) . These coils are driven by the output transformer of the vertical- deflection circuit. As with the horizontal- deflection
coils, connections are made through pins of the yoke connector.
Centering of the sweep normally is obtained by passing dc of the required polarity through the coils, as indicated in Fig. 4 -11.
Fig. 4 -12 shows the basic approach to achieving identical scanning
rasters in a three -tube color camera (all pickup tubes alike) . Deflection
coils are driven in parallel from a common source, as shown in Fig. 4 -12
for vertical deflection.
NOTE: Horizontal- and vertical -deflection amplifiers are covered in Chapter 7. We are concerned at this point only with the yoke assembly.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
155
Vert Output
Red
Green
Blue
Coils
Coils
Coils
Red Height
Green Height .
Red Vert
Centering
Blue Height
Green Vert
Blue Vert
Centering
Centering
Fig. 4 -12. Vertical -deflection coils in color camera.
When different kinds of tubes are employed in a color camera ( for example, an I.O. for luminance and vidicons for chrominance) , different deflection amplitudes must be obtained. Fig. 4 -13 shows a typical arrangement for vertical- deflection circuits. Master vertical size and linearity adjustments affect all four channels. These generally are set for the monochrome tube. Deflection for the color channels then is obtained through a
waveshape and attenuation network.
To
Mono Vert Defl Coil
Master Vert
BRG
Size and Lin
Vert Defl
1500 pF
510
Attenuation and Waves hape Network
Fig. 4 -13. Vertical -deflection circuits for
different tube types.
Fig. 4 -14 shows in simplified form a typical arrangement for a four -tube
camera. Note from this configuration that the green size control becomes
the master for all chroma tubes, with individual controls provided for red
and blue. Linearity controls for red and blue are variable resistors in series
with the deflection coils. Remember that all of this is for the purpose of
most readily obtaining registration.
To get satisfactory registration of all three images, a characteristic known
as skew4 must be controlled. Compensation for this effect is provided by
cross -mixing a small amount of vertical sawtooth into the horizontal saw tooth current, as shown in Fig. 4 -14. In a four -tube camera, skew controls
4Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971) p 202.
156
TELEVISION BROADCASTING CAMERA CHAINS
normally are provided only in the three color channels to match the luminance -tube yoke as a standard. In three -tube cameras, the green yoke is
taken as standard, and only red and blue skew controls are provided. Some
cameras do not use electronic circuitry for skew correction. A mechanical
adjustment is provided in the yoke assembly to permit rotation of the horizontal coil relative to the vertical coil.
Monochrome
Horiz Deft Coil
To
Attenuation Network
Green Size
(Master for BRG)
150 {iF
510
Horiz
Coil
Master
-
"' ° " "'
'
-
Horiz
Coil
Hori
Linearity
Horiz
Size
Green
Horiz Center
Red Size
to Red Coils
Blue Size
to Blue Coils
Coil
From High Side
of Vert Deft
Fig. 4 -14. Horizontal -deflection circuits for four -tube camera.
The alignment -coil arrangement for older I.O. cameras was as shown
in Fig. 4 -4. Only one vertical- and one horizontal -alignment coil was used,
and sometimes it was necessary to rotate the alignment -coil assembly in
addition to adjusting the current in the coils. In more recent camera designs, two series -connected horizontal -alignment coils and two series-connected vertical- alignment coils are employed (Fig. 4-15 ) . The coils are
potted in epoxy and mounted on the inner mandrel of the yoke behind the
deflection-coil assembly. The alignment coils are connected in series with
the I.O. focus coils.
The focus-coil assembly consists of series -connected, pi-wound coils on
a mandrel in the yoke assembly. The front end of the focus -coil winding is
brought out to a protruding pin on the inside of the yoke mandrel. A second pin, located 120° from the first pin, is connected to the focus -current
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
157
source through the yoke connector. The rear end of the focus-coil winding
is connected to the focus -current source through the same connector. A
third protruding pin on the inside of the yoke mandrel has no electrical
connection. This pin and the two current -carrying pins serve to position
and retain an I.O. faceplate coil.
The faceplate coil fits inside the yoke mandrel and is equipped with contacts that engage the protruding pins when the coil is properly positioned
and then rotated counterclockwise. The faceplate coil is thereby connected
in series with the focus-coil winding and must be in place to complete the
focus -current circuit.
Vert -Align Coil
Fig. 4-15. Series -connected
alignment coils.
A
Horiz -Align
Coil
Hon? -Align
Coil
A
Vert -Align Coil
B
B
Focus coils for electromagnetically focused tubes are in series across a
common regulated focus -current supply so that identical focus fields are
obtained. In four-tube color cameras employing three vidicons in the
color channels, the vidicons normally are electrostatically focused, eliminating the focus coils. In this case, the focus -current supply is used only for
the luminance tube. Otherwise, two supplies would be required because of
the different current requirements and the need for individual current
adjustment.
4 -3. YOKE MAINTENANCE
The deflection yoke and focusing coil used with the I.O. must incorporate means for preventing the magnetic field produced by the yoke from
extending into the image section of the tube. Unless proper shielding is
provided, cross talk from the yoke into the image section will cause the
electron image to "jitter." This jitter produces a loss of picture sharpness.
It is common practice to enclose the focusing coil in a cylindrical magnetic
shield. Additional shielding can be provided by fitting the inside portion
of the focusing coil that is directly over the image section of the I.O. with
a copper cylinder having a length of approximately 21/4 inches and a wall
thickness of 1/32 inch. If these camera -design considerations are followed,
the optical focus, image size, and centering characteristics will be uniform
from tube to tube.
TELEVISION BROADCASTING CAMERA CHAINS
158
Modern yoke assemblies are so designed that very little trouble occurs
in this section of the camera head. This is not true of older monochrome
cameras, many of which are still in use. The type of distortion produced
by yoke problems is termed geometric distortion. The five general types
of geometric distortion are illustrated in Fig. 4 -16.
Fig. 4 -16A shows S distortion. You are probably familiar with this type
of distortion in conjunction with the grid -6 ( image accelerator) voltage
control on the camera. You pan the camera along horizontal lines and
observe the departure from straight lines in the reproduction. This departure is the result of a nonuniform axial field in the pickup tube; such
a field causes nonuniform rotation to the electrons in the scanning pattern.
In practice, S distortion results from improper adjustment of pickup -tube
potentials, from stray magnetic fields, or from a magnetized yoke.
(A)
S distortion.
(C) Barreling.
(B) Pincushioning.
(E) Trapezoiding.
(D) Skewing.
Fig. 4 -16. Types of geometric distortion.
Pincushioning (Fig. 4 -16B) normally results from improper distribution of windings in a picture -tube (monitor) deflection yoke. It is quite
common in low -cost yokes, or it may result from an attempt to substitute
a different yoke than that intended for the particular kinescope. Barreling
(Fig. 4 -16C) may result from the same causes.
Skewing (Fig. 4 -16D) , can occur in either the pickup tube or the
monitor kinescope. It results when the horizontal- and vertical -deflection
coils are not perpendicular to each other. Color cameras employ skew controls (either mechanical or electrical) In a monochrome camera, skew
can be caused by a magnetized yoke, or stray magnetic fields.
Trapezoiding (Fig. 4 -16E) can be introduced into either the pickup
tube or the display monitor. It occurs when one set of deflection coils is
not symmetrically placed with respect to the axis of the other; the axes of
.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
159
the horizontal- and vertical -deflection coils should effectively bisect each
other. Trapezoiding also can be caused by a defective capacitor or resistor
network used as built-in compensation for the difference in effective capacitance of the two sides of the coil.
If you encounter a bothersome degree of geometric distortion such as
that in Fig. 4 -16A, 4 -16D, or 4 -16E, you may have a magnetized yoke.
First, note whether the type or shape of distortion changes with change
of location of the camera or monitor. If it does change with location, you
have a stray magnetic field. If it does not change with location, you have a
magnetized yoke, or something in the camera or monitor is strongly
magnetized.
Disconnect the focus coil (not the deflection coil) leads from the
terminal board. If no terminal board is used for the focus -coil leads,
simply disconnect the camera cable, and locate where these leads go
on the camera -cable receptacle.
2. Attach the output of a variable autotransformer (with switch off)
across the focus-coil leads.
3. Set the autotransformer arm on 115 volts, and plug the cord into
an outlet.
4. Turn the variable autotransformer on. Reduce the voltage to zero in
about 5 seconds.
5. Turn the autotransformer off, remove the autotransformer leads, and
restore the focus coil to normal operation. The yoke should be demagnetized.
1.
If the above method does not work, it will be necessary to use the
longer procedure of removing the entire yoke from the surrounding shield,
and demagnetizing it with a degaussing coil of the type used for color picture tubes and receivers. Use the same coil on the entire camera. The small
hand -type degaussers used for magnetic audio -recording heads are not
effective in this procedure.
4 -4. SPECIAL NOTES ON THE 3 -INCH FIELD -MESH 1.0.
A problem that may be encountered is changing over from the non -fieldmesh I.O. (such as the 5820) to a field -mesh tube ( such as the 7293) . For
the field -mesh tube, never have the lens capped when you first apply dc
voltages to the camera. The lens should be uncapped with some light falling on the photocathode, and the beam control should be turned up so that
there is some beam current. This prevents formation of static charges between the field mesh and the target; such charges can cause "sagging" of
elements and consequent damage or shortened tube life.
The field-mesh tube is more critical in beam alignment than is the nonfield -mesh type. It usually will operate at top performance only at one
particular mode of focus. For the 3 -inch field -mesh tube, the grid -4 voltage
TELEVISION BROADCASTING CAMERA CHAINS
160
normally is in the range of 140 to 170 volts. Now look at Fig. 4 -17, which
is the circuit associated with the ORTH FOCUS control in the RCA TK -11
camera. This particular circuit has maximum voltage occuring at maximum
counterclockwise rotation of the control. As the control is turned clockwise, the voltage on grid 4 is reduced; the lowest voltage obtained is
about 130 volts. Thus, in this particular camera, the first mode encountered
when starting from the maximum -clockwise position of the ORTH FOCUS
control is usually the optimum mode of focus. On any other mode of focus,
there may be a coarse -mesh background in picture low lights, and
noticeable with the lens capped.
The field -mesh I.O. has no "white dynode spot" by which beam alignment may be judged. Therefore, the tube can be aligned with the same
procedure as that used for the vidicon. The alignment current is adjusted
so that the center of the picture does not change position as the grid-4
voltage is varied.
To
i.0.
+280V
Orth Focus
50k
Cw
Approx 130 V
Fig. 4 -17. Circuit for orth -focus
control in TK -11 camera.
47k
A much quicker and more accurate method of beam alignment for
these tubes is the wobbulator method. If you feed a synchronized 30 -Hz
square wave to grid 4, beam misalignment will result in a split image of
the entire pattern. It is then only necessary to adjust the alignment controls to converge the pattern into one well- defined image. Older cameras
did not provide this facility, but it is particularly helpful in aligning the
field -mesh I.O. ( Section 4 -1) .
Fig. 4 -18A is the schematic of a practical device for this purpose. It has
only two tubes, since you can "borrow" +280 volts from a regulated supply in your existing installation. The 30 -Hz square -wave output can be
looped through all camera control units as shown in Fig. 4 -18B; thus only
one wobbulator is capable of handling all cameras. Note that this line
should not be terminated. This system requires adding the necessary loop through coaxial connectors to each control unit, and the addition of a
switch as shown.
The reader may have built other wobbulator circuits that have appeared
in various publications in the past, and found they did not work. The
trouble may not have been in the wobbulator, but simply a result of the
ignoring of other aspects. For example, see Fig. 4 -18C. If your camera is
similar to the RCA TK -11, the camera focus circuit has a resistor and by-
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
161
+280 V
+,100HF
47k
400 V
470k
3.6
470k
3.9k
1W
meg
1W
1W
5%
VIA
112
0.1
0.1
12AV7
30-Hz Square Wave
0.1
1
7t
Vert
7i
V1B
1/2 12AV7
V2
0.01
1
270
meg
1
Divide -by -2 MV
meg
100 uF
50V
43k
1
meg
Output
8
12AU7
Drive
1N341
5%
2.2k
25k
1W
Gain
(A) Schematic diagram.
To
Connect to Arm
of Focus Control
Off
Align
I
From Wobbulator
1
Orth
R100
From
Switch
{iF
220k
Added
Focus
Switch
Normal
(Do not
terminate line.)
(Leave switch on
CIOBI0.1
To Next
Camera Control
LOCAL -REMOTE
remote position.)
Camera Control
(B) Control -unit modification.
(C) Modification forTK -11 camera.
Fig. 4 -18. Tube -type wobbulator circuit.
pass capacitor incorporated as shown. As a result, very little of the
wobbulator signal is transferred to the I.O. Fig. 4 -18C shows the modification required for use of the wobbulator signal.
In practice, the switches in both the camera control and camera are
placed in the align -on position, and the beam is quickly aligned. (Remember to have grid 4 on the optimum mode of focus.) Then both switches are
thrown to the off, or normal, position.
4-5. THE IMAGE -ORTHICON COOLING SYSTEM
Fig. 4 -19 illustrates the cooling system of the Visual 3 -inch I.O. zoom
camera. This automatic temperature -control system is designed for fast
warm -up of the image orthicon and accurate control of its temperature as
well as the inside temperature of the camera. Fig. 4 -19 shows a side view
of the camera with the lens on the left separated by a plenum chamber
from the image section of the I.O. tube. The zoom lens has a relatively
long back focal length, so there is room in front of the I.O. photocathode
for this chamber, which contains the light filter wheel, heating elements,
and the outside -air intake.
TELEVISION BROADCASTING CAMERA CHAINS
162
Sliding Door
Plenum Chamber
Q
T2
-T1
-7
Lens
Samples Air
Inside Camera
Air Flow
-i
Deflection Yoke
BI
Chamber With
Power
Motor
Transistors
Camera Air Exhaust
B2
Filtered Outside
R
Heater
Samples Outside
Air In Base
Air In
T- Thermostat
B
Blower
B3
1,
in Base
/
Base, Lens Mount &
Mechanism Housing
Courtesy Visual Electronics Corporation
Fig. 4 -19. Camera temperature -control system.
A mercury- contact thermometer is located inside the focus coil, adjacent
to the image section, with its bulb near the target. The thermometer has
two contacts; one operates at 86° F, and the other operates at 104° F.
When the camera is first turned on, three heaters are turned on for fast
warm -up; these are R3 surrounding the image section, and R1 and R2 in
the plenum chamber. Until the thermometer reaches 86° F, the I.O. beam
is held off by control of grid 2. When the temperature reaches 104° F, the
second contact in the thermometer turns off the three heaters (R1, R2, and
R3) . Air is constantly drawn in through the filters in the bottom of the
camera, through the plenum chamber, through the deflection yoke, and
into the camera interior by blower B1. The motor for this blower is on at
all times. The same motor also drives blower B2, which forces air through
a chamber surrounding the power transistors. Thermostat T2 samples the
air temperature at the top of the camera, and at 86° F it turns on blower
B3, located in the camera base, which exhausts air from the camera.
Thermostat T3 samples the outside air, and if the ambient temperature
is below 14° F T3 turns on an additional heater, R4, in the intake plenum
chamber for extra fast warm-up under very cold field conditions. Also, a
door is provided between the intake plenum chamber and the camera interior; this door can be opened to recirculate camera air to the I.O. tube
for extra cold field conditions.
4 -6. THE VIDICON
There are several basic differences in types of vidicons ,5 as follows:
Physical size:
5For additional information on vidicon tubes, see Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971).
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
1.
1
163
-inch diameter
2. 11/2-inch diameter. The larger tube diameter provides more
TV lines
per picture height (resolution) .
Focus and deflection method:
Magnetic focus and magnetic deflection. This method normally provides highest center and corner resolution but requires the most electrical power and results in cameras having the most weight.
2. Electrostatic focus and magnetic deflection. This type of tube has intermediate resolution capability and requires intermediate electrical
power. It finds use in compact, lightweight cameras and in four -tube
color cameras where a larger tube (such as a 11/2-inch vidicon or
large image orthicon) is used for luminance information.
3. Electrostatic focus and electrostatic deflection. This type of tube requires the least electrical power and permits a considerable reduction
in necessary camera size and weight. Its resolution capability usually
is less than for other types of vidicons when operating voltages are
the same. Normally, it is found only in portable and mobile TV
1.
cameras.
Conventional mesh or separate mesh:
The conventional -mesh type has the mesh (grid 5) and wall (grid
4) electrodes internally connected.
2. In the separate-mesh type, wall and mesh electrodes are not internally
connected and are brought out to separate pins on the tube base.
1.
Grid
1.
2.
3
and grid 4 connections:
Grid
Grid
3
3
and grid 4 internally connected
and grid 4 brought to separate terminals
Figs. 4 -20, 4 -22, and 4 -23 illustrate the basic differences in vidicons.
The structural arrangement of the vidicon shown in Fig. 4 -20A consists
of: a target composed of a transparent conducting film (the signal electrode) on the inner surface of the faceplate and a thin photoconductive
layer deposited on the film; a fine -mesh screen (grid 4) located adjacent
to the photoconductive layer; a beam- focusing electrode (grid 3) connected to grid 4; and an electron gun.
Each element of the photoconductive layer is an insulator in the dark
but becomes slightly conductive when it is illuminated; it then acts like a
leaky capacitor having one plate at the positive potential of the signal
electrode and the other floating. When light from the scene being televised is focused on the photoconductive -layer surface next to the faceplate,
each illuminated layer element conducts slightly, depending on the amount
of illumination on the element. The potential of the opposite surface (on
the gun side) is then caused to rise in less than the time of one frame
TELEVISION BROADCASTING CAMERA CHAINS
164
Horizontal and Vertical
Deflection Coils
Alignment Coil
Focusing Coil
Grid
Grid
3
Grid
2
Cathode
1
Glass Faceplate
''-Grid 4
UE_
Target
11
Target Connection
(A) Physical arrangement.
Basing Diagram
(Bottom View)
Direction of Light : Into Face End of Tube
GI
Pin 1:
Heater
Pin 2:
Grid
Pin 3:
Pin 4:
Internal Connection Internal Connection -
Pin
Grid 2
5:
1
Pin 6:
Pin 7:
Grids 3 and 4
Cathode
Pin 8:
Heater
Do Not Use
Do Not Use
Flange: Target
Short Index Pin: Internal Connection - Make No Connection
(B) Basing diagram.
+290V
Grid
2
5000
Signal Output
to Grid of
tim
Grids
3
and
4
nh
1st Video Amplifier
10k
W
0.1
A
Noninductive
70k
120k
50k
0.1
7038
Bottom View
Noninductive
1000
Target
100k
100k
=0.1
4 VF
200k
2W
I0.1
Grid
1
500k 2W
-110 V
Heater
To
Blanking -
Voltage Supply
(C) Typical voltage dividers.
Courtesy RCA
Fig. 4-20. Vidicon with grids 3 and 4 connected internally.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
165
toward the signal -electrode potential. Hence, there appears on the gun side
of the entire layer surface a positive -potential pattern, composed of the
various element potentials, corresponding to the pattern of light that falls
on the layer.
The gun side of the photoconductive layer is scanned by a low- velocity
electron beam produced by the electron gun. This gun contains a thermionic
cathode, a control grid ( grid 1) , and an accelerating grid ( grid 2 ) The
beam is focused at the surface of the photoconductive layer by the combined action of the uniform magnetic field of an external coil and the electrostatic field of grid 3. Grid 4 serves to provide a uniform decelerating
field between itself and the photoconductive layer so that the electron
beam tends to approach the layer in a direction perpendicular to it
condition necessary for driving the surface to cathode potential. The beam
electrons approach the layer at low velocity because of the low operating
potential of the signal electrode.
When the gun side of the photoconductive layer with its positive- potential pattern is scanned by the electron beam, electrons are deposited from
the beam until the surface potential is reduced to that of the cathode; thereafter, the electrons are turned back to form a return beam, which is not
utilized. Deposition of electrons on the scanned surface of any particular
element of the layer causes a change in the difference of potential between
the two surfaces of the element. When the two surfaces of the element,
which in effect is a charged capacitor, are connected through the external
target (signal -electrode) circuit and the scanning beam, a capacitive current is produced and constitutes the video signal. The magnitude of the
current is proportional to the surface potential of the element being
scanned and to the rate of scan. The video- signal current is used to develop
a signal- output voltage across a load resistor. The signal polarity is such that
for high lights in the image, the grid of the first video- amplifier tube
swings in a negative direction.
As with the image orthicon, alignment of the beam is accomplished by
a transverse magnetic field produced by external coils located at the base
end of the focusing coil. Deflection of the beam is accomplished by transverse magnetic fields produced by external deflecting coils.
The basing diagram for this type of vidicon appears in Fig. 4 -20B. The
diagram shown is a bottom view.
The focusing-electrode (grid 3) voltage may be fixed at a value of about
280 volts when focusing control is obtained by adjusting the current
through the focusing coil. In general, resolution decreases with decreasing
grid -3 voltage. The necessary range of current adjustment depends on the
design of the coil, but should be such as to provide a field -strength range
of 36 to 44 gausses. When a fixed value of focusing-coil current capable of
providing a fixed strength of 40 gausses at the center of the focusing device
is used, the grid-3 voltage is made adjustable over a range between 250
and 300 volts.
.
-a
166
TELEVISION BROADCASTING CAMERA CHAINS
Definition, focus uniformity, and picture quality decrease with decreasing
grid -4 and grid-3 voltage. In general, grid 4 and grid 3 should be operated above 250 volts. The grid-1 supply voltage should be adjustable from
0 to -110 volts. The dc voltages required by the vidicon can be provided
by the circuit shown in Fig. 4 -20C. The Type 7038 vidicon is a typical
tube of the kind being discussed.
A blanking signal is applied to grid 1 or to the cathode to prevent the
electron beam from striking the photoconductive layer during the return
portions of the horizontal- and vertical- deflecting cycles. Unless this is done,
the camera -tube return lines will appear in the reproduced picture. The
blanking signal is a series of negative voltage pulses applied to grid 1, or
a series of positive voltage pulses applied to the cathode.
Beam intensity is controlled by the amount of negative voltage on grid
1. The beam must have adequate intensity to drive the high -light elements
of the surface of the photoconductive layer to cathode potential on each
scan. When this does not occur, and only the low -light elements are driven
to cathode potential, the picture high lights all have the same brightness
and show no detail. Also, when the beam has insufficient intensity, the
photo- conductive -layer surface that normally rises in potential by only a
small fraction of the signal -electrode potential during each scan, gradually rises to a value approaching the full signal -electrode potential in the
high lights. Under this condition, many scans are required to drive to
cathode potential any element that has changed from a high light to a low
light because of movement of the image. As a result, the high lights tend
to "stick." The loss of high -light detail and sticking of the high lights is
referred to as "bloom." At the other extreme, a beam with excessively high
intensity should not be used because the size of the scanning spot increases
with resultant decrease in resolution.
Uniform signal output over the scanned area can be obtained if the
vidicon is operated with a deflecting -yoke and focusing -coil system designed so that no beam -landing errors are produced in the vidicon. If the
tube is to be utilized with focusing and deflecting systems that introduce
such errors (as was common in older cameras, many of which are still in
use) , uniform sensitivity over the scanned area can be achieved by compensating for the beam- landing errors.
Sensitivity variations resulting from beam- landing errors are in the
form of lower signal from the edges of the scanned area than from the
center. However, because of the uniformity of the photoconductive layer,
these variations in sensitivity are the same from tube to tube. Compensation for the beam -landing errors to achieve uniform sensitivity can be obtained by supplying a modulating voltage of a suitable waveform to the
cathode. The desired waveform is parabolic in shape and of such polarity
that the cathode voltage is lowered as the beam approaches the edges of the
scanned area. The modulating waveform contains parabolic components of
both the horizontal- and vertical- scanning frequencies. The horizontal
167
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
component should have the greater amplitude and is the most effective in
obtaining uniform sensitivity.
Fig. 4 -21 shows the amount of parabolic-waveform voltage required and
the method of applying the waveform to the cathode, grid 1, and grid 2 of
a vidicon. The modulating voltage is applied to grid 1 and grid 2 as well
as to the cathode to prevent modulation of the scanning beam.
+300V
7038
5100
30 pF
Grid
1
Gat node
10k
7
Oto-110V
100k
+300 V
Volts
1
pF
1W
Peak -to-Peak
IApproxl
R2'
12A 378
0.1
6
2
Mixed -P*--.71
Parabolic-Waveform 125
Signal Input
88F
If
it
8
Blanking Signal Input
10k
4
1000
1
meg
9
5
10
1000
1
8µF
meg
80 Volts
Peak -to-Peak
IApproxl
'Adjusted
That Grid 1 and Cathode Parabolic
Waveforms Are Equal in Amplitude
So
8200
1W
Courtesy RCA
Fig. 4 -21. Circuit for compensation for beam -landing errors.
The use of this modulating waveform also improves the center -to -edge
focus of the vidicon and assures that sensitivity over the scanned area will
be uniform for the recommended dark current for any specified service.
Care must be taken that identical waveforms are applied to the electrodes
of the three tubes in a three -vidicon color camera to insure good registration of all signals over the entire scanned area.
Fig. 4 -22A is a diagram of a vidicon (such as the Type 6326) with
separate grids 3 and 4. Note that although the screen mesh adjacent to the
photoconductive layer is connected to grid 4, it is given the separate designation of grid 5. Grid 4 is the beam- focusing electrode, and grid 3 can be
used as a vernier focusing electrode. Fig. 4 -22B illustrates the base -pin
designations for this type of tube.
168
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 4 -22C shows a typical voltage- divider arrangement for this type of
vidicon when grid 3 is externally connected to grids 4 and 5. Fig. 4 -22D
illustrates how grid 3 can be individually adjusted for optimum electrical
focus of the image.
Fig. 4 -23 illustrates a more recent vidicon of the electrostatic-focus,
magnetic -deflection type, with a separate-mesh arrangement. Grid 5 is the
additional electrostatic focus grid. Note that grid 5 is brought to a separate
base pin, and that the screen adjacent to the photo- conductive layer is now
designated grid 6.
Horizontal and Vertical
Deflection Coils
Focusing Coil
Grid
Alignment Coil
4
Grid
Grid 2
3
Grid
1
Cathode
1
i
Glass Faceplate
Grid
5
Target
Target Connection
(A) Physical arrangement.
+290 V
Grid
2
1
=0.1
5000
L
rids
3, 4
Signal Output
to Grid of
and 5
1st Video
lOk
20.1
Amplifier
A
Noninductive
72W
0k
50k
0.1
120k
6326
Bottom View
Noninductive
1000
Target
100k
0.1
100k
200k
T41F
2W
Grid
Ì0.l
1
500k
2 W
-110 V
To
Heater
Blanking -
Voltage Supply
(C) Grid 3 connected to grids 4 and 5.
Fig. 4 -22. Vidicon with
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
169
The weight, size, and power requirements of TV cameras employing
this tube are substantially less than the requirements of cameras using conventional magnetic-focus, magnetic- deflection vidicons of comparable size
(1 -inch diameter). Camera size and weight are automatically reduced by
elimination of the magnetic focusing coil. Negligible power is required
for electrostatic focusing. Deflection power is one-fourth of that required
for vidicons using magnetic focusing and deflection, having a separate
mesh connection, and operating at equivalent mesh potential; it is onesixth of that required when the mesh and wall electrodes are connected
Basing Diagram (Bottom View)
Direction of Light: Into Face End of Tube
G4
G3
Pin 1:
Pin 2:
Pin 3:
Pin 4:
G5
GI
Heater
Grid
Grid
1
3
Internal Connection
- Do Not
Use
Pin 5:
Pin 6:
Pin 7:
Pin 8:
Flange:
Grid 2
Grids 4 and 5
Cathode
Heater
Target (Signal Electrode)
Short Index Pin: Internal Connection - Make No Connection
Short Pin
IC
(B) Basing diagram.
+2%
V
Grid
2
20.1
Signal Output
Grids
10k
4
and
to Grid of
1st Video Amplifier
5
50.1
A
Noninductive
50k
6326
Bottom View
Noninductive
Grid
3
10k
Noninductive
Signal
Electrode
1000
100k
0.1
100k
2W
200k
I0.1 Grid
1
500k2W
-110 V
Heater
To
Blanking -
Voltage Supply
(D) Grid 3
as vernier focus electrode.
Courtesy RCA
separate grids 3 and 4.
TELEVISION BROADCASTING CAMERA CHAINS
170
internally or operated at the same potential. In addition, the precision
outer diameter of the bulb permits the use of low- power, close- fitting deflection yokes of small size and low impedance.
This type of vidicon can be operated with "low" electrode potentials, as
described for Figs. 4 -20 and 4 -22, or at "high" potentials, as illustrated in
Fig. 4 -23A. Note the greatly improved resolution at higher voltages shown
by Fig. 4 -23C. This is the uncompensated response. With aperture compensation, excellent resolving power at 800 TV lines can be obtained.
Control over beam landing is obtained by adjustment of the voltages on
grids 6 and 3 and grid 5. This voltage relationship is determined by the
Grid 6 Grid
5
Grid 4 Grid 3
Grid 2 Grid
1
I.
1
To Focus Control
+60 to +160V Typical)
To Beam
1
Heater
Control
-45 to -100 V Typical)
(A) Typical one -inch tube.
(Bottom View)
Pin
Pin
Pin
Pin
Pin
Pin
Pin
Pin
Short Pin
IC
2:
3:
4:
5:
Heater
Grid 1
Grid 4
Grids 3 and 6
Grid 2
6:
Grid 5
1:
7:
Cathode
8: Heater
Flange: Target
Short Index Pin : Internal Connection - Make No Connection
(B) Basing diagram.
Fig. 4 -23. Vidicon with
171
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
camera designer and is not recommended as an operational control. In
general, best geometrical accuracy is obtained with a ratio of grid 6 and
3 voltage to grid 5 voltage of 1.67, and most uniform signal output is
obtained with a ratio of 2.
Definition, focus uniformity, and picture quality decrease with decreasing voltages on grids 6 and 3 and grid 5. In general, grids 6 and 3 should
be operated at or above 300 volts, and grid 5 should be operated at or
above 180 volts. A substantial increase in both limiting resolution and amplitude response is obtained by operating the tube with higher voltages
applied to these electrodes. It should be noted that deflection-current re100
\
- \\\
80
\
\\ \ \
\
I
1
1
Curve A:
Curve B:
Curve C:
Curve.D:
Curve E:
Grid
Grid
Grid
Grid
Grid
Volts 1300, Grid 5
Volts 1000, Grid 5
750, Grid 5
6 & 3 Volts
6 & 3 Volts - 500, Grid 5
300. Grid 5
6 & 3 Volts
6& 3
6& 3
I
Volts 780
Volts - 600
Volts 450
Volts 300
Volts - 180
High -Light Target Microamperes- 0.3
Dark Current - 0.02 Microamperes
Test Pattern:
1
Transparent Square -Wave
Resolution Wedge
I
60
1
1.1111611111
1100kak
40
ELIk%k
20
\\` \
0
200
400
i
N`\N
\ \\'
TV Line Number
600
800
(C) Resolution curves.
Courtesy RCA
additional focusing grid.
172
TELEVISION BROADCASTING CAMERA CHAINS
3 are increased when the voltages
are increased.
A composite of manufacturers' instructions concerning operation of the
vidicon in studio and film cameras follows. We will explore advanced
maintenance techniques for the vidicon further in Chapter 11.
The target connection is made by a suitable spring contact that bears
against the edge of the metal ring at the face end of the tube. This spring
contact may be provided as part of the focusing coil.
The deflection yoke and focusing coil used with the vidicon are designed
to cause the scanning beam to land perpendicularly to the target at all
points of the scanned area. This is done to obtain minimum beam -landing
error and resultant uniformity of sensitivity and focus over the scanned
area. The recommended location of these components for monochrome
cameras is shown in Fig. 4 -20A.
The deflection yoke and focusing coil in color cameras should extend
1/4 to 1/2 inch beyond the faceplate of the tube, as shown in Fig. 4 -22A.
The yoke must have a minimum inside diameter of 1Y8 inch to provide
clearance for the side tip. A long yoke, in comparison with a short yoke,
deflects the beam through a narrower angle, which effectively gives better
center -to -edge focus and reduces geometric distortion of the image. Freedom from such distortion is particularly important in color cameras utilizing the method of simultaneous pickup, in which three images must be
identical for proper registration.
The polarity of the focusing coil is such that a north -seeking pole is
attracted to the image end of the focusing coil, with the indicator located
outside of and at the image end of the focusing coil.
The scanning speed must be constant in order to obtain good black level reproduction when the vidicon is operated at high dark current with
resultant higher effective sensitivity. (Dark current is the current with the
lens capped and target voltage applied.) The dark-current signal is proportional to the scanning speed. Therefore, any change in scanning speed will
produce a nonuniformity in black level in direct proportion to the change
in scanning speed.
quirements of the tube on grids 6 and
NOTE: A constant scanning speed means that the current through the
deflection coils describes a pure sawtooth waveform. Any departure results in picture nonlinearity. This emphasizes the importance of proper
sweep -linearity adjustments in the operation of vidicon cameras.
The alignment coil should be located on the tube so that its center is at
a distance of 311% inches from the face of the tube. This coil should be
positioned so that its axis is coincident with the axis of the tube, the deflection yoke, and the focusing coil.
Electrostatic shielding of the target from external fields is required to
prevent interference effects in the picture. Effective shielding from the
fields produced by the deflection components ordinarily is provided by
173
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
grounding a shield on the inside of the faceplate end of the focusing coil
and by grounding a shield on the inside of the deflection yoke at a point
near the input of the video amplifier.
The temperature of the faceplate should not exceed 60 °C (140 °F) during either operation or storage of the vidicon. Operation with a faceplate
temperature in the range from about 25 to 35 °C (77 to 95 °F) is recommended. The temperature of the faceplate is determined by the combined
heating effects of the incident illumination on the faceplate, the associated
components, and the tube itself. To reduce these heating effects in film pickup cameras and permit operation in the preferred temperature range
with a high value of illumination, it is normal to use an infrared filter
between the projector and faceplate and to provide a blast of cooling air
across the faceplate.
The dark current is doubled for every 10 °C rise in the temperature of
the faceplate, and halved for every 10 °C decrease in the temperature of
the faceplate. To obtain optimum performance, it is desirable to operate
the vidicon at a pre -established value of dark current. Therefore, if the
temperature of the faceplate is allowed to vary, it will be necessary to adjust
the target voltage to maintain the desired dark current, as shown in Fig.
4 -24. Since the sensitivity of the tube decreases with increasing temperaCurve A: Relative Target Voltage Required to Maintain
Dark Current of 0.2 pA
Curve B: 28701( Incandescent Illumination Required
to Produce Signal-Output Current of 0.2 pA
Curve C: Persistence (Lag) Characteristic for an
Initial Signal- Output Current of 0.2 pA
100
High -Light Signal Output 0.2 Microampere
Dark Current 0.2 Microampere
Scanned Area of Photoconductive
Layer
112
"x318"
80
Zr
A
60
.'.
40
C
20
0
30
40
50
Faceplate Temperature I0C1
Courtesy RCA
Fig. 4 -24. Characteristics of typical vidicon.
174
TELEVISION BROADCASTING CAMERA CHAINS
ture, the amount of faceplate illumination necessary to produce a given
signal as a function of faceplate temperature also is shown in Fig. 4 -24. In
addition, the lag decreases with increasing temperature, as shown by curve
C in Fig. 4 -24. For live pickup, it is desirable to select an operating temperature that provides the best balance between lag and sensitivity. The
faceplate should be held close to this temperature to assure stability of
black level and signal -output level.
The target voltage is obtained from an adjustable dc source. As the target voltage is increased, the dark current increases (Fig. 4-25). The target
voltage must be adjusted to produce the desired value of dark current for
the type of operation. The target -voltage range of the vidicon for a given
value of dark current is small, as shown in Fig. 4 -25. This feature permits
utilization of simplified circuits in cameras where automatic change in
target voltage is desired to compensate for varying light levels.
0.4
Scanned Area of Photoconductive Layer
Faceplate Temperature 30°C Approx
112" x 318"
`
.,..'
e
0.1
.
n-
'{,
0.2
,r
lµ?
0.08
0. 06
d
o-
0. 04
.
E
r;
-
:)
"4;-'
l
d
T
J,:f}.it
0. 02
?
'
YDi¡1r
ry5
.
Á
O
'
0.01
t
0.008
Vs
p,r'
ï
,_a=
',. '
1
:
0.006
fy'
0.004
r.'ÿ
*v.
.i2'. P
0.002
4
6
8
10
.:#.
20
40
60
80
100
200
Target Volts
Courtesy RCA
Fig. 4 -25. Range of dark current for typical vidicon.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
175
The illumination incident on the faceplate ranges from relatively high
values for film pickup to relatively low values for direct pickup. For satisfactory operation of the vidicon at these extremely different light levels,
it is essential that the target voltage be properly adjusted with reference
to the curves in Figs. 4 -26A, 4 -26B, and 4 -26C to give the proper value of
dark current for the desired service. (Adjustment of the target voltage to
obtain the desired dark current is covered later.)
For practical purposes, the illumination on the tube faceplate can be
calculated from the following relationship:
E
E8RT
where,
E is the tube -face illumination in foot -candles,
E. is the scene illumination in foot -candles,
R is the reflectance of the scene,
T is the transmission of the lens,
f is the f-stop number of the lens.
Assuming a lens transmission of 80 percent and a scenic high -light reflectance of 60 percent, with a scene illumination of 300 foot- candles and
the lens stopped to f/2.8:
(0.8)
E= (300)(0.6)
4 (2.8)2
=
144
31.4
4.6 foot -candles on tube faceplate
for high lights
For live pickup involving low illumination levels, a- good picture can be
obtained with a high -light illumination of 1 to 3 foot-candles on the faceplate of the vidicon. Such a low illumination level, however, requires
maximum -sensitivity operation of the tube. For this type of operation, a
dark current of 0.2 microampere is required. This value is obtained with
a target voltage in the range of 60 to 100 volts. Under such low -level
illumination conditions, the lag will be greater and the black -level uniformity will be poorer than for live-pickup conditions with higher faceplate illumination and lower dark current.
When the vidicon is used for live pickup with illumination levels of 10
to 20 foot- candles on the faceplate, a dark current of 0.02 microampere is
required. This value is obtained with a target voltage in the range of
30 to 50 volts.
For film pickup, an average high -light illumination of 50 to 200 foot -
candles is required on the faceplate of the average vidicon for minimum
lag and best black -level uniformity. For this range of illumination, a dark
176
TELEVISION BROADCASTING CAMERA CHAINS
\
\
100
I
1
111111
1
I
Illumination:
2870K Incandescent
I
High-Light Signal Output 0.3 Microampere
Scanned Area of Photoconductive Layer 112'
Faceplate Temperature- 30oC Approx
I
e 3
8"
60
40
20
\\
10
6
4
1
\
NN%
t
0.001 0.002
0.04
0.1
0.02
Dark Cur ent IMicroamp resi
0.004
(A) Illumination
1
1::1riiI
ö
iú
iiiii
mIIIn
d
É
2t3,
0.2
0.4
vs dark current.
7::
Diii s;1iiiie i
ï Sill..IIIIII
:iii
a:illl0d111111i11111111111
MI
0.4
0.2
0.01
1
i
iii
t+`
Ì
i
MINN!
EEro_
. v1:ü===1:1::;1i:i:.ii
_reou w
--ou
m./I.,mUIIm./1
im...., IImma/111
0 01
--c u--r--o
um
mum IIIn1111
11,1111121111!i ÌI1IIIIIIIII1IIIIIII1IIIIIII
0.004
2ti:uiimasueoiiiiiseo:ut:::iiii1i:i iiii
0.1
F
$
0.04
â
0.02
0. CO2
.1111111 IIIIIIIII111111
//IIII
11111
,11'IIII
0.001
0.01 0.02 0.04
0.1
0 2
Ilumination: Uniform ver Photoconductive Layer/I1al
Scanned Area of Photoconductive Laye
Faceplate Temperature - 300C Approx
0.4
1.0
2870K Tungsten
2
4
40
20
10
112"
318" 11111
x
11III
100
200 400
1000
Illumination on Tube Face (Foot- Candles)
(B) Light transfer characteristic.
30
Microampere
n
Scanned Area of Photoconductiv
ayer - 112' x 318"
300C Approx
Faceplate Temperature
5
1111
0.002
0.004
I
I
I
1
I
0.04
0.01
0.01
Dark Current (Microamperes)
-
1
0.1
0.2
0.4
(C) Persistence characteristic.
Courtesy RCA
Fig. 4 -26. Typical vidicon characteristics.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
177
current of about 0.004 microampere is required, and the target voltage is
between 10 and 20 volts.
The exact value of target voltage to give the required dark current depends on the individual tube and on the temperature at which its faceplate
is operated. It is important that the tube be allowed to reach a stable operating temperature before the operating dark current is determined; other
wise, the dark current will change as the temperature of the tube changes.
In all cases, the illumination level and/or dark current must be limited
or adjusted so that the peak signal-output current does not exceed 0.3
microampere for the 1 -inch tube, or 0.6 microampere for the 11/2-inch
vidicon. In order that the signal-output current and dark current will be
known at all times, the camera sometimes is provided with a microammeter in the target circuit of each vidicon to read average target current,
or a calibration pulse of the proper magnitude is fed into the input of the
video preamplifier to indicate peak target currents.
The maximum amount of illumination on the photoconductive layer is
limited primarily by the temperature of the faceplate, which should never
exceed 60 °C and should preferably be maintained within the operating
range from 25°C to 35°C for most satisfactory performance.
Signal output as a function of uniform 2870K tungsten illumination on
the photoconductive layer for different values of dark current is shown in
Fig. 4 -26B. Note that these curves are for a typical tube under the conditions indicated. Because the target voltage needed to give maximum sensitivity at a dark current of 0.2 microampere may range between 60 and 100
volts, it is essential that the best operating target voltage be determined
for each vidicon. From these curves, it also should be noted that the illumination must be increased about 30 times to produce an increase of 10 times
in signal-output current for any given value of dark current.
The average gamma, or slope, of the light- transfer characteristic curves
in Fig. 4 -26B is approximately 0.65. This value is relatively constant over
an adjustment range of 4 to 1 in target voltage, or 50 to 1 in dark current,
for a signal- output current range between 0.01 and 0.3 microampere. Close
uniformity in the value of gamma between individual tubes is maintained
to insure satisfactory operation of color cameras in which the signal- output
currents of the three vidicons must match closely over a wide range of
scene illumination. Because its transfer characteristic is approximately the
complement of the transfer characteristic of a picture tube, the vidicon can
produce a picture having proper tone rendition.
The spectral response of the typical vidicon is shown by curves A and C
in Fig. 4 -27. Curve A is on the basis of equal values of signal -output current at all wavelengths, whereas curve C is on the basis of equal values of
signal- output current with radiant flux from a tungsten source at 2870K.
For comparison purposes, the response of the eye is shown in curve B.
Full -size scanning of the 1/2 ' X %g" area of the photoconductive layer
should always be used. This condition can be assured by first adjusting the
TELEVISION BROADCASTING CAMERA CHAINS
178
deflection circuits to overscan the photoconductive layer so that the edges
of the sensitive area can be seen on the monitor. Then, after centering the
image on the sensitive area, reduce the scanning until the edges of the
image just disappear. In this way, the maximum signal -to -noise ratio and
maximum resolution can be obtained. It should be noted that overscanning
the photoconductive layer produces a smaller -than- normal picture on the
monitor.
Underscanning of the photoconductive layer, i.e., scanning an area of the
layer less than I/2" X 3/8 ", should never be permitted. This condition
(which produces a larger- than-normal picture on the monitor) not only
causes sacrifice in signal -to -noise ratio and resolution, but also may cause
permanent change in sensitivity and dark current of the underscanned area.
Curve A: For Equal Values of Signal-Output Current at All Wavelengths
Signal Output From Scanned Area of 112" o 318" - 0.02 Microampere
Dark Current - 0.02 Microampere
Curve B: Spectral Characteristics of Average Human Eye
Curve C: For Equal Values of Signal Output Current With Radiant Flux From Tungsten Source at 2870K
120
0.030
_._
Range of
Max Value
AA
100
0. 025
/
il
U
rn 0. 020
/
`\t
:B
i
o
1
/
I
x
1
.
1
11
///
/
0.015
1
II
1
I
1
I
1
4-,
1
1
1
0.010
1
1
t
1
t
0
300
/
/
I
t
li
t
/
I
/
20
t
, 11
/
\
..,'1/
-1 ,
400
\
t
l3
0. 005
\
1
I
1
\
l
t
500
600
700
800
0
900
Wavelength lnml
i
I
I
Courtesy RCA
Fig. 4 -27. Spectral sensitivity of typical vidicon.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
4
179
An underscanned area showing such a change will be visible in the picture
when full -size scanning is restored.
Failure of scanning even for a few seconds may permanently damage
the photoconductive layer. The damaged area shows up as a spot or line in
the picture during subsequent operation. To avoid damaging the vidicon
during scanning failure, it is necessary to prevent the electron beam from
reaching the layer. This can be accomplished conveniently by increasing
the grid -1 voltage to cutoff.
The sequence of adjustments in operating the vidicon for live pickup is
as follows: With the grid -1 voltage control set for maximum negative
bias (beam cutoff) , the target voltage control set for minimum voltage
( about 10 volts) , and the deflection controls set for maximum overscan,
apply other voltages to the tube. Next, with a %" X %8" mask centered on
the face of the tube, and with the iris set for minimum opening, decrease
the grid -1 bias to just bring out the high -light details of the picture on the
monitor. Adjust the beam -focus voltage control, the lens stop, and the optical focus to obtain the best picture. Reduce horizontal and vertical scanning
so that the edges of the image extend just outside the scanned area on the
monitor. Then adjust the alignment field so that the center of the picture
does not move as the beam -focus voltage is varied. Some readjustment of
horizontal and vertical centering may be necessary after alignment.
For maximum -sensitivity operation of the vidicon in live -pickup service, proceed as follows: With no illumination on the face of the tube, increase the target voltage until a dark current of 0.2 microampere is measured. The current should be measured with a sensitive microammeter, or
by another procedure, as outlined in Chapter 11 of this book. Next, open
the lens and adjust the aperture to give a peak signal -output current of
0.2 to 0.3 microampere. A good procedure for doing this is to focus the
camera on a uniform white area having the same brightness as the high
lights in the scene to be televised. The image of this white area must at
least cover the scanned area of the tube face. The current read on the microammeter will be the dark current plus the peak signal -output current,
i.e., high -light target current.
A waveform monitor can be used to compare the peak signal- output
current produced by any scene to the peak value measured with the microammeter when the camera is focused on a uniformly bright scene. When
a camera is adjusted in this manner, video gain should be kept constant,
and the light level on the tube face should be controlled to maintain the
constant predetermined value of peak signal as observed on the oscilloscope.
After the light level is adjusted to obtain the correct signal -output current, the grid -1 bias voltage should be adjusted to just discharge the high
lights. Too much current will result in poor resolution and poor picture
quality. After the grid -1 bias is adjusted properly, it will be necessary to
check and readjust the dark current and the peak signal -output current.
Proper adjustment of the dark current, the peak signal-output current, and
180
TELEVISION BROADCASTING CAMERA CHAINS
the grid -1 bias will result in a picture of good quality with minimum
smearing of moving objects.
For average- sensitivity operation of the vidicon in live-pickup service,
the adjustments are similar to those for maximum-sensitivity operation,
except that the target voltage should be adjusted to produce a dark current
of 0.02 microampere. When sufficient light is available, decreased lag can
be obtained by operating with this lower value of dark current.
For film-pickup operation of the vidicon, the adjustments are similar to
those for live pickup, except that the target voltage should be adjusted to
produce a dark current of 0.004 microampere, and the peak signal- output
current should be adjusted to the desired value by controlling the light level
on the faceplate of the tube.
In setting up three vidicons in a color camera, particular attention must
be given to proper alignment, best obtainable focus, and identical centering
of scanned areas on the photoconductive layers. For best color balance and
color tracking over a wide range of light levels, the light level in each color
channel should be controlled so that each vidicon develops the same value
of peak signal output for white portions of a scene. Observation of these
operating conditions should assure good registration and good color balance.
4 -7. THE LEAD -OXIDE VIDICON
The lead -oxide vidicon (Plumbicon) is a photoconductive tube with
higher sensitivity than the conventional vidicon, virtually zero dark current, and a gamma of close to unity. Being the most recently developed
pickup tube, it is still more "experimental" than either the image orthicon
or the vidicon described in the previous section. It is, however, rapidly
becoming a widely used pickup tube in television cameras. The term
Plumbicon is a registered trade mark of N. V. Philips of Holland. "Lead
oxide vidicon" is the RCA terminology for the same type of pickup tube.
The lead -oxide tube is similar in electrode arrangement to the conventional vidicon, uses the same general type of focusing and deflection coils,
and has similar grid characteristics. The major difference is in the properties
and construction of the photoconductive layer.
The Plumbicon element consists of three layers. The middle layer is a
relatively thick lead oxide (PbO) acting as an intrinsic ( designated i)
semiconductor. The outer and inner layers are very thin. The layer toward
the electron gun is doped into a p -type semiconductor. The layer on the
signal -electrode side is doped into an n -type semiconductor. Thus, the
three layers form a p -i -n diode ( positive- intrinsic -negative diode) This
diode is connected in the "reverse" direction; the p -type material is toward
the tube cathode, and the n -type material is biased by the positive signal electrode potential.
Although the basic action of the Plumbicon is identical to that of the
vidicon, the intrinsic region of the p -i -n structure contributes largely to the
.
181
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
c
`\
Ñ
O
c-
o
oN
O
II
ñ
p
$
- 8E
0 7
U
O
8
O
II
O
"n
N
O
O
-Y
C
C
v
'
>
om
o
Q
C
rn
A
c0
ñ
t;
II
á
N
.-
O
Ñ
,o
U_
N
O
O
h
41
N
N
CO
m
V
.O
O
Ñ
i
O
Q
C
v
O
S.;:-;
-Ó
O
12
ó
h
8
v
r>
E
N
O
ÓI
LO
S
oo
o
.-
P
o
g
-
o
vi
ci
o.
h
v!
ry
ci
_
.-
O
o
C
O
o
,p
co
E
ó
O
p,
Lo
O
>
á
II
O
`=
C
ô
M
Ñ
á_
Ñ
.d
T
0
ó
E
roi
ó
,000
<
E
ó
ó
ro
vo
O.
E
o
V
rei
d
_o
Ñ
.
E
aó
E
0
O
ó
cc:
E
E
ó
v
q
8
m
on
in
á
o
co
ó
N
o
a.
r,
ál
o
_°;
0
JC
J
Ñ
J
O
V
E
m
o
N
O
Qy
ó
io
0
Q
iñ
o
al
c
u
n
ON
C
C
o
ú
ú
C_
do
W
O.
C
-O
ti
Ñ
¿
CD
o
L_
o
C
á
I
h
C
o
g
j=
J
w
oo
o
E
C
O
o?
O:
. t!-1
u
O
C
6?.
o o.
=_
E>
C
Zi
C
A
d
m
a
>
O
E
m
U'
()
Y
og.
o,
"2
TELEVISION BROADCASTING CAMERA CHAINS
182
relatively high sensitivity of the Plumbicon. In this region, the conductivity
is low while the electrical field strength is high. Thus, all the liberated carriers in this region contribute to the photocurrent when the target potential
is sufficiently high. A much higher ratio of signal current to dark current
is obtained than is the case for the vidicon described previously. Table 4 -3
compares typical characteristics of three available sizes of Plumbicons with
typical characteristics of 1 -inch and 11/2 -inch conventional vidicons.
Note that in Table 4 -3 luminous sensitivity is given in microamperes
per lumen (µA/1m). Sometimes this type of information in tube specification sheets is given in terms of microamperes per watt (µA /W) Usually
the light source is specified as a tungsten light of 2870K color temperature.
This light source emits 20.4 lumens per watt of total radiant flux. Thus,
you can convert microamperes per lumen to microamperes per watt by
using the 20.4 multiplication factor. For example, to convert 300 µA /Im to
.
µA /W:
300 µA /Im X 20.4
= 6120 µA /W
= 6.12 mA /W
Conversely, you can convert microamperes per watt to microamperes per
lumen with the relationship:
A/lm
4 -8.
µA/W
20.4
CAMERA OPTICS
It is important that the reader already be familiar with the fundamentals
of lenses, including lens angles, field of view and depth of field for a given
lens and pickup -tube useful scanned area, the differences between the
turrent- mounted lens and the zoom lens, etc.° It now remains to examine
the more advanced applications of camera optics, and servo control of lens iris functions.
Fig. 4 -28 illustrates the problem of bringing three images into optical
focus at exactly the same time and with exactly the same orientation with
respect to the scanning rasters. The lens in Fig. 4 -28 may be either an
objective lens or a field relay lens. It would be possible to mount a number
of these lenses of different focal lengths on a turret, but the back focal
distance must be rather large, limiting the installation to lenses that are
of very large focal length and therefore are unsuitable for average studio
pickups. So consider the lens to be a field relay lens. This means that an
objective lens is ahead of the relay lens, or a film -projector lens is throwing
a real image on the relay lens. In this case, each pickup tube would have its
OSee, for example, Harold E. Ennes, Television Broadcasting: Equipment,
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co.,
Inc., 1971), Sections 4 -1 and 4 -2.
183
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
Color-Selective
Reflectors (Dichroics)
Green
Objec ive or
Relay Lens
Fig. 4 -28. Basic color- separating optical system.
own objective lens ( normally termed reimage lens) focused on the realimage plane of the relay lens. This is the basic type of color-camera optical
system for both studio and film applications.
Since the images on all three pickup -tube photocathodes must have
identical size and must come into focus at one common adjustment, paths
1 -6 -7 ( blue) , 1 -2 -3 -4 (red) , and 1 -2 -5 (green) must be of identical
length. This is the reason for the physical placement shown.
Dichroic mirrors divide the incoming light roughly into blue, red, and
green components. The mirror at point 1 reflects blue and passes red and
green. The mirror at point 2 reflects red and passes the remaining light,
which is primarily green. Front -surface mirrors are located at points 3 and 6.
Fig. 4 -29 shows typical transmission and reflection characteristics of a
pair of dichroic mirrors. These curves reveal why it was stated that dichroic
mirrors split the light "roughly" into the three primary color components.
It is obvious that additional trim filters in the light paths following the
dichroic mirrors are necessary. The characteristics of dichroic surfaces are
always given for a fixed angle of incidence.
100
Blue (Reflection)
80
/
\
i\
I
1
60
1
'
il
0
\
.....s.."
,
,
i
!
i
450
500
550
,
__-
600
Wavelength in Nanometers
Fig. 4 -29. Typical dichroic- mirror
"
\
..\
,. y_.-._..`
400
ection)
\
A
40
r.`,
%
Green (Transmission)
1U.j
á
20
1
Red (Ref
characteristics.
650
--'
700
184
TELEVISION BROADCASTING CAMERA CHAINS
It is now time to examine the basic problems involved in color separation, and the reasons behind color optics. This will make it easier to understand which problems are under our control and which problems are not.
Conventional dichroic mirrors have the following drawbacks:
The dichroic interference layers are evaporated onto a plane -parallel
glass plate. Certain aberrations (coma and astigmatism) are introduced and make compensating optical elements necessary.
2. Reflections from the rear surface of the glass plate result in a faint
ghost image. This image is particularly noticeable under high- contrast
conditions.
1.
Fig. 4 -30. Influence of angle
of incidence.
Courtesy RCA
Properties of the interference layer depend on the angle of incidence,
as shown by Fig. 4 -30. Rays p and q from two points of scene T meet
dichroic mirror S at different angles. Since the spectral reflection
characteristic depends on the angle of incidence, color rendition at
points P and Q is different, resulting in a spurious shift of color across
the image. Rays q and q', originating at the same point of the scene,
likewise strike the mirror at different angles. Consequently, their contributions to Q are governed by slightly differing spectral reflection
characteristics. The result is a loss of color discrimination.
4. In practice, light can become polarized by reflection from polished
objects, specular surfaces, or even perspiring foreheads of persons in
the scene. Red and blue reflecting surfaces of color splitters in the
camera are sensitive to the plane of polarization, since they are at
angles with the incoming light. (Green passes straight through and
therefore is not affected by polarization). Fig. 4 -31 shows the result
of a red -reflecting dichroic mirror under polarized-light conditions.
This shows in essence that the perpendicular component ( dash line)
reflects at a shorter wavelength than does the horizontal component
(solid line)
3.
.
NOTE: Both the color discrimination of Fig. 4 -30 and the difference -ofwavelength effect of Fig. 4 -31 become worse with increasing angles of
incidence.
185
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
100
-
- -- Perpendicular
Component
Parallel Component
75
50
25
0
400
500
700
600
Wavelength Inml
Courtesy RCA
Fig. 4-31. Influence of
light polarization.
The foregoing drawbacks are minimized by prism -type optics ( Fig.
4 -32) Note that the pickup tubes in this example are not parallel to each
other, but are "fanned." This arrangement minimizes the required angle
of incidence of all optics. The front part of the prism optical assembly contains a 1/4- wavelength plate, in diagonal position, such that the characteristics of perpendicularly polarized and parallel polarized light are averaged.
The effectiveness is dependent on having minimum angles of incidence of
the following reflective surfaces.
.
Compensating Lenses
Front -Surface Mirror
Prism Optics
Field Lens
Relay Lenses
Neutral Density Filter
%Trimming
Filters
Neutral-Density Filter
rapGreen
Objective Lens
Orbiting
Remote -Controlled
Optical Wedge
Iris
1.0.
Neutral- Density Filter
Trimming Filter
Front- Surface Mirror
Courtesy RCA
Fig. 4 -32. Optical system of RCA TK -41 camera.
TELEVISION BROADCASTING CAMERA CHAINS
186
Without this correction for polarization, the red and blue outputs are
reduced at certain critical angles of view and with variations in scenic
reflectance or light levels, leaving a predominantly green high light. This
is sometimes noticed as a greenish flesh tone of a person at (for example)
the left of the screen, while flesh tones in the rest of the scene are reproduced properly.
The primary advantage of the parallel pickup-tube assembly of Fig. 4 -28
is that the influence of any magnetic field will affect all tubes by the same
amount; hence compensation can be made easily with centering adjustments. The primary disadvantage is that this is not the optimum arrangement for an optical system employing a prism block, because of the long
back focal length required and the relatively large angles of incidence for
the red and blue reflecting surfaces.
One type of prism -block assembly is illustrated in Fig. 4 -33. With this
arrangement, the vertex angle of the central prism is reduced, decreasing
the angle of incidence to the red and blue tubes to approximately 30 °. The
tubes are fanned as in Fig. 4 -32. Fig. 4 -34 is a view of the Marconi Mark
VII color camera; note the fanned yoke assemblies.
Another type of prism-block assembly is shown in Fig. 4 -35. To gain
efficiency in the separation of red from green, the angle of the ray incident
on the red reflecting surface at the central point is only 13 °. The angle of
incidence onto the blue reflector is approximately 25 °. An air gap is provided between the blue- reflecting and red -reflecting prisms to reflect the
red component internally. The back focal length required with this assembly
is less than half that of the methods previously described. The fanning
angle of the pickup-tube yoke assemblies is greater than for the former
methods. Fig. 4 -36 illustrates the yoke configuration in the RCA TK -44
color camera.
Blue
Incident Lignt
-.Green
Prism Block
Red
Fig. 4-33. Prism -block assembly with 30° angle of incidence.
187
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
Courtesy Ampex Corporation
Fig. 4 -34. Interior view of color camera.
Fig. 4 -37 illustrates the zoom and focus controls normally mounted on
the RCA TK -42 color camera. These controls are supplied by Albion Optical Co. and are adaptable to all makes of cameras. In the RCA TK -42, the
zoom lens assembly is internal. The focus and zoom control cables feed
through two holes in the side of the casting below the camera as shown.
To Red
I
Incident Lignt
- .-To
Green
Fig. 4 -35. Prism -block assembly with 13° (red) and 25° (blue) angles
of incidence and reduced back focal length.
188
TELEVISION BROADCASTING CAMERA CHAINS
Courtesy RCA
Fig. 4 -36. Fanning of color -camera yoke assemblies.
illò
Item
1
2
3
Description
control
Twist-grip focus control
Universal mounting clamp
Two -speed zoom
5
Arm assembly
Focus control cable
6
Zoom control cable
4
Courtesy RCA
Fig. 4-37. Zoom and focus controls.
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
(
A
I
189
Side view.
(B) Front view.
Courtesy RCA
Fig. 4 -38. Camera with external zoom lens.
A two-speed lever permits fast or slow zoom control. In the fast position,
one turn of the handle zooms through the entire range. In the slow position,
slightly more than two turns of the handle are required for full zoom range.
The zoom lens assembly on the RCA TK -44 color camera is external
( Fig. 4-38 ) Thus, the focus and zoom control cables connect directly into
the zoom assembly.
.
TELEVISION BROADCASTING CAMERA CHAINS
190
The Philips PC -100 color camera (Fig. 4 -39) employs a lens- mounting
design that automatically couples internal mechanical drive shafts as well
as completing the electrical connections. There are no external lens cables.
Courtesy Philips Broadcast Equipment Corp.
Fig. 4 -39. Camera with quick- connect lens assembly.
4 -9.
THE
IRIS -CONTROL
SERVO
Most modern cameras employ remote control of the lens iris from the
operating panel. Servo functions for this purpose generally are located in a
servo module of the camera head.
Fig. 4 -40A represents the basic servo function. The motor is geared to a
follower potentiometer which turns with the motor rotation. The voltage
applied to the follower also is applied to a control potentiometer. When the
net voltage at the amplifier input is zero (equal and opposite voltages from
the two potentiometers), the motor cannot receive power. Assume the
control is turned toward a positive voltage. The amplifier now has an input
that is amplified and applied to the motor. When the motor rotates, the
follower rotates in the opposite voltage direction, in this case toward a
negative voltage. When the amplifier input again receives equal and opposite voltages ( net zero) , the motor stops. Bridge circuits normally are
involved in practice.
Fig. 4 -40B is a functional block diagram of an iris servo. When either
of the silicon controlled rectifiers is "fired," its associated diode bridge circuit is able to pass ac. When this happens, the circuit of the winding of the
two -phase motor is completed to ground directly through the bridge, and
191
PICKUP TUBE, YOKE ASSEMBLY, AND OPTICS
Input
Motor
+12.5
V
12.5
V
Follower
Control
(A) Basic action.
To^^ward^^f12
Jv
Firing -Pulse
DM Amp
Q1
Gen
Q3
SCR1
Toward f132
Diff Amp
Firing -Pulse
Gen
Q4
Q2
SCR2
Iris Control
+12V
12V
N\
Larger
AC Common
or/
Motor Not
V
Running
Amplitude
or
AC Hot
Motor Running
Follower
Coupled to Motor Shaft
(B) Typical circuit.
Fig. 4-40 Principle of lens -iris servo.
Motor Winding
TELEVISION BROADCASTING CAMERA CHAINS
192
the circuit of the other winding is completed to ground through the bridge
and a capacitor. The direction in which the motor runs depends on which
winding is in series with the capacitor, and this in turn depends on which
bridge is conducting (and therefore which SCR has fired)
When the inputs to the differential amplifier are equal, neither bridge
conducts. A change at the Q2 base causes an opposite reaction on Q1. The
change occurs because of operation of the iris control; one extreme of this
control is the maximum stop opening, and the opposite extreme is the
minimum stop opening. An opposite rotation of the iris control causes the
opposite SCR to fire, and the motor turns in the reverse direction.
The waveforms on the drawing give a clue as to how to signal trace in
probing for troubles. The two types of signals depend on the circuitry
involved; these are the most prevalent. Remember, you will always see the
60 -Hz power waveform on the scope at the point indicated whether the
circuit is completed to ground or not. The clue to completion of the circuit
for the motor is the increase in the amplitude clipping point or the presence
of the SCR "spikes" on the waveform.
.
EXERCISES
What
is the purpose of the "pancake" coil in front of the 41/2-inch
image orthicon?
Q4 -2. How is electrical focus obtained in the image section of the image
orthicon?
Q4 -3. How is focusing achieved in the scanning section of the image
orthicon?
Q4 -4. Does the multiplier section of the image orthicon contain any beam focusing element?
Q4 -5. What is the purpose of grid 5 in the image orthicon, and how is the
proper voltage adjustment for this grid determined?
Q4 -6. Compare I.O. curve B in Fig. 4 -6 with vidicon curve A in Fig. 4 -23C.
The I.O. amplitude response at 350 lines is approximately 70 percent.
The vidicon amplitude response at 350 lines is slightly less than 60
percent. Does this mean the resolution capability of the vidicon camera
is less than that of the image- orthicon camera?
Q4 -7. What adjustment in a vidicon camera generally assures good uniformity in black level (lack of picture shading) ?
Q4 -8. In the vidicon, what is the relationship between dark current and
target voltage?
Q4 -9. In the Plumbicon, what is the relationship between dark current and
target voltage?
Q4 -10. What is the difference between the Plumbicon and the lead -oxide
vidicon?
Q4 -1.
CHAPTER
5
Video Preamplifiers
Shunt circuit capacitances, both stray wiring capacitance and tube or
transistor interelectrode capacitance, inevitably result in a signal loss that
increases as the frequency increases. Therefore, if a video amplifier is to
have a "flat" frequency response, "peaking" circuitry or heavy amounts of
inverse signal feedback must be used. The coupling between the pickup
tube and the preamplifier input, since it occurs at high impedances, is also
a cause of high- frequency losses and phase shift.
Video -peaking circuitry to obtain a flat frequency response results in a
certain amount of phase shift across the video passband. Therefore, additional compensation circuitry is required for phase correction. Adjustable
controls for this type of signal correction are termed high- peaker or phase correction controls. Sometimes both terms are used for separate controls,
depending on the stage in which the correction is applied, and on the time
constant of the correcting network. High -peaking and phase-correction
controls normally are found in the video preamplifier, which receives the
signal output from the pickup tube for further amplification.
5 -1.
THE VACUUM -TUBE VIDEO AMPLIFIER
Vacuum -tube circuitry depends more heavily on peaking and phase correction than is true of the most recent solid -state amplifiers. Since there
are many tube -type amplifiers still in daily service, we would be remiss in
our duty if we omitted them from our studies.
Vacuum -Tube Peaking Circuits
Fig. 5 -1A illustrates the shunt -peaking method of compensating for the
usual high -frequency losses. In the equivalent circuit (Fig. 5 -1B), CT is
the total of the output capacitance of V1, the stray capacitance of the circuit, and the input capacitance of V2. Electrically, the circuit amounts to a
parallel resonant circuit, designed so that the resonant frequency is approximately 1.41 times the highest frequency to be amplified. Thus, a boosting of the high frequencies occurs, but there is no effect on the lower fre193
TELEVISION BROADCASTING CAMERA CHAINS
194
quencies. In practice, peaking coils may have values between one and several
hundred microhenries. Ten to fifty microhenries is the average range found
in commercial equipment.
Fig. 5 -2A illustrates the series type of peaking circuit. The series coil, in
combination with the effective circuit capacitances, forms a low -pass filter
network ( Fig. 5 -2B) . At first thought, it might appear that such a circuit
defeats the purpose intended, an increase in the efficiency of amplification
at higher frequencies. A basic analysis is therefore necessary.
Pickup
Tube
Tube
R
B
-p
(B) Equivalent circuit.
(A) Location of coil.
Fig.
5
-1. Shunt peaking.
It is necessary that Cl have twice the capacitance of C2. Load resistor RL
is always connected to the low- capacitance side of the circuit. In practice,
therefore, a small physical capacitor may be found in the circuit where effective capacitance Cl would appear; this physical capacitor is considered to
be in parallel with Cl, effecting a 2 -to -1 ratio in effective capacitances.
L
Series
L
Tube
C17.
Series
C2
RL
B+
B-Q
(A) Location of coil.
Fig. 5 -2. Series
(B) Equivalent circuit.
peaking.
Inductor L and capacitance Cl form a series resonant circuit with an
effective increase in current as the frequency increases. Capacitance C2 is
separated from Cl by inductance L, with a resultant reduction in shunting
effect across L at high frequencies in the passband. Therefore, since the
voltage drop in the series resonant circuit increases with an increase in
frequency, and this voltage is applied across load resistor RL shunted by
C2, the voltage developed in the load likewise increases for high frequencies
in the passband. The increase of high- frequency voltages as a result of the
VIDEO PREAMPLIFIERS
I
95
V2
Pickup
Tube
QB-
(A) Locations of coils.
(8) Equivalent circuit.
Fig. 5 -3. Shunt-series peaking.
resonant effect of L and Cl more than offsets the effect of decreased reactance of C2 with increasing frequencies. The resulting video voltage is
coupled to V2 in the usual manner.
Fig. 5 -3 illustrates the most efficient design used in video amplifiers to
increase the high -pass range. This is the shunt - series peaking circuit, which
can be seen to be a combination of the two methods just discussed. The
increased voltage gain from such a circuit is aided materially by the fact
that it is possible to use a load resistor of around 80 percent greater value
than can be used with a simple shunt -peaked circuit. Since the gain of a
stage is equal to the transconductance of the tube times the value of RL, it
may be seen that the gain is increased appreciably by this means alone. It
should be remembered that the value of plate load resistance in ordinary
amplifiers is limited by the bandpass required; too great a value of load
resistance reduces the bandpass capabilities of the stage.
The relative gains of video amplifier circuits may be tabulated as follows:
Uncompensated
Shunt Peaked
Series Peaked
Shunt -Series Peaked
0.707
1.0
1.5
1.8
Low frequencies are of just as much importance in the video amplifier
5 -4 illustrates the addition of a low- frequency
as are high frequencies. Fig.
Fig.
5
-4. Low -frequency compensation.
196
TELEVISION BROADCASTING CAMERA CHAINS
filter circuit to an amplifier stage. At low frequencies, the voltage across RG,
the grid resistor of V2, decreases because of the increased reactance of
coupling capacitor Cc. However, the reactance of CF also increases at the
lower frequencies, and the combined impedance of RF and CF is added to
that of load resistor RL. Thus, as low- frequency signals become somewhat
attenuated by the limited value of coupling capacitance, the total plate load
impedance becomes greater, and the resulting boosting effect helps maintain constant gain across the entire passband.
NOTE: The reader should be cautioned that the frequency- compensation
circuitry presently being discussed is for normal losses that result from
circuit characteristics. The aperture distortion previously discussed is a
loss of high- frequency definition without phase distortion. Correction for
this effect is made by aperture- correction circuitry in processing video
amplifiers (Chapter 6) .
Typical Vacuum -Tube Preamplifiers
The video preamplifier shown schematically in Fig. 5 -5 consists of eight
stages (six tubes) . The first four stages, V1 to V4, are video amplifiers
using Type 6AH6 tubes with shunt peaking in the plate circuit of each
amplifier. The cathodes of the first and fourth stages are unbypassed. RC
high -peaking networks in the cathode circuits of the second and third
stages, with a variable resistor in each cathode circuit, permit adjustment
of the response in two distinct steps. This peaks the higher frequencies to
compensate for high- frequency attenuation associated with the imageorthicon output.
The four feedback and output stages are built around two dual tubes,
a Type 6U8 pentode -triode (V5) and a Type 5687 dual triode (V6)
Feedback from one cathode (pin 6) of V6 to one cathode (pin 7) of V5
is adjustable by means of a 7 -35 pF variable capacitor (CT) to obtain flat
overall response for the feedback pair. The feedback circuit also provides
sending -end termination and isolation for the viewfinder output.
Horizontal shading (Chapter 6) in the viewfinder picture is corrected
by feeding a positive or negative sawtooth voltage to the cathode of V I.
The video preamplifier in Fig. 5 -5 has no gain control. The image orthicon output is controlled by varying the dynode gain by means of the
ORTH GAIN control. This allows adjustment for variations in image orthicons in order to keep the video input to the preamplifier constant.
Note that the preamplifier input signal from the image orthicon is about
0.05 volt peak -to-peak. Two outputs are provided: a 1.0 -volt peak -to -peak
signal for the viewfinder and a 0.4 -volt peak -to-peak signal to the 50 -ohm
coaxial section of the camera cable. The ORTH GAIN (dynode gain) control in the camera head is adjusted so that these output levels exist for high
lights in the scene.
Observe in Fig. 5 -5 the small values of the cathode -bypass capacitors in
the second and third stages. The greater the value of cathode-return resis.
VIDEO PREAMPLIFIERS
197
Courtesy RCA
Fig. 5 -5. Preamplifier in RCA TK -11 camera.
198
TELEVISION BROADCASTING CAMERA CHAINS
tance ( variable) , the greater is the degree of low- frequency degeneration
relative to higher frequencies. Sometimes a fixed cathode -return resistance
is used with variable bypass trimmer capacitors.
Fig. 5 -6 is a simplified schematic diagram of a more recently developed
tube -type video preamp. The signal voltage to feed the preamplifier input
is developed across R1, the I.O. load resistor. Note that this load resistor is
returned to ground for ac through Cl. This is normal in most I.O. output
coupling circuitry.
Shading or calibration signals generated in rack -mounted processors in
the control room are terminated by R4 (51 ohms) and fed into the bottom
of the I.O. load (R1) through Cl. These signals may be disconnected and
a test signal inserted across R2 (10k)
The video signal is amplified in the first cascode amplifier and then
coupled into the second cascode amplifier, which incorporates the high peaker circuit. The signal then is amplified further by two feedback pairs
as drivers for the output line driver.
Most modern plate-coupled, low- impedance output stages take the form
of the single -ended push -pull circuit, as shown for V7 and V8 in Fig. 5 -6.
This circuit provides the advantages of push -pull amplification while supplying a single -ended output for the coaxial cable. Typical values of output
coupling capacitors range from 200 to 400 F.
The cascode input stage for video preamplifiers has become quite common in both tube and solid -state circuitry. The preamplifier and frequencyphase compensation methods are similar for all pickup tubes whether image
orthicon, vidicon, or lead-oxide vidicon.
.
5 -2.
SOLID -STATE VIDEO PREAMPLIFIERS
A solid -state video preamplifier frequently used in vidicon channels of
RCA color cameras is illustrated in Fig. 5 -7. The input amplifier unit and
preamplifier module provide current amplification for low -level signals
from a vidicon tube to drive the 93 -ohm input of the following video
amplifier stage.
The input amplifier, consisting of a field -effect transistor (FET) and a
resistor, is a separate unit that is designed for mounting on a vidicon yoke
assembly, thereby minimizing input circuit capacitance. The output from
the FET amplifier provides the input signal for the preamplifier module.
When used together, the input amplifier and the preamplifier function
as a signal amplifier with selectable current gain of 50, 100, or 150. The
video output is passed through a peaking network in the preamplifier
module to compensate for input source capacitance that causes high frequency attenuation of the video input signal from the vidicon. Ac feedback from the preamplifier to the input amplifier is employed to cancel
the capacitive loading effect on the transistor elements; dc feedback is employed for bias stabilization.
199
VIDEO PREAMPLIFIERS
fro
F
á
{,
U
-ev\e's,-
ó
E
Ú
Ñ
od
'
1
4
>
_
O
8bo
FO
C
Q
+
O
Q
V=1
'"'
°cc
"
1
co
L
Fig. 5 -6. Simplified diagram of tube -type video preamplifier.
200
TELEVISION BROADCASTING CAMERA CHAINS
aL
O
Os.
8 Ñ
r+
C
In
v
vH
S
.17
E
co á
'v
>
Courtesy RCA
Fig.
5
-7. Simplified diagram of video preamplifier for vidicons.
VIDEO PREAMPLIFIERS
201
A field -effect transistor (FET) , which has characteristic high input impedance and low noise figure, operates as a current amplifier in the input
amplifier. A 4.7- megohm resistor in the input gate circuit provides the
path for degenerative video- signal and bias -stabilization feedback.
The signal current from a vidicon target (constant- current device) is
passed through the gate-biasing resistor. The signal voltage developed
across the resistor is applied to the gate (G) terminal of the FET, setting
up within the transistor space charges that vary with the signal. These
varying space charges modulate the current in the path between the drain
(D) and source (S) terminals. The result is amplification of the input
signal current without polarity inversion.
The output from the field -effect transistor is coupled to cascode amplifier
Q1 through the input amplifier cable. Transistor Q1, a grounded -base amplifier with low input impedance and high output impedance, provides
neutralization of the degenerative capacitance effects between the drain and
gate materials of the FET. The video output signal from Q1, developed
across resistor RL and high- frequency peaking coil L1, is applied to the base
of Q2, an emitter follower with a high input impedance and a low output
impedance. The video signal at the emitter of Q2 is amplified and inverted
by Q3, and is applied to a feedback amplifier pair, Q4 and Q5.
One output from the emitter of Q5, the degenerative video feedback
signal, is coupled through C4 and R4 to the gate terminal of the FET. A
second output from Q5 is coupled through C5 and R5 to a peaking and
gain -selection network. High -frequency peaking trim is adjusted by means
of capacitor Cl to compensate for losses caused by vidicon - target capacitive
loading at the input to the input amplifier. Gain -selection switch Si sets the
current gain at 50 when in position 2, at 100 when in position 1, and at
150 when in position 3. The high- frequency -peaking trim requires adjustment when the gain is changed.
Note that when S1 is in position 2, the output is through R1 and high peaker trimmer Cl. When SI is in position 1, R2 is paralleled with R1.
Since the resistances of R1 and R2 are equal, the current gain is doubled.
When the switch is in position 3, R3 is paralleled with R1; the value of R3
is such that the current gain is triple that for position 2.
Capacitors C2 (across R2) and C3 (across R3) are compensating capacitors to help maintain the proper frequency -phase response for the various
switch positions. However, as stated before, the high peaker (C1) normally
must be readjusted when the switch position is changed.
The feedback pair (such as Q4 and Q5 in Fig. 5 -7) is a common circuit
in modern cameras, where it finds use particularly in drivers for 50- or 75ohm lines. The voltage amplification is approximately equal to the ratio of
feedback resistance to signal -source resistance (Fig. 5 -8) . In Fig. 5 -7, the
source resistance for the feedback pair is the effective output impedance
of inverter stage Q3. Since the emitter resistance (RE) is large and degenerative (unbypassed) , this source resistance (Rs) is essentially the value of
TELEVISION BROADCASTING CAMERA CHAINS
)0)
Rf
AV
RS
Fig. 5 -8. Typical feedback pair.
Load
Build -Out
Resistor
500
ax
RL
50 to
75 Ohms
(Coaxial Cable Termination)
Rc, or 1000 ohms. Thus, the voltage gain expected in the feedback pair is
5100/1000, or 5 (approximately) . Both the input and output resistances
are low so that capacitive effect at high frequencies is negligible.
Observe that in both Fig. 5 -7 and Fig. 5 -8 a resistor is placed in series
with the load feed. All modern video -amplifier outputs provide a source
impedance equivalent to the load impedance. Let us examine now why we
normally find a build-out resistor in the output stage of a coaxal line driver.
Recall that the emitter -follower circuit is characterized by a relatively
high input impedance and a relatively low output impedance. Therefore,
it is a "natural" for feeding low- impedance video lines. The input and output impedances of an emitter follower ( feeding a low impedance load) are
somewhat interdependent. This is because the input load is part of the
output, and the outut load is part of the input. If the reader will observe
Figs. 5 -7 and 5 -8, he will note that the same is true of the feedback -pair
circuit.
Observe Fig. 5 -9A. The approximate internal output impedance of this
circuit, as given on the drawing, is the source impedance divided by the
sum of beta of the transistor plus 1. Assuming a beta of 50, we can see that
this output impedance becomes 680/51, or 13, ohms (approximately) .
For the coaxial -line termination of 50 ohms to "see" a sending-end impedance of 50 ohms, a 37 -ohm resistor must be used as shown in Fig. 5 -9B.
Now see Fig. 5 -9C. In this circuit, emitter follower Q2 is driven by
emitter follower Q1. Assume that Q1 sees a source resistance of 1000 ohms
and that beta is 50 for both transistors. The output resistance of Q1 is:
ZorT
50
+
1
= 20 ohms (approx)
Then the output resistance of Q2 is:
ZoTJT
= 5020+ 1 = 0.4 ohm (approx)
VIDEO PREAMPLIFIERS
203
The equivalent circuit of the output stage is shown in Fig. 5 -9D. To
feed a 50 -ohm load, a build -out resistance of 50 ohms is needed so that the
effective internal output impedance (looking back from the load) becomes
50 ohms. With such a low- impedance transistor output, it is simply necessary to use a 75 -ohm build-out to feed a 75 -ohm coaxial line. In either case,
the build-out and load combine to form a 2 -to -1 voltage divider. Therefore, the signal voltage at the emitter of Q2 is twice as great as the signal
voltage across the coax -line termination ( input to following video amplifier). This is a normal condition for modern video line drivers, and it
should be understood in particular by the maintenance personnel.
R
Source
Impedance
lApproxl
ZollT
Build-Out
Resistor
Camera
RS
Cable Coax
MOO
370
R1500
47051
(A) Single emitter -follower stage.
j130
500
(B) Connection of A to line.
t
O.
(C) Two emitter -follower stages.
500
44
(D) Connection of
500
C to line.
Fig. 5 -9. Drivers for coaxial lines.
The ideal phase -frequency characteristic is a straight line that rises
from a value of pi radians (180° ) at zero frequency (frequency plotted
along the x axis and phase angle plotted along the y axis). The scale is
linear, which indicates that the phase angle increases linearly with frequency, maintaining the same time delay for all frequencies. Departure
from this characteristic results in phase distortion that is most noticeable
as trailing white edges following black edges on a gray or white background.
If the phase characteristic departs from the ideal straight slope and bows
upward away from the frequency (x) axis, the increasing slope with
frequency is an indication that the time delay is increasing with frequency.
The shifted components making up the pulse now add in such a manner
TELEVISION BROADCASTING CAMERA CHAINS
204
a leading -edge overshoot and a trailing -edge undershoot on the passed
pulse occur. Conversely, if the phase characteristic bows toward the x
axis, the decreasing slope with frequency indicates a decrease in time delay
with frequency. The shifted pulse components now add to produce a leading -edge undershoot and trailing -edge overshoot. Note that one type of
phase distortion may be compensated by an equal and opposite phase correction. A lagging phase shift is corrected by an equal leading phase shift,
and vice versa.
that
C
(B) Peaking and phase controls.
(A) Conventional emitter peaking.
Emitter
Follower
Phase
0 47
Streaking
1
270
1
Q2
meg
High Impedance
5k
(C) Phase -split correction method.
Fig. 5 -10. Representative high -peaking and phase circuitry.
In Fig.
5 -10A,
the emitter resistor forms an inverse- feedback path that
is incompletely bypassed by the small value of capacitance. Signals of
higher frequencies are bypassed, but signals at lower frequencies must pass
through the inverse feedback path. Thus, gain is reduced at lower frequencies but not at higher frequencies.
In Fig. 5 -10B, the LC circuit is resonant near the top of the intended
passband. The variable resistance (peaking control) determines the magnitude of the resonant peak introduced. The variable trimmer capacitor
(streaking control) adjusts the resonant frequency to introduce the required lead or lag in phase.
VIDEO PREAMPLIFIERS
205
In Fig. 5 -10C, Q2 is fed from both the emitter and collector of Q1.
Thus, Q2 is driven from signals 180° apart at high frequencies. Phase
changes in Q1 itself are prevented by the low values of the emitter and
collector resistors. Transistor Q2 presents a high impedance to Q1 and
the coupling networks. Note from the relative values of the coupling
capacitors that low- frequency gain is higher in the collector circuit than
in the emitter circuit, to maintain amplitude response independent of
frequency under adjustment of the phase control. The relative phase shift
at high and low frequencies is adjustable by means of this control.
5 -3.
MAINTENANCE OF VIDEO PREAMPLIFIERS
We are now ready to study procedures for maintenance and adjustment
of the video preamplifier. Older tube -type preamps require a more consistent maintenance schedule than do the latest solid -state versions. In fact,
some of the most recent solid -state amplifiers have few, if any, adjustable
peaking coils in the circuitry. This is because of the very high gain -bandwidth product of transistors and integrated circuits in the latest designs.
Also, the stability of performance has been improved greatly over that
possible in vacuum -tube circuitry.
Preparation for Video Sweep
Older tube -type video preamps normally require a video -sweep alignment on regular preventive -maintenance schedules about every 90 to 120
days. In any case, the maintenance technician should be familiar with
video -sweep techniques. Even newer solid -state amplifiers require such testing whenever components are replaced.
Preliminary steps are necessary before video sweeping is attempted.
This is particularly true for those stations still using the older TV oscilloscopes, such as the Tektronix Model 524. First, we will review briefly
the practical use of the oscilloscope with normal probes and with the
video -sweep detector probe.
Because of capacitive loading effects, the direct scope probe is severely
limited in its application to TV equipment maintenance, even when it is
applied directly across 50- or 75 -ohm terminations. A direct probe should
never be used where frequency response or transient response is a factor;
therefore, use of this type of probe is limited to certain applications in
which the IRE response is employed.
For most applications, the 10 -m -1 capacitance divider probe should be
used. For a scope with 1- megohm input shunted by a 40 -pF capacitance,
the simplest 10 -to -1 probe consists of a series -connected 9- megohm resistor
shunted by a trimmer capacitor of 3 to 12 pF. When the probe is connected to the scope, the input impedance from the probe tip becomes 10
megohms shunted by approximately 12 pF. The trimmer capacitor is adjusted so that the RC product of the probe is equal to the RC product of
TELEVISION BROADCASTING CAMERA CHAINS
206
the scope input, thus making the voltage division independent of frequency. This is done by touching the probe to the scope calibration -pulse
output (or the output of a square -wave generator set to about 1 kHz) and
adjusting the trimmer so that the leading edge of the pulse is not rounded
on the top (undercompensated) and does not overshoot (overcompensated). This adjustment should be checked often, and it always must
be checked when the probe is used with a different scope, even one of the
same make and model. The frequency response and transient response of
the scope itself should be checked with this probe so that all variables are
calibrated.
1N96A
1N34A
1
Probe
00- 0 pF
1N34A
82-
47k
Probe
Scope
I
0 pF
22k
820pF
1N96A
Ç43O pF
100k
To
S cope
O
(B) 50- percent output.
(A) 75- percent output.
Fig. 5 -11. Video -sweep detector probes.
A logical step-by -step initial calibration of the scope can be outlined
as follows:
1.
Video Sweep (Detected) . Terminate the video -sweep generator in
75 ohms directly at the generator output connector. Use a video detector probe (Fig. 5 -11). The probe in Fig. 5 -11A will read approximately 75 percent of the actual peak -to -peak output signal,
whereas the higher-isolation probe in Fig. 5 -11B will read about 50
percent. Adjust the output amplitude for 1 volt, which will read
approximately 0.75 volt with the probe of Fig. 5 -11A or 0.5 volt with
the probe of Fig. 5 -11B. Also, adjust the scope gain to provide a
convenient display (Fig. 5 -12) . This makes possible a check of the
flatness of the sweep generator itself, since the detected sweep envelope does not depend on the high-frequency response of the scope.
The Tektronix scope may be used on any response position; or a
scope with limited response can be used, provided it has reasonably
good low -frequency square -wave response. If the video -sweep generator cannot be made perfectly flat, as observed on the scope, the
Fig.
Detected video -sweep
signal waveform.
5 -12.
207
VIDEO PREAMPLIFIERS
(A) Scope on normal response.
(B) Scope on fiat response.
Fig. 5-13. Display of undetected video -sweep signal.
deviations must be plotted as a correction factor for equipment
checks and scope calibration.
2. Video Sweep (Wideband). (This should be observed only after
determining the flatness of the sweep generator as in Step 1.) Although this method is used only in very special cases (and with
extreme care) , the rf envelope may be observed directly without detection as a "quickie" check on scope-amplifier response (Fig. 5 -13).
This check, however, is valid only if the probe (10 to 1) to be used
for equipment checks is used on the scope and a signal of the same
amplitude .is employed so that the scope compensated attenuator is
at the same setting as that to be used. It is good engineering practice
to run these checks with all probes in stock, and through the scope
preamp as well as directly to the vertical-amplifier output. Use varying levels from the sweep generator to permit use of convenient
scales on the scope with different attenuator positions that might be
incorrectly compensated. An attempt to employ correction factors for
different attenuator settings becomes both cumbersome and inaccurate in system measurements. Normally there will be some correction factor when using the preamp and when feeding the vertical
amplifier directly. Plot these responses either on a graph or by tabulation in peak -to -peak values. Normally the detector probe is employed when video sweep is used. The wideband display provides
a quick check of scope response to single- frequency sine waves or in
similar applications.
3. Single- Frequency Sine -Wave Checks. The most accurate method of
checking scope- amplifier frequency response is to run single -frequency sine -wave checks over the range of 100 kHz to 10 MHz. The
same generator and probes that will be used for system checks
should be used for scope calibration. Commercial sine-wave generators such as the Hewlett- Packard 650-A incorporate a frequency compensated metering circuit at the output to aid in maintaining a
constant input to the scope or equipment at all frequencies. If a
generator of this type is not available, a VTVM with good response
to 10 MHz can be used across the terminated generator output. As
208
TELEVISION BROADCASTING CAMERA CHAINS
in Step 2, it is good practice to check all probes and all attenuator
settings that are likely to be used in system checks. When the calibration is posted on the scope, the particular generator, meter, and
probe should be identified, unless all such items have been found to
be directly interchangeable. Such checks normally should be made
about twice a year, or at any time that considerable maintenance ( tube
or component changes, etc.) on the scope or signal generators has
been required.
4. Low -Frequency and Transient Response. Determine the rise time
and percent of overshoot of the square wave as read on the scope,
both through the preamp and directly into the main amplifier, at the
frequencies normally used. Unless a generator with short rise time
is available, higher- frequency square waves (above 75 kHz) are
not particularly useful, because for response checks at the higher
frequencies the rise time of the pulse must be shorter than the rise
time of the amplifiers to be checked. A 60 -Hz square wave fed to the
Tektronix 524 AD (dc position) should have an absolutely flat top,
as shown in Fig. 5 -14A. Fig. 5 -14B shows the normal amount of
tilt introduced by the input coupling capacitor when the scope selector is on the ac position. Remember that the last two (highest gain) positions of this particular scope are ac only, since the preamp
is used on these positions. An adjustable grid time constant (low frequency compensation control) is used in the preamp; this control
should be adjusted according to the manufacturer's instructions.
When a scope employs either external or plug -in preamps, always
include these units in all scope- calibration procedures.
NOTE: More recent Tektronix television oscilloscopes, such as the Type
545 (30 -MHz bandwidth) or Type 547 (50 -MHz bandwidth) , are dc
coupled even at the highest gains. These scopes require checking only
about once a year unless obvious performance deterioration is noted.
It is possible for a wideband scope amplifier to exhibit a leading -edge
overshoot caused by a vacuum-tube defect known as cathode interface.
This low- frequency phase shift results from series -resistance and capacitive-
(A) 60 Hz, scope on dc position.
(B) 60 Hz, scope on
Fig. 5 -14. Square -wave response patterns.
ac position.
209
VIDEO PREAMPLIFIERS
bypassing effects of a chemical interface layer that forms between the
sleeve and oxide coating of the cathode. Since some tubes have been
known to develop cathode interface in less than 500 hours of operation,
the scope should be checked about every two months for this type of tube
defect. The following procedure may be used:
1.
Adjust the frequency of the square -wave generator to 500 kHz. The
waveform should have a rise time of 0.2 microsecond or less.
1
Microsecond --a
Overshoot
Duration
Fig.
5
-15. Interface distortion of
square wave.
2
2.
Microseconds
Adjust the time base so that several cycles of the square wave are
displayed. If an overshoot with a duration of 0.2 to 0.6 microsecond
appears (Fig. 5 -15), one or more tubes in the vertical amplifier may
have cathode interface. (Overshoot duration, or time constant, is the
time required for the overshoot to decay to the final flat -top value.)
A 500 -kHz square wave completes one cycle in 2 microseconds; thus
it has a pulse width of 1 microsecond, as shown in Fig. 5 -15. The
overshoot duration normally is between 20 and 60 percent of the
total pulse width when cathode interface is present. As a double
check, plug the scope into a variable autotransformer and increase
the line voltage to the upper limit allowed. If cathode interface is
present, the increased tube -heater voltage will reduce the overshoot; a
decrease in the voltage will increase the overshoot. When this occurs,
it is best to replace all tubes in the vertical amplifier with new ones;
then substitute the old tubes one at a time while observing the square
wave. Discard any tube that tends to show this effect. Leave the line
voltage at the lower limit (105 -108 volts) to emphasize the effect of
cathode interface.
210
TELEVISION BROADCASTING CAMERA CHAINS
Video Sweeping the Preamp
A sweep generator consists of a fixed -frequency oscillator the output
of which is beat with the output of a sweep oscillator that is frequency
modulated at 60 Hz. The frequency modulation is such as to cause the beat
frequency to swing over a usable range from about 100 kHz to approximately 10 or 20 MHz. The frequency swing may be produced by a reactance -tube circuit with 60 -Hz excitation from the power line, or, in many
cases, by a motor -drive capacitor in the oscillator tank circuit. A 3600 -rpm
motor provides a 60 -Hz sweep of the oscillator frequency. Such a sweep
generator usually incorporates an absorption marker generator that places
a notch (or series of notches) at any reference frequency (or frequencies)
over the usable range.
The fundamentals of checking the high -frequency characteristics of a
video amplifier are illustrated in Fig. 5 -16. The output frequency of the
sweep generator is swept over a range of 100 kHz to 10 or 20 MHz, with
a tunable frequency maker (notch) placed at any desired frequency. The
sweep is repeated 60 times per second. This test signal is applied to the
amplifier to be tested. A detector of the type shown in Fig. 5 -11 is connected to the output of the amplifier. This detector rectifies the signal output as shown (in this case the amplifier is considered to be theoretically
ideal: no distortion has occurred) , and the output of the detector is fed
to the vertical input of the oscilloscope. By this means, the oscilloscope
traces a graph of output voltage versus frequency over the passband above
100 kHz. The scope for this test should have excellent low- frequency
response so that no distortion of the 60 -Hz square wave takes place. High frequency response need extend no farther than 50 kHz.
It is very important not to overload the amplifier ( s) when using video
sweep. If the normal output is a 0.7 -volt (peak to peak) signal, feed just
enough input sweep level to result in a 0.5 -volt (peak to peak) amplifier
output. Remember to calibrate the detector probe so that you know how
much loss occurs in the probe. For example, if the detector -probe gain is
50 percent, a 0.5 -volt (peak to peak) actual output level reads 0.25 volt
( peak to peak) through the probe.
Some engineers prefer to check the output level by using the regular 10to-1 probe with the scope in the wideband position (Fig. 5 -13) . When
using this method, adjust the sweep-generator output so that the 2 -to -3
MHz region of the sweep signal is at reference level at the output.
When the more modern oscilloscopes with 30- to 50 -MHz bandwidths
(flat response across the passband of interest) are used, the detector probe
need not be employed if care is taken. Even here, however, it is advisable
to use the detector probe to minimize hookup and grounding problems.
Most modern video-sweep generators for broadcast service have an internal output impedance (sending -end impedance) of 75 ohms. In making
response adjustments, which may require feeding an interstage circuit, the
VIDEO PREAMPLIFIERS
Fig.
5
211
-16. Method of testing video amplifier for high -frequency characteristics.
212
TELEVISION BROADCASTING CAMERA CHAINS
video -sweep generator feeds a high impedance unless the particular instruction manual for the unit specifies otherwise.
In checking individual units, the coaxial output cables should be disconnected and replaced with 50 -ohm terminations (for preamplifiers) . The
detector probe is placed directly across the termination and retained in this
position for response alignment. NOTE: When the video preamplifier
feeds a following amplifier in the camera head (rather than the camera
cable) , the output impedance may be something other than 50 ohms. A
93 -ohm coaxial cable sometimes is used.
Let us assume we are going to sweep and align the preamplifier of Fig.
5 -5. The output coaxial cable at J3 is removed, and J3 is terminated with
a 50 -ohm resistor (actual value is 51 ohms) The scope probe or detector
probe is connected across this termination, where it remains throughout
the alignment procedure. The alignment procedure may be outlined as
follows:
.
1.
2.
3.
4.
5.
Feed the sweep-generator signal to the grid (pin 2) of V5. Adjust
the sweep generator (in this example) to obtain a signal of no more
than 0.4 volt at the termination feeding the scope.
Note that the only adjustable component in the last two stages is
the trimmer capacitor (CT) . Adjust this capacitor for the flattest response possible.
Feed the sweep signal to the grid (pin 1) of V4. Reduce the generator output to obtain the reference level at the output. Adjust L4
for flattest response.
Note now that the grid of V3 (where the sweep -generator signal is
to be fed next) has a dc component from the return to the high side
of the high -peaker control. Therefore, it is necessary to feed the
sweep signal to this grid through a capacitor of about 0.1µF. This
capacitor is needed because most video-sweep generators do not employ a blocking capacitor at the output, since the actual output normally is through a built -in pad to obtain the desired level.
Also in this stage is the frequency -phase compensation network
associated with the high -peaker. This circuit must be bypassed
temporarily with a capacitor of 0.25 to 0.47 p.F. The final adjustment of high -peaker and phase controls must be made only with
pickup -tube signals to eliminate smear or streaking in the picture.
The purpose of video alignment is to make the amplifier response
flat, by using a video -sweep signal or single- frequency sine -wave signals, with any special phase- correction circuitry bypassed. Peaking
coil L3 is now adjusted for flattest sweep.
Note that the conditions for the circuit of V3 also apply to the circuit of V2. Bypass the cathode of V2 with another 0.47 -µF capacitor,
leaving the bypass for the V3 cathode in place. Feed the grid of V2
through a 0.1 -µF capacitor, and adjust L2 for flattest response. Re-
213
VIDEO PREAMPLIFIERS
member to reduce the gain of the video -sweep generator in all steps
to maintain the reference output level.
6. This will be a rough adjustment of the phase and high -peaker controls, which will be adjusted finally with a pickup -tube picture as
mentioned above. Remove the temporary bypass capacitors at the
V2 and V3 cathodes. Feed the sweep -generator signal to the grid of
VI through the circuit of Fig. 5 -17B (described below) Adjust
L1, R2, and R4 for proper frequency response.
.
If circuits such as the cathode circuits of V2 and V3 (Fig. 5 -5) are not
bypassed with a capacitance of around 0.47µF, the video sweep through
the amplifier is distorted (Fig. 5 -17A) and is meaningless. There is practically no response below about 2 MHz. This is where the term "high peaker" originates, but it is a misleading term. To see the actual effect on
I
I
I
Sweep
Input
I
I
I
6.2k
I
Sweep
I
Input
1
6.2k
5.6k
To
75
ß
47
PF
Video
Amp Input
Sweep Output
(A) Effect of unbypassed cathode
(B) Typical image- orthicon circuit
network on sweep.
simulator.
'
High-Peak
1--.
s-.T.
Ideal
7
MHz
Phase
(C) Effects of phase and high -peak controls.
Fig. 5 -17. Modifications in sweep response.
frequency response, you need to feed the video -sweep signal through an
image -orthicon circuit simulator, an example of which is illustrated in Fig.
5 -17B. Note that the so- called "high peaker" affects only the very low end
of the video sweep (see Fig. 5 -17C, which shows the amplifier output
with the video -sweep signal fed through the I.O. circuit simulator) . The
"phase" control in this instance affects a higher -frequency range up to
around 5 MHz. The main purpose of such controls is phase correction
made necessary by the image-orthicon input resistance -capacitance network. Obviously, the same type of correction is required for vidicons or
any other type of pickup tube.
214
TELEVISION BROADCASTING CAMERA CHAINS
Output
150
150
180 pF
Cl
C2
1200pF
R4
High-Peaker
Fig. 5 -18. Phase and high -peaker controls.
You can get a rough idea of the intended range of effect on the length
of a smear (or streaking) by noting the maximum time constant of the
correction networks (Fig. 5 -18) For example, the maximum time constant
of the phase network (R1, R2, C1) is 0.12 microsecond. Consider the
active line interval to be 53 µs (63.5 p.s minus the horizontal -blanking
time). Note that 0.12 µs is approximately equal to 0.00226 of 53 µs.
Assume the raster is 18 inches wide. Then:
.
0.00226 X 18
= 0.04
inch, or about 3/64 inch
This type of circuit with this time constant corrects short trailing smears.
The maximum time constant of the high -peaker network (R3, R4, C2)
is about 2:5µs. This is approximately 0.047 of the raster width. Then on
a raster 18 inches wide:
0.047 X 18 = 0.846 inch, or about 13/16 inch
Therefore, this time constant corrects long streaking of trailing edges.
All pickup -tube preamps do not have the same specified frequency
response. Usually, the luminance -channel preamp will be specified as
"flat" within 1 dB to around 7 MHz. In 4- channel cameras, the charinance tubes have preamps with a much narrower bandwidth, such as to 4
MHz, whereas the luminance channel has response to 7 or 8 MHz. Remember that in the 4- channel camera, the luminance channel carries the high resolution information, and the color channels provide a relatively broad
"paint brush" that requires less high- frequency response.
Figs. 5 -19 through 5 -22 illustrate various detected sweep curves. Fig.
5 -23 identifies the components discussed in connection with these curves
in the following paragraphs.
If plate -load resistor RL (Fig. 5 -23) should increase from the normal
value, the trace obtained would appear similar to curve 2 in Fig. 5 -19.
Remember that a higher value of coupling resistance causes a departure
215
VIDEO PREAMPLIFIERS
v
Response Curve of Properly
Functioning Video Amplifier
O
Load Resistor Too Large
Load Resistor Too Small
O
100 kHz
8
MHz
10 MHz
Fig. 5 -19. Effect of load resistor on demodulated sweep -generator test signal.
from flat response at both high and low frequencies. In this case, we are
observing the high passband from 100 kHz to 8 MHz, and the droop
toward the upper end of the band is noticeable. If the slope is very pronounced, phase distortion is bound to occur, and loss of resolution is apparent in the picture. Although this effect might be caused by an actual
change in value of RL, this is not necessarily the sole cause. Anything
that would affect the dynamic plate load so as to increase its effective impedance over the passband would have the same result. For example, observe curve 2 in Fig. 5 -22. This is essentially the same trace as curve 2
of Fig. 5 -19, and it is caused by reduced inductance of the shunt peaking
coil. The use of the proper value of shunt peaking coil allows a higher
value of plate load resistor to be used than when peaking is not employed.
Thus reduction of the inductance of this coil results in a condition similar
to that of curve 2.
Understanding of these basic circuit relationships materially aids the
maintenance engineer in interpreting resultant scope traces. Curve 3 of
Fig. 5 -19 is a typical trace when RL has decreased from its normal value.
Observation of curve 1 in Fig. 5 -22, for which Ll is larger than the
optimum value, reveals the effect on plate -load impedance at the higher
frequencies, effectively decreasing the value of RL. The traces are therefore
similar in appearance.
The effect of the series -peaking coil is shown in Fig. 5 -20. When the
series coil is larger than the optimum value (curve 1) , a gradual upward
slope occurs from 100 kHz (low end of sweep) toward the mid -range of
sweep. The larger the value of inductance, the farther the hump is shifted
to the left. Compare this with curve 3 in Fig. 5 -19, which indicates RL is
too low in value. The major difference in the resulting traces is the extremely reduced cutoff level (at start of maximum downward slope of
I
216
TELEVISION BROADCASTING CAMERA CHAINS
Series-Peaking Coil Too Large
- -- Cutoff
I
I
I
I
Series- Peaking Coil
I
100 kHz
6
MHz
Too
Small
--Cutoff
10 MHz
Fig. 5 -20. Effect of series -peaking coil on response curve.
curve) indicated in Fig.
5 -20 (series coil too large) . This causes the
notch to "slide down" on the sloping portion of the curve at the high end.
From the slope of this curve, it may be inferred that the effective load is
reduced as the series -peaking coil is increased in value, just as in the case
of the shunt- peaking coil. Increasing the value of the series coil lowers the
resonant frequency (greater LC ratio) at which the series coil performs.
This causes the hump in amplitude response to move to the left (lower in
frequency) , and the effective load at the highest frequencies in the desired
passband is reduced.
If the series peaking coil is too small in value (curve 2 in Fig. 5 -20) , the
response from 100 kHz to midrange of the sweep is too large, and the
amplitude at cutoff is increased. This condition also may be caused by a
reduced value of damping resistor shunted across the series coil.
The effects of increased values of damping resistance are shown in
Fig. 5 -21. Curve 1 is displayed when the resistance value has increased to
the point at which insufficient damping of the resonant peak occurs. This
Damping Resistor Increased in Value
O
Damping Resistor Open
Fig.
5
-21. Effect of damping resistor on response curve.
VIDEO PREAMPLIFIERS
'/\
O
Fig.
217
5
Shunt Peaking Coil
Too Large
Shunt Peaking Coil
Too Small
-22. Effect of shunt -peaking coil on response curve.
is one possible cause of transient oscillation. Curve 2 indicates an open
damping resistor that allows the resonant peak to appear.
The peaking circuits of most video amplifiers are adjustable, as indicated
by the variable inductances in Fig. 5 -23. The proper alignment of these
stages constitutes an important function of the maintenance engineer both
in initial setup of amplifiers and in routine and priority checks of equipment. With the sweep generator connected at point 1 in Fig. 5 -23, the
effects of varying the adjustments of L1 and L2 may be noted. It will be
observed that varying series -peaking coil L2 mostly affects the trace at the
right of the pattern, and varying L1 mostly affects the trace through the
center of the pattern. The marker -notch frequency should be set so that
it appears at approximately the assumed limit of the fiat portion of the
curve, such as 7 or 8 MHz. If, on varying L2, the peak starts moving to the
left, the adjustment should be made in the opposite direction to obtain as
flat response as possible. Similarly, while the scope pattern is observed,
L1 is adjusted to obtain the ideal response characteristic. No more than
approximately 2 percent variation should occur from 100 kHz to the limit
of the flat portion of the curve. Always compare results with manufacturer's specifications.
The typical traces shown in Figs. 5 -19 through 5 -22 assume only one
component fault, as is usually the case in preventive maintenance or in
LI
Fig.
5
-23. Components for shunt and
series peaking (Figs. 5 -19)
through 5 -22).
Point
Shunt
Peaking
Coil
12 Series
Peaking Coil
1
Damning
Resistance
TELEVISION BROADCASTING CAMERA CHAINS
218
trouble occcurring during operation. If a number of amplitude variations
show up in the pattern, several defects may exist simultaneously. In this
case, the engineer familiar with the effect of any given adjustment on the
corresponding pattern will establish a basis from which to proceed. It is
important that the setup of test equipment and test leads produce no
spurious response on the screen. Experience with any particular installation is necessary before the engineer can readily determine whether a
trace is normal or abnormal.
If a great number of pronounced "humps" or "wiggles" is observed on
the scope screen, there probably is a poor ground connection. Always use
the shortest possible ground leads in video -sweep measurements. If changing a ground connection to a different point changes the pattern, the
grounding arrangement is faulty.
Adjustment of High -Peaker and Phase Controls
As emphasized earlier, final adjustments of the high -peaker and phase
controls are made with the camera looking at a test pattern. The horizontal
bars in the standard test pattern serve as a reference for adjusting preamp
controls. Fig. 5 -24A shows slight positive streaking (black after black
or white after white) , and Fig. 5 -24B shows negative streaking (white
after black or black after white). Fig. 5 -24C illustrates the appearance of
(A) Slight positive streaking.
(B) Negative streaking.
(C) Severe negative streaking.
Fig.
5
-24. Examples of streaking.
219
VIDEO PREAMPLIFIERS
lettering under severe negative- streaking conditions. A condition this
severe may originate in the preamp or in following video amplifiers that
incorporate clamping circuitry (Chapter 6)
It may be asked how "anticipatory" streaking can occur before (for example) a white area in the scene, causing streaks all the way from left to
right on the raster as in Fig. 5 -24C. In Fig. 5 -25A, the build -up of low frequency response causes a gradual decline toward black after the white
.
White
White
I
Black
.,J
Streaking
Region of
-
Black
Region of Streaking
(A) Waveform for positive streaking.
(B) Waveform for negative streaking.
White Window
(C) Severe negative streaking.
Gray Background
I
!Black Streaking Across Entire Raster
--
-White
I
(D)Waveform for condition in C.
Fig.
5 -25.
Ref Black
Conditions of streaking
signal, resulting in white streaking of the white image. In Fig. 5 -25B, the
loss of low- frequency response causes an overshoot into the black region
on the trailing edge of the signal, causing black streaking after the white
image. These waveforms are representative of "short streaking" (as in
Figs. 5 -24A and 5 -24B). Control of such streaking normally is within the
range of high -peaker and phase controls in the preamplifier.
Fig. 5 -25C illustrates a white window on a gray background (analogous
to the lettering of Fig. 5 -24C) with severe negative streaking. The reason
for this may be made clear if we consider more than one line of information, as in Fig. 5 -25D. A severe loss of low- frequency response not only
affects the shape of the white signal ( ideally a flat- topped pulse) , but also
affects the base line as shown. Thus, the black streaking occurs all the way
across the raster, not just following the white signal.
Obviously, the best way to check the low- frequency response of the preamplifier is with a square-wave signal. Always be certain of the back -toback response of your square -wave generator and scope (as described
earlier in this chapter) before measurement is attempted. Also, since the
TELEVISION BROADCASTING CAMERA CHAINS
220
input amplitude of a preamplifier normally is very small, 75 -ohm pads
for the input will be needed.
Fig. 5 -26A illustrates a video pad suitable for test -signal hookups to
preamplifiers. With all the switches in the "up" position, the signal is
passed without attenuation. Any pad or combination of pads may be
selected to obtain a number of attenuation values from 3 to 53 dB. The
resistors are carbon film, 1 percent, V2 watt, and should be mounted in
a shielding box with a spacing of at least two inches from the shield to
avoid capacitive effects. The proper coaxial connectors, either BNC or UHF,
should be used for input and output.
Shield
3
20dB
10dB
dB
20dB
To
750
Input
From 750
of
Gen Output
Amplifier
26.42
-438.75-
I
I
371.25
106.73
371.25
-91.65-
-91.6
-144.38-
(A) 75 ohms,
3 to 53 dB.
Nearest EIA Value . 3900
Shield
From 754
Gen Output
To
375011
%I
750 Input
of
Amplifier
(B) 75 ohms, 40 dB.
Fig. 5 -26. Video attenuators.
Fig. 5 -26B illustrates a simple 40 -dB pad. The series resistor may be
either the 1- percent value of 3750 ohms or (when absolute gain measurements are not required) the nearest EIA value of 3900 ohms. In the
latter case, the attenuation is slightly more than 40 dB.
In running square -wave response checks, small -value cathode (or emitter) bypass capacitors, such as in Fig. 5 -18, must be bypassed with a large value capacitor as described previously.
It is extremely important that the square -wave input amplitude be sufficiently attenuated that no overload occurs. Overloading obviously would
flatten the waveshape and lead to an erroneous interpretation. The 40 -dB
pad is valuable for this purpose. For example, a 1 -volt signal into the pad
will give a 10- millivolt signal into the preamp. When observing the output of the preamplifier with the scope, always reduce the input level and
VIDEO PREAMPLIFIERS
221
check for any effect on the output waveform, so that compression can be
avoided.
Video -sweep and square -wave response checks on amplifiers employing
clamping circuits are covered in the next chapter.
EXERCISES
Q5 -1.
Q5 -2.
Q5 -3.
Q5 -4.
Q5 -5.
Q5 -6.
Q5 -7.
Why are phase controls necessary?
Does the output of a video preamplifier always feed a 50 -ohm (actually 51 -ohm) coaxial cable?
Do peaking circuits in the preamplifier compensate for scanning -beam
aperture distortion?
When the video preamplifier does not incorporate a gain control (usually the case), what adjustment is set to give the normal output level
of the preamplifier?
Why is it important to have a 40 -dB video pad available to the maintenance department?
What could cause white compression in the pickup tube even though
the associated amplifiers are linear?
What is the result of using excessive beam current in the pickup tube?
CHAPTER
6
Video Processing
In this chapter, we will consider the special kinds of video- signal
processing in amplifiers following the video preamplifier. This includes:
1.
Resolution and transient response (bandwidth)
2. Detail contrast (aperture compensation)
3. Cable- equalizing amplifiers
4. The dc component (clamping)
5. The transfer curve (gamma correction)
6. Blanking and sync insertion onto the video signal
7. Testing and maintenance
6 -1. THE TRUE
MEANING OF BANDWIDTH
The bandwidth of a video amplifier normally is defined in terms of the
frequency span between the -3 -dB amplitude points of the response curve.
However, there is an additional characteristic of prime importance to the
television engineer; this is the shape of the rolloff of the response curve for
a given gain- bandwidth product.
Fig. 6 -1 illustrates the point of interest at this time. The solid line indi-dB point.
cates a gradual rolloff starting at some frequency below the
The dash line indicates a rolloff starting at a frequency closer to the
-dB point (extended flat response) but with a much more rapid fall in
-3
-3
response. We will now examine the effects of this most important characteristic.
NOTE: It is important that the reader understand the following relationships. (See for example, Harold E. Ennes, Television Broadcasting:
Equipment, Systems, and Operating Fundamentals [Indianapolis: Howard
W. Sams & Co., Inc., 19711.)
1.
Horizontal resolution (ability to resolve sharp vertical transitions in the picture) is dependent on bandwidth, and 1 MHz corresponds to approximately 80 TV lines.
222
223
VIDEO PROCESSING
OdB
3dB
`\
Fig. 6 -1.
Rolloff of
frequency response.
1
I
I
c
2. Vertical resolution is fixed by the FCC standards and is
limited by
the number of active scanning lines, not bandwidth. Because of
the scanning factor, vertical resolution becomes about 320 lines,
somewhat less than the actual number of active scanning lines.
3. One cycle of video signal (whether pulse or sine wave) equals two
picture elements, one black and one white.
The High- Frequency Spectrum of Bandwidth
We know that high definition of sharp vertical transitions in the image
requires a well focused scanning beam and sufficient amplifier bandwidth.
A perfect reproduction would require a waveform with zero time for the
vertical transition; it would require an infinitely small scanning beam and
infinite amplifier bandwidth, both impossible in practice. This is to emphasize that high horizontal resolution requires a pulse concept of effective
bandwidth rather than a sine -wave concept. This applies just as well to
the low- frequency spectrum of bandwidth, which we will cover later.
Since the video signal must be considered a "pulse" signal, we know that
the pulse rise time versus amplifier bandwidth is important. Fourier
analysis tells us that for an amplifier to pass a pulse with the same rise
time and shape as the input pulse, the amplifier bandwidth must be:
BW =
1
2RT
where,
BW is the bandwidth,
RT is the rise time of the pulse.
This says that the bandwidth must be equal to half the reciprocal of the
rise time. For a pulse with RT = 0.02 µs:
BW=
1
1
2(0.02)
0.04
=25 MHz
(Since the rise time is in microseconds, the result is in megahertz.)
224
TELEVISION BROADCASTING CAMERA CHAINS
Since the limit of bandwidth normally is taken to be the -3 -dB point
(Fig. 6 -2A) , the frequency at this point is designated the cutoff frequency (fe) . Assuming the input pulse does not have a faster rise time
than the amplifier, phase shift is proportional to frequency ( Fig. 6 -2B) ,
with uniform time delay at all frequencies in the passband. Thus, the reproduced waveform (Fig. 6 -2C) is the same as that applied, with no overshoot or undershoot. Bear in mind that the curve in Fig. 6 -2A has a
gradual rolloff beyond the -3 -dB amplitude -response point. This is normally specified as a 6 -dB /octave rolloff.
For the wideband curve of Fig. 6 -2A, since:
BW =
2RT
it follows that:
RT
- 2BW
1
Thus, if the wideband response is 10 MHz, the rise time of the amplifier is:
RT = 20
= 0.05 µs
-3dB
á
fC
(B) Phase response for A.
(A) 6-dB/ octave rolloff.
-3dB
11
fC
(C) Step waveform.
(D) Fast rolloff.
(E) Phase response for D.
(F) Ringing of pulse.
a
Fig. 6 -2. High- frequency response.
225
VIDEO PROCESSING
If the applied pulse waveform (Fig. 6 -2C) has a rise time of no more
than 0.05 µs, the output will be free of overshoot or undershoot. If the
wideband response with the rolloff of Fig. 6 -2A is 25 MHz, the amplifier
will pass a pulse with a 0.02 /Ls rise time without distortion.
In practice, designers make a compromise in wideband amplifiers, and
the shape of the rolloff is made only sufficient to allow no more than a 2
to 3 percent overshoot for a pulse rise time equivalent to the amplifier
rise time. For the purpose of visualizing this relationship, a k factor is
defined as follows:
(BW) (RT)
=k
where,
BW is the bandwidth in megahertz (to the -3 -dB point) ,
RT is the rise time in microseconds (measured between 10 and 90 percent of peak value),
k is a factor between 0.3 and 0.5, depending on the type and amount
of high- frequency compensation.
The limit on factor k is that the overshoot on the leading edge of a pulse
must be less than 3 percent. In fact, a system has an equivalent bandwidth
and rise time only within the limits of 3- percent overshoot.
The most typical value for k is 0.35, and the equation may be expressed
in three possible ways:
(BW) (RT) =0.35
(BW)
0.35
(RT)
0.35
(RT) = (BW)
Table 6 -1 is based on this relationship and shows the equivalent TV lines
of horizontal resolution for different values of rise time and bandwidth.
Table 6 -1. Resolution and Amp ifier Response
Equivalent
BW (MHz)
2
3
4
5
6
7
8
9
10
RT
(µs)
0.35
0.175
0.1166
0.0875
0.07
0.058
0.05
0.0437
0.039
0.035
TV Lines
80
160
240
320
400
480
560
640
720
800
226
TELEVISION BROADCASTING CAMERA CHAINS
Notice that a pulse with 0.035 µs rise time will be passed with no more
than 3 percent overshoot by an amplifier that has a 10 -MHz bandwidth.
For example, an oscilloscope ideally must show all TV waveforms, including sync pulses, exactly as they are, without introducing any distortion
in the scope amplifier. The "wideband" position of the Tektronix 524
scope has a bandwidth to 10 MHz ( -3 -dB point) with a gaussian rolloff
curve. With this characteristic, the -3 -dB point occurs at 10 MHz, and
the -12 -dB point occurs at 20 MHz. This is a 9 -dB /octave rolloff. (When
a frequency is doubled, it has been increased by one octave.) This type of
rolloff is suitable for scope amplifiers used to display television waveforms.
More recent oscilloscopes, such as the Tektronix 545 (30 MHz) and 547
(50 MHz) , have much shorter rise times (11 nanoseconds and 7 nanoseconds, respectively) .
A gaussian response curve, while essential in oscilloscope vertical amplifiers, is not found in television camera chains or in video distribution amplifiers. This is because of the limitation of rise time in a series of amplifiers forming a cascaded system. The rise time of the original waveform is
reduced by the square root of the sum of the squares of the individual
amplifier rise times.
For example, suppose we pass a signal through two identical amplifiers
with 10 -MHz bandwidth and gaussian response. The combined rise time
of the amplifiers is:
RT = N/0.0352 + 0.0352
_ \/0.00245 = 0.05 µs
(approx)
This is a 40- percent increase in rise time as a result of passing the signal
through just two cascaded 10 -MHz gaussian amplifiers. In practice, many
video amplifiers are cascaded in forming a complete system.
Thus, the practical video amplifier must have a flat frequency response
up to and including the highest anticipated frequency, with a relatively
rapid rolloff beyond this frequency. It can be shown from pulse theory that
rise time is proportional to the area under the amplitude -frequency response curve; hence cascading such amplifiers does not appreciably affect
the rise time. However, such an amplifier will not reproduce a step transition without overshoot, ringing, or other transient distortions at the output.
It is actually the shape of the curve that is being changed when video peaking coils are adjusted. Leading and trailing transients of a rapid
transition in picture content must be adequately controlled. Hence it is
necessary for maintenance personnel to have complete familiarity with the
scope- amplifier and video -amplifier characteristics.
Fig. 6 -2D illustrates a frequency- response curve which more nearly approaches that of the average video amplifier. The resultant phase response
is shown in Fig. 6 -2E. Remember that a "pulse" (step transition) requires
transmission of the higher -order harmonics (which may actually be above
227
VIDEO PROCESSING
the passband intended) to be free of waveshape distortion. Fig. 6 -2F
shows the resultant ringing that occurs. The amplitude of such ringing depends on the step- transition rise time for a given amplifier bandwidth and
rolloff characteristic. The distribution of ringing ( leading and trailing) is
an indication of direction and degree of phase shift. Actually, Fig. 6 -2F
shows a "phase corrected" signal, as indicated by the even distribution of
ringing at the leading and trailing edges. Late arrival of high -frequency
components causes most of the ringing to occur on the trailing edge.
whereas early arrival of these components causes most of the ringing to
occur at the leading edge.
The Sine- Squared Pulse
Thus far, we have learned that the sine-wave response of a video ampli-
fier does not provide a complete story of the amplifier performance for a
video signal. Likewise, a step transition (or a square -wave signal) is not a
particularly useful signal for evaluation unless the exact rise time of the
pulse is correlated properly with the intended passband of the amplifier.
The very important transient response, which accounts for the degree of
picture ringing, smearing, or streaking, requires a rather precise analysis
method to assure valid tests in practice.
It is pertinent to recall here the actual pickup -tube output when a
"square- wave" test pattern is scanned. Remember that, because of the
round shape and finite size of the scanning beam, the actual output more
nearly approaches a sine wave (actually a sine-squared wave, which we will
develop) rather than a square wave. Therefore, the practical video amplifier is not required to handle square -wave video information (discounting
blanking and synchronizing pulses in a composite picture)
The problem of pulse rise times that are not related to the actual picture
transmission spectrum is the reason why, sooner or later, we will be concerned with the sine -squared (sin2) pulse. The basic usefulness of this
test signal is in complete system testing, and therefore a complete study
of the uses of the sin2 pulse is more appropriate for a text on system maintenance. However, we will go through an introductory examination here,
because the sin2 pulse provides an excellent means for revealing the practical requirements of a video amplifier, and because it has certain limited
applications in camera-chain maintenance, as covered later in this chapter.
First of all, be sure to understand what a "picture element" is. A picture
element is determined by the available bandwidth. Our complete TV
system is fixed by FCC standards that allow the video transmitter only a
4 -MHz bandwidth for the picture signal.
One cycle occurs in a time equal to the reciprocal of its frequency; for
example:
.
1
cycle at 4 MHz =
microsecond
4x110 = 0.25
TELEVISION BROADCASTING CAMERA CHAINS
228
Therefore, a black -to -white transition with a width representing 4 MHz
will occur in 0.25 microsecond. But black is one picture element, and white
is one picture element. Therefore, a picture element in a 4 -MHz system
represents 0.125 microsecond ( one alternation of the complete cycle) . In
the sin2 technique, a time duration equivalent to one picture element is
given the symbol T, whereas a time duration equivalent to two picture
elements ( for the system bandwidth under test) is symbolized by 2T.
An explanation of the sin2 pulse is shown in Fig. 6 -3. In Fig. 6 -3A,
note the conventional continuous sine wave with a frequency of 4 MHz
0.25 µs
i
\
x
0
\
Axis
(A) Sine wave.
0.25 ys
O
Axis
90o Phase
Shift
(B) 90° shift.
0.25 µs
Base of Pulse
0.125 ys
(C) Axis shifted.
1.0
T
Pulse (4 MHz)
E 0.5
(D) Spectrum.
2T
Pulse
2
4
8
Frequency lMHzl
Fig. 6 -3. Sine- squared pulse.
VIDEO PROCESSING
229
( one cycle occurs in 0.25 /Is). We realize from fundamental theory that
any phase shift of a continuous sine wave can be measured only by
laborious methods not suitable for routine testing of transmission facilities. Also, the amplitude- frequency characteristic of a system simply shows
the amplitude of the continuous sine wave relative to the amplitude at a
reference frequency, unless we are equipped to measure the phase relative
to a known reference.
Observe Fig. 6 -3B. If we shift the waveform 90 °, we have one complete
cycle of an inverted 4 -MHz cosine wave (starting and finishing at negative peaks). Now with an added dc component of such value as to raise
the negative peaks to the zero line, we have the T pulse for a 4 -MHz
system (Fig. 6 -3C) . Note that the half -amplitude duration (h.a.d.) is
0.125 µs, equivalent to one picture element for a 4 -MHz bandwidth. Fig.
6 -3D shows that the significant energy of the T pulse is down 50 percent
(6 dB) at 4 MHz, and there is practically no energy beyond 8 MHz. The
energy of the 2T pulse (h.a.d. of 0.25 µs) is 50 percent (6 dB) down at
2 MHz, with no significant energy beyond 4 MHz. Thus, the system
can be checked with a pulse that essentially duplicates actual picture conditions and that provides known frequency content upon which to base
judgement of system performance. Please note that any similarity to the
sine wave no longer exists; a pure sine wave has no harmonic content at all.
Fig. 6 -4A shows the preceding definition in terms of T and system
bandwidth. Fig. 6 -4B shows the terminology used with a pulse that has
been passed through an amplifier or (more usually) a complete system.
The first lobe is a negative overshoot, and the second lobe is a positive overshoot, either preceding or following the pulse.
The sin2 pulse generator normally also generates a window signal following the pulse (Fig. 6 -4C) so that an amplitude reference for low
frequencies is established. The sin2 pulse appears as a thin, vertical, white
line on the left of the raster immediately following blanking, and the
white window is on the right side, centered vertically on the raster as indicated by the field display. This signal is used frequently for system and
line checks. The window signal has a rise time equivalent to that of the
sin2 pulse.
The amplitude- frequency and amplitude -phase response at higher frequencies ( beyond about 100 kHz) will be most evident in the measurement of the sin2 pulse. Amplitude -phase response is most evident in the
measurement of the window signal.
NOTE: Some generators place the sin2 pulse following rather than preceding the window. This has no effect on a basic understanding of the
measurement principles.
Distortions at low frequencies produce waveform changes with a long
time constant as, for example, in streaking. This is most evident by window
measurement. Distortions at higher frequencies produce waveform changes
230
TELEVISION BROADCASTING CAMERA CHAINS
with shorter time constants as, for example, in smearing, loss of resolution, or "edge effects" from bad transient response. This is most evident
by sin2 pulse measurement.
High- frequency rolloff results in loss of amplitude. Loss of amplitude results in a widening of the pulse, since the area of the pulse represents a
h.a.d.
0.063
0.125
0.125
0.250
Microsecond
Microsecond
Microsecond
Microsecond
IT Pulse for 8-MHz System)
IT Pulse for 4 -MHz System)
)2T
(2T
Pulse for 8 -MHz System)
Pulse for 4-MHz System)
(B) Distortion lobes.
(A) T -pulse definitions.
Vertical -Rate
Horizontal -Rate CRO Display
o
c
íg
c
co
ó
ó
î
CRO Display
P
114
1/2
114
__.._
I._...
IRE Units
100
Window
Window
--!
l0
(Line)
L
0
L
Field)
s
sin2 Pulse
(C) Sine- window signal.
Fig. 6 -4. Television test signals.
constant dc component. A slow rolloff within the video band produces
large reduction in amplitude (and pulse -width increase) with little or
no ringing. A rapid rolloff close to the top of the band but still within the
desired video bandwidth produces both ringing and a reduction (perhaps
slight) in amplitude. A rapid rolloff (almost a cutoff) just above the
video bandwidth concerned results in practically no effect on amplitude,
but does produce ringing. The shape of the rolloff and whether the result-
231
VIDEO PROCESSING
ing phase shift is leading or lagging is revealed by the distribution of
ringing before and after the pulse.
The window detects low- frequency distortion which has practically no
effect on the sin2 pulse. The window shows undershoot, overshoot, and
horizontal tilt depending on the time constant of the impairment. The
window when used with the sin2 pulse has the same rise time as the
pulse so that no frequencies beyond the system test reference are introduced.
The Low -Frequency Spectrum of Bandwidth
See Fig. 6 -5A. The rise and decay times of a pulse depend on the system
high -frequency response and the shape of the passband response curve.
We are concerned now with the duration response (td) , which depends
on t /RC, or time divided by the RC product. This, of course, is the low frequency characteristic in practice.
The output voltage as a function of t /RC is shown in Fig. 6 -5B. As t
increases from 0, the factor t/RC increases and the output voltage decreases until, at t /RC = 1, the output voltage drops to 0.37 of the initial
voltage.
Since the pulse durations required in a TV system are known, it is most
convenient to use the reciprocal of the above relationship in thinking of
practical RC- coupled circuits. Fig. 6 -5C shows a plot of the output voltage
during a pulse in relation to the RC /td ratio. Note that it is necessary to
have an RC product of 10 times the pulse duration (td) to avoid more
than a 10- percent tilt over the duration of the pulse. It is obvious that the
time-constant problem becomes severe in a practical circuit when the
duration of a field is 16,666 µs (the reciprocal of 1/60 second) The time
constant (TC) is given in seconds when R is in ohms and C is in farads
or when R is in megohms and C is in microfarads. It is given in microseconds when R is in ohms and C is in microfarads or when R is in
megohms and C is in picofarads.
For example, the combination of a 0.1-SF coupling capacitor and a
1-megohm grid resistor results in a time constant of 100,000 µs. You can
see that this value is not 10 times the field duration. The TC value in
practical circuits is limited by the stability factor (motorboating, large
capacitances to ground, etc.) ; this is the reason why either negative feedback to flatten the lows (as well as the highs) is used, or the low -frequency boost circuit is employed (previous chapter) In amplifiers incorporating clamping circuits, the low- frequency characteristics are almost
entirely dependent on proper operation of the clamp -pulse former and
clamping circuit (discussed later)
See Fig. 6 -5D. This cathode-follower circuit has a sine -wave frequency
response that is only 3 dB down (relative to 1 MHz) at 5 Hz. But, for
all practical purposes, the time constant of this circuit is 1000 X 12, or
12,000 µs. This is an RC /td ratio of less than 1 for the field duration in a
television signal. The circuit in Fig. 6 -5D is that of an oscilloscope probe.
.
.
.
TELEVISION BROADCASTING CAMERA CHAINS
232
Low
Frequency
Input Pulse
High
Frequency
(A) Pulse nomenclature.
90%
Output Pulse
--- - - -1 -I
l
I
Duration
Rise Time
Decay Time
+0.8
1.0
+0.6
0.8
+0.4
á
+0.2
0.6
'
0
c't 0.4
-0.2
-0.4
0.2
0
0.2
0.6
0.4
0.8
t
1
0
-0.6
Time
(B) RC discharge curve.
-
(C) Square -wave tilt.
(D) Typical
Fig. 6 -5. Factors in low- frequency response.
stage.
VIDEO PROCESSING
233
Since capacitive loading of the circuit by the probe must be minimized,
a small coupling capacitor must be used. This probe permits making circuit checks at medium and high frequencies without a sacrifice in gain such
as occurs when a 10 -to -1 capacitance -divider probe is used. It is not intended to be used in those instances in which low- frequency duration
checks are important.
This example is intended to form a sharp demarcation line in your
mind between "sine-wave response" and "transient response." An amplifier that checks "flat" down to 5 Hz may produce excessive "tilt" in a low frequency step -response waveform.
6 -2. THE
TELEVISION- CAMERA RESOLUTION CHART
The resolution chart was designed to provide a standard reference for
measuring resolution of television cameras and as an aid in testing for
streaking, ringing, interlace, shading, scanning linearity, aspect ratio, and
gray -scale reproduction. It contains low- frequency, midfrequency, and high frequency information.
Resolution
The horizontal resolution that may be obtained from many camera chains
limited by the resolving capabilities of the camera tube, and not by the
bandwidth of the video amplifiers employed. Therefore, much useful information concerning the limiting resolution, percent response at various
line numbers, and degradation of resolution with aging of camera tubes
and components can be obtained from a test chart containing a high
number of lines. Fig. 6 -6 illustrates a typical resolution chart with high frequency resolution detail up to 1600 TV lines.
The types of resolution charts vary with the intended purpose. For example, the usual station -identification resolution chart often broadcast just
prior to sign -on has vertical- and horizontal -resolution wedges going
only to 320 lines. This is the limiting vertical resolution set by FCC
standards, and the limiting horizontal resolution (also from FCC standards)
corresponding to the bandwidth limitation of 4 MHz at the visual
transmitter.
The chart shown in Fig. 6 -6 is more useful at the studio, where the
bandwidth normally extends to 10 MHz, for the purpose of evaluating
and adjusting camera chains. The center horizontal and vertical wedges are
composed of four black lines separated by three white lines of equal
width. The numbers printed alongside the wedges correspond to the total
number of lines (black and white) of the indicated thickness that may
be placed adjacent to one another in the height of the chart. For example,
if black and white lines having the same thickness as those indicated at
the 300 position were placed adjacent to one another, a total of 300
(black and white) lines could be fitted into the height of the chart. Since
is
234
TELEVISION BROADCASTING CAMERA CHAINS
the aspect ratio of the chart is 4 to 3, a total of 300 X 4/3, or 400, lines
of this thickness could be placed in the width of the chart.
The fundamental frequency (based on FCC television standards) developed in scanning through the 300 position of one vertical wedge may
be calculated as follows:
Horizontal scannning frequency (nominal) = 15,750 Hz
H = time for active scan + horizontal blanking = 63.5 µs
Assume horizontal blanking = 0.17H
Therefore active scan = 0.83H
Active scanning time = 52.7 µs
The total number of vertical black and white lines, having the thickness
indicated at the 300 -line position on the chart, that could be placed adjacent to each other in the width of the chart is 400 (see preceding paragraph) Since a complete cycle includes one black line and one white line,
there should be 400/2, or 200, cyclic variations in scanning this pattern
(200 cycles in 52.7 µs). Thus, the time to scan (horizontally) one black
and one white line at the 300 -line position is 52.7/200, or 0.26 µs. The
fundamental video frequency is:
.
1
f
-t
=
1
0.26
3.8 MHz
Ea200
200
8
0
Courtesy TeleMation, Inc.
Fig. 6 -6. Typical resolution chart.
VIDEO PROCESSING
235
The limiting resolution as a result of the TV- transmitter bandwidth
is 320 TV lines (from Table 6 -1) Therefore, ideally, sufficient
aperture correction should be applied to the pickup -tube response to obtain 100 -percent response at 300 to 320 lines on the resolution pattern.
Circuitry and problems of aperture correction are covered later in this
chapter.
The fundamental video frequencies generated by scanning through different parts of the vertical wedges may be determined from the formula:
(4 MHz)
.
N =K
T
where,
N is the indicated line number on the chart,
f is the fundamental video frequency in MHz,
K is 80 (320 lines/4 MHz = 80),
When solved for frequency, the formula becomes:
f
=
80
Table 6 -1 lists the TV line number for whole multiples of
1
MHz.
Shading
Shading may be checked by visual inspection of the picture monitor to
determine if the background is an even gray, and if the same number of
steps is discernible on all four gray scales. Also, a waveform monitor may
be used to determine if the average picture -signal axis is parallel with the
black-level line at both line and field frequencies.
Streaking
Streaking of the horizontal black bars at the top or bottom of the large
circle is an indication of low- frequency phase shift or of poor dc restoration. The black bars are also useful in adjusting the high -peaking circuits
that are used in camera chains to compensate for the high - frequency roll off of the coupling network between the camera tube and first video
amplifier.
Interlace
The four diagonal black lines inside the square formed by the gray
scales may be used to check interlace. A jagged line indicates pairing of the
interlaced lines.
Gray Scale Reproduction
The transfer characteristic of the camera, for given operating conditions,
may be determined by using an oscilloscope with a line detector (discussed
later in this chapter) . The gray -scale reproduction achieved depends on
TELEVISION BROADCASTING CAMERA CHAINS
236
the amount of gamma correction employed, the manner in which the
camera tube is operated, and the adjustment of the picture monitor. The
user will have to standardize these operating conditions if comparative
subjective measurements are to be made. ( Gamma correction is covered
later in this chapter.)
The four ten -step gray scales cover a contrast range of approximately
30 to 1. The reflectance of step 1 is determined by the reflection density of
the chart material forming the center circle. The nine -step "paste -on"
gray scales cover a nominal contrast range of 20 to 1, step 2 having a
reflectance of 60 percent and step 10 a reflectance of 3 percent. The steps
are arranged in logarithmically decreasing values of reflectance such that
the difference in reflection density between adjacent steps is 0.16. Table 6 -2
gives the reflectance and reflection density of the steps on the gray scales.
The background reflectance of the outer useful area of the chart is 40
percent ± 5 percent.
Table 6 -2. Specifications for Gray Scales
Gray Scale
Number
Nominal Reflectance
Nominal Reflectance
Relative to MgO (Percent)
Density
1*
>60.0
<0.22
60.0
41.7
28.2
0.22
0.38
0.55
19.5
13.5
9.3
6.3
0.71
4.4
3.0
1.36
1.52
2
3
4
5
6
7
8
9
10
0.87
1.03
1.20
*Center circle
6 -3.
CLAMPING CIRCUITRY
This section assumes the reader has a background in the basic reasons
for and theory of clamping circuitry.' It remains to explore the more
advanced features of various types of dampers found in both tube -type and
solid-state equipment.
It is important to understand the difference between "clamping" and the
broader term "dc restoration." The clamper restores the dc component by
means of a keying signal such that a line -to-line reference is established
'See, for example, Harold E. Ennes, Television Broadcasting: Equipment,
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co.,
Inc., 1971).
237
VIDEO PROCESSING
at either sync -tip level or back-porch blanking level. Other, more simple
forms, such as unkeyed diodes, are grouped under the general classificacation "dc restoration."
There are two basic types of dampers: the "fast" clamp that practically
eliminates any 60 -Hz hum component, and the "slow" clamp that prevents
a change in baseline reference under varying picture levels (APL's) , but
is sufficiently slow to allow 60 -Hz hum components to pass. The latter
form is used in modern camera -waveform monitors to prevent the blanking level from shifting under varying APL's, but to allow a hum component
to pass with very little attenuation so that this trouble is apparent to the
operator. In this section, we will consider only the fast clamp as used in
camera chains. The slow clamp as used in camera-waveform monitors is
covered in Chapter 8.
Clamping is performed in any circuit in which a video reference black
must be established. Examples are blanking- or sync -insertion stages, gamma- correction circuits, sync -stretching circuits, etc.
Fig. 6 -7A shows the basic form of keyed clamping. The charge time of
coupling capacitor Cc is limited only by the value of source resistance
Rs. If a small -value coupling capacitor is used together with a low source
Clamped Element
(Control Grid of Tube
or Base of Transistor)
\
Clamped
Video Out
(A) Basic form of damper.
Electronic Switch at 15.75 kHz
Ground or
Video
I
DC Ref
nput
(Low R51
P1
C1
Clamped Signal
Output
15.75 -kHz
Keying Pulse
1JZr
Blanking Input
C2
(B) Tube -type clamping circuit.
Fig. 6-7. Clamping circuits.
DC
Ref
238
TELEVISION BROADCASTING CAMERA CHAINS
resistance ( such as from a cathode follower or emitter follower) , Cc is
charged rapidly when the electronic switch is closed. The switch is closed
only for about 1.5 µs either at sync -tip time or during the back -porch
blanking times. Discharge time is limited only by the impedance in the
switch.
Fig. 6 -7B illustrates a basic tube -type clamping arrangement. In this
example, blanking is inserted on the suppressor grid of the clamped tube.
A dc reference of picture black (or camera blanking) must be established
at the control grid for the start of composite blanking so that this reference
does not change with average picture level.
Negative pulses from the cathode circuit and positive pulses from the
plate circuit of the clamp-driver tube supply clamping pulses at line frequency for the four-diode clamper. Point A is the point to be clamped;
that is, point A must be maintained at a predetermined reference level on
successive pulses of the video signal. The diodes function as an electronic
switch triggered by the line- frequency pulses. When pulses are applied,
point B becomes positive and point C becomes negative, creating the condition for diode conduction to occur. This is equivalent to closing the
switch of the simple circuit in Fig. 6 -7A. When the diodes conduct, a low resistance charge (or discharge) path for capacitor Cc results. During the
active line -scanning interval, the diodes do not conduct. During this time,
should point A develop a negative dc potential, conduction will occur
through X1 to point B. If point A should develop a positive dc potential
during this time, conduction will occur through X2 to point C. In either
case, the clamped point is maintained at a certain reference level that determines the dc level of the video signal. Resistor R3 prevents point B
from drifting in the positive direction and point C from drifting in the
negative direction, which would render the clamp inoperative. Thus, the
clamped point (A) is maintained at reference level during the line -scanning interval, and it is discharged (or charged) at the conclusion of each
scanning line. By this means, interaction on the video signal information
is prevented, since the time constant (product of Cc and the forward
clamp resistance) is long compared with the line interval.
Potentiometer R2 and capacitor C3 are used for balancing the clamp
circuit. The need for this may be realized by noting that point A is at high
impedance, and therefore a small component of the pulse appearing at
point B will appear at point A through the cathode -to -anode capacitance
of X1. From the same principle, a pulse from point C will also appear at
point A. If these pulses are effectively balanced in phase, amplitude, and
rise time, their effect at point A is nullified, since they are equal and opposite in polarity. Without R2 and C3, however, a slight unbalance would
occur, since the cathode clamping pulse tends to rise more rapidly than
the plate clamping pulse. By adjusting R2 to balance the amplitudes, and
C3 to delay the cathode pulse slightly, an exact balance may be achieved
in practice.
239
VIDEO PROCESSING
The preferred pulse polarity for the input of the clamp- driver tube is
such that the tube conducts during active line -scan intervals (diodes held
open) and is cut off for the pulse duration. The resultant high source resistance for the pulses gives a more balanced drive condition for the
pulse- coupling capacitors, Cl and C2. Time constants R1C1 and R2C2
must be long compared to the pulse time. The clamping pulses are large
compared to the video signal amplitude so that the video signal does
not cause diode conduction during the active line interval. As a rule of
thumb, the video amplitude at the clamped point is one-third to onefourth the amplitude of the clamping pulses.
Fig. 6 -8A illustrates the usual transfer through an RC- coupled circuit
of video waveforms with different APL's. This illustration also indicates
exact proportionments of white to black durations for 50, 10, and 90 percent APL. Thus, if the active line -scan interval is 0.83H, a white pulse
of half this duration (0.415H) simulates an average picture level of 50
percent (average scene) . With the same active line -scan interval, a white
pulse of 0.083H simulates a 10 percent APL (dark picture) , and two
50%APL
-0.415H-
10% APL
90% APL
O.
0.83H
083H
r_
AC
Axis
0.83H
U
(A) Capacitively coupled.
AC Axis
_U
Sync Tip
(B) Dc rertored.
Fig. 6 -8. Effect of clamping on transferred waveform.
0.083H
240
TELEVISION BROADCASTING CAMERA CHAINS
white pulses with a black interval of 0.083H between them simulate a
90 percent APL (almost all -white picture) Note the marked difference in
.
sync levels.
Fig. 6 -8B shows the transfer of the same waveforms through a clamped
RC- coupled stage. Since the active line -scan signal always starts from
the reference dc potential at the beginning of each line, sync and blanking
levels remain the same regardless of APL; thus the video- signal ac axis
is effectively changed.
Fig. 6 -9 illustrates the pulse- transformer type of clamping circuit. The
base of clamp driver Q1 receives a sharp negative pulse to drive it from
cutoff to saturation. The sudden collector current through the primary
of the pulse transformer rings this circuit because of inductive kickback.
But after the first positive alternation, diode X5 clamps the negative portion of the ringing waveform by shorting out the primary when the
collector attempts to swing negative. Note that this polarity would be
reversed if an npn transistor were used. Note also the polarity of the resulting pulses on the secondary and how these pulses result in forward
biasing of the quad diode circuit, closing the "switch" and applying -12
volts to the Q2 clamped base. Coupling capacitor Cc is always small, since
it must be charged or discharged quickly (during the approximately 1.5 -.ts
duration of clamping) to the reference -12 volts. Time constant R1C1
Fig. 6 -9. Solid -state clamping circuit.
241
VIDEO PROCESSING
must be long compared to a line interval so that the charge on Cl will
hold the switch open (nonconducting) between pulses during the active
line (video) interval.
Back- Porch -Clamp Timing Circuitry
It has been mentioned that clamping may occur either at sync -tip level
or immediately following sync at the blanking level of the back porch. We
will now examine how the clamping pulses are delayed to the back -porch
interval.
Horiz Sync
Horiz Sync Differentiated
Fig. 6 -10. Back -porch timing
Clamp-Pulse Formation
waveforms.
Back -Porch Clamping
-Clamping
Reference Point
Horizontal -sync pulses normally are used to form clamping pulses for
either sync -tip or back -porch clamp timing. See Fig. 6 -10. Horizontal sync
( waveform 1) is differentiated ( waveform 2 ) . The following circuit is
sensitive only to a positive -going pulse so that a clamp pulse is formed at
the time of trailing edge of sync ( waveform 3) Waveform 4 shows the
resultant timing of the clamping action.
See Fig. 6 -11A. Since the ratio of RB to RL is only ten to one, we should
recognize the familiar "boxcar" circuit ( transistor saturated in quiescent
operating condition) The value of coupling capacitance C is small so that
the time constant results in differentiation of the input pulse.
At the base of the transistor, the waveform received through the coupling
capacitor would be a negative -going excursion followed by a positive going excursion if it were not for the base clamping action; all negativegoing excursions are clipped because of the low forward impedance of
the junction. Since the transistor is already saturated, negative -going excursions have no effect. At the end of the input pulse, the positive -going
excursion drives the transistor toward cutoff, and C begins to charge
through RB toward -10 volts. Thus, the output pulse occurs at the trailing
.
.
242
TELEVISION BROADCASTING CAMERA CHAINS
+10V
-10
1k
Small
1 Value
/ r'
PNP
Boxcar
Small
Value
=
Input
Input
Base
Base
Output
Output
--i r(A) Negative pulses.
Delay
NPN
Boxcar
Delay
- (B) Positive pulses.
Fig. 6 -1 1. Pulse -delay circuits.
edge of the input pulse. The width of the output pulse is approximately
0.7RBC. Obviously, if RE is made variable, the pulse width can be adjusted
as desired.
Thus, although the input sync pulse is about 5 µs in duration, the width
of the output pulse is dependent on time constant RBC. If C is 170 pF, the
output pulse width is:
(0.7) (0.01) (170) =
1.2
µs (approx)
Fig. 6 -11B illustrates the corresponding circuit for a positive -going input pulse. Since the npn transistor is already saturated, the positive -going
input excursion has no effect. The trailing -edge negative -going pulse (produced by differentiating action) cuts the transistor off, resulting in a positive -going pulse. As in the previous case, the width of the output pulse
is determined by the input time constant, which determines the discharge
time of the coupling capacitor.
We will go through the circuit of Fig. 6 -12 to show that it is possible to
analyze a chain of pulse circuitry rapidly by applying fundamental knowledge. This analysis will show how to determine what waveforms to expect
if the circuit is functioning as intended.
Since the circuit in Fig. 6 -12 is a back -porch clamper, we know that
clamping pulses must occur after the trailing edge of horizontal sync.
Comparison of the circuit of Q1 with Fig. 6 -11B reveals how the delay is
obtained.
Note that Q1 is actually a form of boxcar, since the emitter of the npn
transistor is returned directly to -20 volts and the base is grounded
through its resistor. The delayed negative -going base pulse drives Q1 to-
243
VIDEO PROCESSING
-12 V
-20 V
Amp
Delay Gen
QI
Hr
I
Z
Sync
70
60k
Clamp Driver
Q3j
Q2#.
". rr
12k
Video Input
co
1k
I
J¡C-
12 V
-zu V
D
1E-12
V
Fig. 6 -12. Fundamentals of
back -porch damper.
ward cutoff, but the Q1 collector is directly coupled to the Q2 base. The
positive-going pulse excursion at this point is "caught" at -12 volts by
the base-emitter clamp of Q2, and can go no further in the positive
direction.
We know that prior to the pulse, Q1 is saturated, and its collector is at
-20 volts (switch closed) . At this same time, since the Q2 emitter is at
12 volts and its base is at -20 volts, Q2 is cut off (the base is negative
relative to the emitter) . The positive-going pulse at the Q2 base then
drives this transistor on, and since there are no limiting diodes and no
emitter resistance for self-bias, Q2 saturates. So we should expect a Q2
collector pulse from 0 volts (cutoff) to -12 volts (saturation) .
The unbypassed emitter resistance of phase -splitter Q3 tells us that the
input impedance of Q3 is quite high (very low base current) , so we
should expect about the same pulse amplitude directly at the Q3 base as at
the collector of Q2.
Now analyze Q3, first the interval between pulses, then the pulse interval. Between pulses, the Q3 base is essentially at zero voltage, so the
transistor is cut off. During the pulse interval, since the Q3 emitter and
collector loads are identical, we expect essentially unity gain. Therefore,
we expect the Q3 emitter pulse to extend from about 0 (cutoff) to -12
volts (unity gain) . The emitter current at the -12 -volt peak is:
-
V
1E
12
R
;
300
= 40 mA
For a quick analysis, assume the collector current is equal to the emitter
current. Then the signal -voltage swing during the pulse is:
Vc
= (0.04) (300) = 12
volts
Therefore, the collector pulse swings 12 volts from -20 volts (cutoff) , or
up to -8 volts.
In practice, the collector current is slightly less than the emitter current
(Ic = ale) , and for this reason you will normally find the collector load
244
TELEVISION BROADCASTING CAMERA CHAINS
resistor of a clamp driver slightly higher in value than the emitter resistor.
For a 300 -ohm emitter resistance, the collector load usually is about 330
ohms. Thus, the slightly smaller collector-current swing develops a voltage
swing the same as that at the emitter.
The above analysis should serve to emphasize an important servicing
technique: Always use the dc- amplifier position when making scope
checks. This tells you considerably more about circuit operation than does
the ac- coupled position. For large- signal operation, this is very convenient.
In case of suspected clamping problems, always check the clamp pulses
themselves at the anode and cathode of the diode being driven by the
pulses. In some circuits, the pulses are particularly critical in rise and
fall times and the shape of the pulse tops. Any difference in rise and fall
times between the two opposite -polarity pulses sometimes can result in
the appearance of spikes at the clamped grid. A marked difference in pulsetop shapes will appear as a signal voltage at the clamped grid. In practice,
a significant difference in amplitude can exist between the pulses without
causing trouble. This is true up to the point at which the amplitude is not
sufficient to drive the diode into conduction.
Loss of clamping normally results in very long streaking of horizontal
lines in the image. This type of signal impairment is much more severe
than misadjustment of high -peaker stages in video preamplifiers.
6 -4. APERTURE -CORRECTION CIRCUITRY
"Phaseless" aperture correction normally is achieved by a delay -line technique in which one end of the line is terminated and the other end is effectively open. The "open" end actually is a high - impedance point in tube
or transistor circuitry. A solid -state aperture- correction circuit based on
this principle is shown in Fig. 6 -13. The problem with transistors is that,
because of the finite input impedance, an effective open circuit is hard to
obtain if the transistor is used as in a conventional tube circuit. Fig. 6 -13A
illustrates how this problem may be solved. The collector of Q1 provides a
sufficiently high impedance to serve as the open circuit for the delay line,
in this case at the sending end. The line is terminated in its characteristic
impedance at the collector of Q2.
The signal is divided into two parts: the delayed signal from the Q1
collector at the Q2 collector, and the undelayed signal from the Q1 emitter.
Since the collector signal is delayed, the undelayed signal of reverse polarity
at the output is, in effect, anticipatory. Now note that the undelayed component appearing at the collector of Q2 goes back down the delay line and
is reflected from the relatively high (unterminated) impedance at the Q1
collector. The reflected signal returns to the Q2 collector as a second component of reverse polarity. Thus, this single delay line supplies both an
anticipatory component and a following overshoot on transitions, resulting
in symmetrical aperture correction. Fig. 6 -13B shows how the aperture
245
VIDEO PROCESSING
correction is made to compensate the scanning-aperture rolloff. Fig. 6 -13C
shows the sine2 pulse response indicating phaseless aperture correction.
Recall that conventional peaking circuits introduce phase distortion, which
must be compensated by a phase-correction stage.
In Fig. 6 -13A, R3 is the correction amplitude adjustment. It sets the
ratio of main signal current to aperture- correction current.
(A) Circuit.
Correction
y,
i
Result
Scanning- Aperture
Rolloff
Input
Frequency
(B) Response.
Output
C) 0.1 µs .ring pulse.
Fig. 6 -13. Aperture correction.
The gain of this circuit at low frequencies is essentially unity. It is best
checked with a sine2 pulse; equal preshoots and overshoots indicate phaseless response. It also can be checked with the conventional video -sweep
technique. The frequency at which the boost is peaked varies between 2.4
MHz and 6 MHz, depending on the specifications of the delay line used.
Measuring Detail Contrast
The fact that one person can "see" 600 lines of horizontal resolution but
someone else can "see" only 500 lines on the same test -pattern reproduction
is not meaningful. There are too many variables involved, including eyesight, psychology, and the condition of the display device. There is only
one way to put resolution on an absolute and measurable basis: the use of
the line -selected sweep.
TELEVISION BROADCASTING CAMERA CHAINS
246
The conventional horizontal -rate scope sweep results in a pattern that
contains all the lines of the field. It is, however, possible to observe only a
single line of the field on, for example, the Tektronix 524 scope by turning
the TRIGGER SELECTOR switch to the delayed -sweep sector. By rotation of
the SWEEP DELAY control, any line or lines may be observed. This sweep
is obtained internally from the composite television signal by establishing
a coarse time delay from a vertical -sync pulse ( from the sync -separator
circuit) and then actually triggering the sweep from a selected horizontal sync pulse. Since the scanned line interval is 63.5 microseconds, if the time
base is adjusted to 6.35 microseconds per centimeter, a single line will
occur in the full -scale 10 centimeters of the scope graticule.
When the SWEEP DELAY control is used in this manner, the sweep is
triggered only 30 times per second. The resulting display is correspondingly
dim, and, when much ambient light exists, the screen should be viewed
through the hood provided for this purpose.
The particular line being observed on the CRO may be determined by
connecting a spare video monitor to the LINE INDICATING VIDEO output
jack at the rear of the scope. The picture on the monitor is brightened
during the time of the sweep gate. The SWEEP DELAY control is rotated
until the desired line of the picture is selected (brightened on the monitor) Thus, the amplitude of test -chart bandwidth wedges may be measured relative to gray (100- percent) areas.
Fig. 6 -14A shows the picture- monitor display of a multiburst test pattern
with the strobe line indicating the point of observation on the waveform
monitor. (The scope time base in this instance was adjusted to five TV
lines to make the indicating line apparent in the photo.) The highest burst
.
(A) Multiburst test pattern.
(B) Uncompensated waveform.
(C) Compensated waveform.
Fig. 6-14. Strobe -line technique of measuring detail contrast.
VIDEO PROCESSING
247
frequency in this particular test pattern is 8 MHz, which corresponds to
640 TV lines (Table 6 -1). Fig. 6 -14B shows the resulting waveform presentation with the aperture- correction circuitry switched off. Fig. 6 -14C
shows the waveform with the aperture- correction circuitry switched on;
note the increased response at mid and high frequencies.
The amount of aperture correction that can be employed is limited because high -frequency noise is increased along with high- frequency picturesignal content. In the image orthicon, the greatest amount of noise is toward
the black -level picture information. For example, note the increase in noise
at the black level of Fig. 6 -14C compared to that in Fig. 6 -14B. Where
amplitude controls are incorporated for the amount of aperture correction
inserted, the noise amplification becomes the limiting factor.
Amplitude- Limited Aperture Correction
Since the human eye is most sensitive to fine detail in picture high lights
and less sensitive to detail contrast in shadows, an amplitude -limited aperture correction sometimes is employed. This technique makes aperture correction of the signal effective only over the top 25 or 50 percent of the total
video- signal amplitude, thus effectively equalizing only the high lights of
the signal. This is sometimes termed a high -light equalizer.
Fig. 6 -15 is a simplified schematic diagram of such an aperture- compensation circuit. The video- signal amplifier feeds two separate circuits, a
white -clipper stage and one input of a difference amplifier. The white
clipper is clamped so that the video level does not vary through the RCcoupled stage with changes in APL. The clipping level for the diode is
adjusted by the clipper -level control. The output of this stage in turn feeds
two paths, a delay stage and the other input of the difference amplifier.
Thus, the difference amplifier has two inputs, the full video signal and
the high -light -clipped portion of the signal. Note that the full video signal
applied to the lower section of the difference amplifier goes through a
polarity reversal at the plate. The clipped signal is applied in effect to a
cathode follower; hence this signal does not go through a polarity reversal.
Therefore, at the output of the difference amplifier, the clipped signal
( which contains no high lights over the amplitude set by the clipping
level) is subtracted from the full signal. Thus the remainder is the same as
the high -light portion of the signal that was cut off in the clipper stage.
This high -light portion is aperture compensated and delivered to one input
of a summing amplifier.
Since the aperture equalizer presents a delay to the signal, an equal
amount of delay must be presented to the unequalized low -light portion of
the signal. This delayed low -light signal then becomes the second input to
the summing amplifier.
The summing amplifier has the same configuration as the difference amplifier. Note, however, that the two input signals are now of opposite
polarity; hence the output is the sum of the two inputs. The variable resis-
248
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 6 -15. Simplified schematic of amplitude -limited aperture equalizer.
249
VIDEO PROCESSING
tors between plate and cathode of the difference and summing amplifiers
allow exact equalization of gains between the two sections. The gain is
slightly less than unity.
If the clipping level is adjusted so that no clipping occurs, complete
cancellation results at the difference amplifier, and nothing is equalized.
At the other extreme, if the entire signal is clipped, no subtraction occurs
at the difference amplifier, and the entire signal is equalized. The clipping level control normally is adjusted so that about 50 percent of the total
signal amplitude is clipped at this input of the difference amplifier.
Contours out of Green
The contours- out -of -green technique is employed in some three- Plumbicon color cameras to provide "crispening" of the luminance signal similar to
that obtained from aperture compensation. Sharpened edges ( both vertical and horizontal) produced from the green channel of a three -tube
camera are fed into the red, green, and blue channels. This provides the
Green
Con our
Correction
Red
Blue
Blanking
Fig. 6 -16. Contours out of green.
same tolerance to misregistration that a fourth tube provides. Remember,
however, that this applies only to a monochrome receiver reproducing a
color program; misregistration will not result in a loss of sharpness on the
monochrome receiver, but it will cause color fringing, and therefore loss of
sharpness, on color receivers.
See Fig. 6 -16. Enhanced contours in the image from the green tube
correct all three channels and become the derived luminance signal. Since
the enhanced contours come from a single tube, these contours cannot be
degraded by registration errors. The delay of the green signal and the contour signal relative to uncompensated red and blue signals is corrected
automatically in the normal registration procedures.
Vertical Aperture Correction
The contour signal is enhanced by employing both vertical and horizontal aperture correction. We know that horizontal aperture correction is
250
TELEVISION BROADCASTING CAMERA CHAINS
limited by the bandwidth of the system, including the home receiver.
Vertical aperture correction (being a low- frequency correction) produces
a noticeable increase in sharpness not limited by bandwidth. For example,
enhancement of a vertical transition from black to white is obtained by
slightly darkening one line and whitening the succeeding line in the
transition.
Total Contour
Video
L
-1_
Horiz Correction Signal
Output
L.
I
Video Signal
1Ve t
Correction Signal
100 ns
I
100ns
t
100 ns
1
Video
Horiz Delay
Input
63.
3.sus
Horiz Delay
63.5 as
Fig. 6 -17. Vertical and horizontal aperture correction.
To accomplish this result some means of delaying a line of video information is necessary. Fig. 6 -17 shows that the video is delayed 1 line (1H)
in block 1, and another line in block 2. The signal that is delayed one line
is the main uncorrected signal. This signal goes to a subtractor, an adder,
and a 100 -nanosecond (one picture element) delay line. Thus, it serves as
the input to the horizontal aperture corrector. The same unit results in
both horizontal and vertical aperture correction.
The second line delay (block 2) is used to achieve the functions described as follows by the Philips laboratories:
For large areas in which each line is like every other, subtracting signals
from adjacent lines is like subtracting a signal from itself. To get more
equalization, more adjacent -line signal is subtracted, and less overall large area signal remains. This means changing equalization would also change
gain. To avoid this problem, a detail signal that has no large -area information is made. This is done by subtracting enough adjacent signal information to cancel the signal completely when adjacent lines are alike. The contour signal is a signal containing only vertical- and horizontal- transition
VIDEO PROCESSING
251
detail. This contour signal is then added in any desired amount to make the
main signal without changing gain.
6 -5. CABLE EQUALIZATION
See Fig. 6 -18. In the camera on the studio floor, amplifiers are made to
have a flat frequency response by means of peaking circuitry and negative
feedback-that is, by design. Because of the scanning- aperture effect,
aperture compensation is employed to maintain this "flat" amplifier
response.
The cable that connects the camera with the studio rack or console equipment may be of any length from 50 to 2000 feet. Since the coaxial element
within the camera cable exhibits a high -frequency loss approximately proportional to the square root of the frequency, an equalizer must be used to
maintain flat response. Since the equalizer attenuates the signal by the
amount of equalization required, it is generally followed by an amplifier
that compensates for this amplitude loss.
6 -6. GAMMA CORRECTION
In the ideal situation, the light output of the kinescope would be directly
proportional to the light input from the televised scene. But the kinescope
is nonlinear in the direction of compressing blacks and stretching whites.
The I.O. tube is nonlinear in the direction of stretching blacks and compressing whites. Also, the transfer characteristic of the I.O. is dependent in
a complex manner on whether the scene is high -key or low -key, which is
another way of saying that the dynamic transfer curve varies somewhat
with APL.
The resultant overall characteristic of an uncompensated system is black
compression, because the kinescope is more nonlinear than the I.O. tube.
The same is true for the vidicon. The Plumbicon has an almost linear
transfer characteristic, having a gamma of about 0.95, or close to unity.
Be sure of terminology. Amplitude linearity is a measure of the shape
of the transfer curve. It is a function of output luminance levels versus
input luminance levels of the system. Gamma is the exponent of the
transfer characteristic. This is the slope of the transfer characteristic plotted
on a log -log scale. A gamma of unity (dash lines in Fig. 6 -19) is a strictly
linear transfer slope. If the slope is greater than unity ( kinescope) , blacks
are compressed and whites are stretched. If the slope is less than unity
( pickup tube) , blacks are stretched and whites are compressed. Overall
system gamma is the product of the individual gammas.
For example, the vidicon has a relatively constant gamma of 0.65 over
the normal beam -current operating range. The average kinescope gamma is
around 2 (color standards assume an exponent of 2.2) , which means that
the picture -tube high -light brightness increases approximately as the square
252
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 6 -18. Function of frequency-compensation circuitry.
253
VIDEO PROCESSING
of the applied video voltage above cutoff. Then assuming all other parts
of the system have unity gamma, the overall system gamma is:
(0.65) (2.2) = 1.43
This value is greater than unity. The amount of gamma correction necessary
to obtain a unity exponent is:
1
1.43
= 0.7
The product of the system gamma (1.43) and the gamma correction (0.7 )
is 1, or unity.
(A) Pickup tube.
(B) Picture tube.
Fig. 6 -19. Transfer curves of terminal devices of TV system.
A transfer curve is graphically illustrated by the system response to a
stair -step signal. Such a curve might be strictly linear except at one extreme
end, either black or white. The term "gamma" in photography refers to
the maximum slope of a curve showing density versus exposure (exposure
on a logarithmic scale). The term "gamma" in television usage (and this
term is abused by many engineers) should refer to the overall system transfer characteristic on a log -log scale. Although an oversimplification, the
overall transfer characteristic is given an exponent that relates the relative
reproduced output luminance to input luminance. Gamma is not a constant
in a television system, but varies over the contrast range from low lights to
high lights.
Although there is considerable variation in gamma- correction circuits,
nearly all work on the principle of adjusting the threshold of diode conduction, as shown basically in Fig. 6 -20. Some gamma circuits work only
in the gray -to -black region; multiple diodes are used so that the transfer
curve can be made to "break" at different conduction levels. The circuit of
Fig. 6 -20 employs both black stretch and white stretch.
When the white - stretch control is rotated clockwise ( direction of arrow) ,
a point is reached at which X1 is forward biased. Since white is a positive going signal at the emitter, this voltage fixes the point of "bend" in the
white region. When X1 conducts, the emitter resistance is partially bypassed, reducing degeneration and increasing gain. See the white -stretch
curve in Fig. 6 -20B.
254
TELEVISION BROADCASTING CAMERA CHAINS
lk
Gamma -Corrected
Output
lk
White
White
Stretch
Black
Black
Stretch
(A) Circuit diagram.
Output Amplitude
(B) Transfer curves.
Fig. 6 -20. Basic circuit for black and white stretch.
Similarly, as the black- stretch control is adjusted clockwise, the back bias
on X2 is reduced and it conducts. Note that a black signal is a negative going signal at the emitter. Therefore, the gain increases in the gray -toblack region, stretching blacks. See the black -stretch curve in Fig. 6 -20B.
Checking Gamma Circuitry for Film Cameras
Regardless of the type of circuitry employed, the amplifier containing
gamma correction is most conveniently checked with the aid of a stair -step
generator. The following is a complete step -by -step procedure that can be
used regardless of the method in which gamma correction is obtained:
Observe the linearity- generator output directly with the scope; check
for proper linearity of the steps. If the steps from the generator are
not perfectly linear, take this condition into account when checking
the amplifiers.
2. Remove the camera -signal input from the control chassis, and feed
the stairsteps into this point. Increase the amplitude of the stair-step
signal until clipping starts to show at the output. This tells you the
maximum peak -to -peak signal the control ( processing) amplifiers
can handle without compression. Keep a record of this amplitude for
future reference. Be sure the gamma switch is on unity (no gamma
correction) Remove any aperture correction employed.
In a vidicon film- camera chain, fixed values of gamma correction
normally may be switched into or out of the circuit. Most units have
1.
.
255
VIDEO PROCESSING
switch with three positions: unity, 0.7, and 0.5. The 0.5 gamma
often necessary for films originally processed for theater projection
in order to fit the wide dynamic range of the film gray scale into the
range that can be transmitted without severe compression of low grays
and blacks. The vidicon has no knee and will not compress whites
(assuming that the beam current is sufficient to discharge the highest
high light)
3. For accuracy in checking linearity without danger of erroneous measurements as a result of excessive levels, reduce the input level to
one -half the value recorded in Step 2.
4. When observing linearity of steps at the output of the control chassis,
be sure the pedestal control is adjusted with sufficient setup so that
the bottom (black) step is not compressed. If the black steps are still
"pulled down," check any transient -suppressor controls for incorrect
adjustment. These circuits affect the black region.
5. Now place the black -stretch (gamma) switch on the position to be
checked. Fig. 6 -21 shows the values of output steps you should obtain
( with linear input) for the two most common values of correction.
As an example of the mathematical relationship involved, assume
gamma is 0.5, and figure where the step for a 0.1 -volt input step
should be:
a
is
.
(0.1)1 =v'0.1 =0.316
For the 0.2 -volt input step:
(0.2)1 =x/0.2 =0.447
To make computations for the 0.7 gamma correction, it is necessary
to use logarithms. For the 0.1 -volt step, it is necessary to find the value
of ( 0.1)17. The method is as follows:
log N = 0.7 log 0.1
_ (0.7) ( -1) = -0.7, or -1 +0.3
Since log N
= -1 + 0.3, N
N=
6.
is 0.1 times the antilogarithm of 0.3:
(0.1) (2.0) =0.2 (approx)
If the step -voltage response is distorted or more than a few percent
in error with the gamma correction in, the reference diodes ( when
used) are the most likely source of trouble. A slight departure from
theoretical step response is normal in circuits employing diodes that
are biased to conduct at various levels of video. This is because the
resulting response is in incremental steps from black to white rather
than a smooth curve.
NOTE: It was pointed out previously that the camera chain employs high
peakers and phase- correction circuits with small trimming capacitors across
a cathode (or emitter) resistance. These circuits may cause some tilt on
256
TELEVISION BROADCASTING CAMERA CHAINS
the stairsteps, but normally will not prevent an accurate measurement.
If excessive distortion is present, it is a simple matter to bypass such a
cathode correction circuit with a capacitance of about 0.47 ltF.
7.
Now run the gray-scale slide or test loop through the complete film
chain. This will give a good indication of the operation of the
camera head as it responds to the black -to -white pattern viewed by the
vidicon. If any compression (with gamma correction removed) is now
present, either the trouble is in the camera head itself, or the peak -topeak video level from the camera is excessive. Check this level at the
input to the camera control chassis. Always keep this input level
below that found in Step 2.
Always remember this: The black stretch of gamma -correction circuitry
applies to a linear input signal. This is the basic function of the amplifier.
Then, with the pickup tube looking at a logarithmic gray scale (standard)
the output should be linear.
,
Input
Output
Output
(Gamma - 0.7)
(Gamma - 0.5)
I.0
1.0
1.0
0.94
0.93
0.9
0.89
0.855
0.84
0.8
0.7
0.78
0.77
0.70
0.70
i'
'
0.615
0.6
0.632
i
,
i
0.5
0.4
i
0.547
Fig. 6 -21. Gamma correction.
0.526
i
0.447
0.43
i
i
'
0.324
0.316
0.3
i
0.2
i/
0.2
i
0.1
0
'
257
VIDEO PROCESSING
Checking Gamma Circuitry for Live Cameras
Gamma circuitry is best checked with the camera looking at the crossed
gray scale in the studio. The scope will reveal the range of control, which
should be checked against the specification sheet for the camera involved.
Obviously, if the gamma correction range of any one channel is low, color
balance with the other channels cannot be obtained. In an emergency, you
can operate without gamma correction in the channels required to obtain
balance.
A troublesome gamma stage is most often the result of faulty transistors
or diodes. It is best to make direct substitutions of parts as a check.
Be certain you can distinguish between a gamma fault and some other
fault. Fig. 6 -22A shows one possible CRO presentation when the crossed
gray scale is scanned at full raster. Note that the ascending luminance steps
are linear, showing that gamma is proper when a logarithmic chart is
observed.
Possible Beam-Alignment Error
Shading
(A) Full- raster display.
(B) Centered in raster.
Fig. 6 -22. Waveforms for crossed gray scale.
The mismatch at black level should be corrected with the horizontal
shading control. The white levels will still be unequal.
Be certain that the incident -light meter shows equal illumination at the
lower left and upper right white chips. Also, be sure the reflected light is
the same. Sometimes a white chip can be dulled by imperfections or dirt.
The next thing to check is beam alignment. The color camera normally
has a "G4- rock" circuit, which is simply a 30 -Hz multivibrator; its signal
is applied so as to result in a split field if the beam is not properly aligned.
Aim the camera at a registration chart, and adjust both the vertical- and
horizontal -alignment controls until the images on the viewfinder are superimposed.
You may find that, with the particular pickup tube involved, it is possible
to obtain good beam alignment around the central area but not at the
corners and edges. This is just one of the factors that can cause unequal
TELEVISION BROADCASTING CAMERA CHAINS
258
white levels at the two extremes of the pickup area. The presence of this
effect can be verified by zooming out or moving away from the chip chart
so that it is in the center of the raster as in Fig. 6 -22B. (The chart should
be displayed against a neutral background.) If the levels become even, it is
likely that a beam- alignment problem (or possibly black -level shading)
is present.
IMPORTANT NOTE: Always check the manufacturer's specifications for
the specific pickup tube involved. Some tubes have a photocathode or
target sensitivity specification that can vary as much as 20 percent between
left and right areas and still meet standards.
6 -7.
BLANKING AND SYNC INSERTION
Fig. 6 -23 illustrates one commercial method of blanking insertion: suppressor -grid injection of the blanking pedestals. It should be noted that
negative -black video is combined with negative pedestal voltages. The
polarity is important since the video-signal maximum black must occur
just under the blanking level. The composite blanking signals from the
sync generator are fed to a blanking amplifier, and then to the suppressor
grid of the mixing stage. The large negative pulses at this grid result in
plate -current cutoff during each pulse. The video signal and the resulting
pedestal formed by plate -current cutoff appear as positive excursions (positive black) in the plate circuit, and are impressed on the 1N34 clipper
stage. The clipping level (hence pedestal level) of this rectifier is determined by the setting of the pedestal-height control, since any signal amplitude above the dc voltage on the arm of this control is not passed by the
clipper diode.
The purpose of clamping at this point may now be more clearly understood. It is recalled that clamping occurs during the blanking intervals of
the pickup tube. The grid of the blanking -insertion amplifier is the clampCamera Blanking Intervals
BlankingInsertion
Amplifier
Blanking
Clipper
Pedestal Height
To Video Amp
íM
Negative -Black Video
+120 V
1N34
Pedestal -Height
\
Control
Clipping Level
for Pedestal Height
Negative Blanking Pulses
Positive Video and Blanking
to Clipper Stage
Fig. 6 -23. Basic technique for blanking insertion.
259
VIDEO PROCESSING
ing point, being held to a predetermined grid-voltage level. Therefore, the
pedestal control determines the clipping level at a voltage that is fixed in
reference to the blanking level of the camera. Thus, any necessary change
in the video gain control of the first stage usually does not necessitate
another adjustment of the pedestal control, since the ratio between black
signal level and pedestal height is not affected by the gain control. In this
way, the pedestal is fixed with reference to black in the signal rather than
with reference to the average signal content.
The synchronizing signal (composite sync from the sync generator) is
inserted and controlled by the same technique.
Fig. 6 -24 illustrates a common method of blanking insertion in Marconi
cameras. The first stage provides a clamp that is dependent on a variable
dc potential from the lift control ( "lift" is the British terminology for
blanking or black-level control) The signal is passed to a three -diode
gate, where system blanking is inserted and white clipping occurs.
The gate circuit is followed by a black clipper with a fixed clipping
reference. It is interesting to note that in this arrangement the level between the settings of the white clipper and that of the black clipper remains constant regardless of the setting of the black control, and the
stability of the two clipping levels is independent of the amplitude of the
blanking signal inserted. With the signal at a level of 4 volts, nonlinearity
introduced by the diodes in the clippers is kept to less than 3 percent of
picture amplitude.
Diode X6 is added to keep the number of forward -biased diodes between
point A and the black reference (point B) the same as the number between point A and the lift potential ( point C) . This together with the
use of low- temperature -coefficient components in the derivation of the
potentials at points B and C, provides stability in the black level-about
a 0.5 percent change for a 20 °C change of temperature. Slight transient
.
- Blanking
Input
Output
Ll
j-
X2
X6'
X1
X5
X3
Lift
I
-
Differentiated
Video
N
V
White -Clipping Ref
Courtesy Ampex International
Fig. 6 -24. Blanking insertion in Marconi camera.
260
TELEVISION BROADCASTING CAMERA CHAINS
breakthrough at black, caused by the finite switching time of the diodes, is
largely offset by adding a small amount of inverted, differentiated signal
onto the black reference.
6 -8. TESTING AND MAINTENANCE
A detailed description of normal setup and electronic adjustment of the
camera chain was presented in a previous volume2 and will not be repeated
here. It remains to examine the special techniques required in testing and
maintenance of video processing amplifiers. (The testing of gamma- correction circuitry has been covered already in preceding Section 6 -6.)
A typical video path for a camera, or for one channel of a color camera,
is shown in Fig. 6 -25. The sequence of functions is different for every
manufacturer, but the basic function of each block is the same regardless
of the sequence. These functions can be outlined briefly as follows:
Preamplifier. This stage normally is used only for vidicons or Plumbicons. Its purpose is to convert the video signal current from the pickup
tube to an amplitude equal to that from an I.O., or sufficient for satisfactory
processing of the signal.
Video Amplifier. Normally this is the first amplifier for an I.O. tube.
In the example of Fig. 6 -25, the stage includes aperture correction and remotely controlled gain. In a four -channel camera employing an I.O. and
three vidicons, aperture correction sometimes is included only in the
monochrome channel, since the three chroma channels are narrow -band.
Processing (Prot) Amplifier. This amplifier involves voltage amplification to boost the signal voltage to a level suitable for gamma correction.
It includes manual level sets for black and peak white.
Cable Compensation. This section corrects for camera -cable losses at all
frequencies across the video passband, for cable lengths of 200 to 1000 or
2000 feet.
Output Amplifier. This amplifier provides multiple 75 -ohm outputs for
distribution of the video signal to the system, and a 50 -ohm output to feed
back to the view finder through the camera cable. It also contains circuitry
to limit peak black and peak white signal excursions to preselected values
under overload conditions. There is provision for insertion of system blanking, and sync insertion is optional at this point.
Buffer Amplifier. This element of the system provides isolation for signal
feed to the view finder. It includes cable compensation for the view -finder
coaxial line.
View - Finder Amplifier. This section amplifies the video signal to a level
suitable to drive the display kinescope. Also, it normally employs a switch able "crispener" circuit as a focusing aid for the cameraman.
2Harold E. Ennes, Television Broadcasting: Equipment, Systems, and
Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971).
261
VIDEO PROCESSING
Aperture
Gain
Preamplifier
-
Video
Amplifier
Black Level
--
Camera -Head
Video Output
Processing
Amplifier
Pickup Tube
Gamma
Remote
Gain
White
Level
-
Test Pulses
Blanking
(Generated in
Camera Head)
Cable Length
1
Output
t
Level
7552
Output
Outputs
o op) o
Station Blanking
Cable
Compensation
Encoder
Amplifier
Rack Input
Cr spener
White Clip
Pedestal
Test From
Processing Amp
Out
o
In
Camera Cable
5012
Output
Buffer Amplifier
and Cable
Compensation
Kinescope.---View inder
Amp
ifier
lin Cam ra
Head)
Fig. 6 -25. Typical video path for camera channel.
Video Amplifiers
Three typical video-amplifier circuits common to almost all modern
cameras are shown in Fig. 6 -26. Fig. 6 -26A illustrates the well-known
method of base biasing in conjunction with emitter -resistance stabilization.
This is a popular voltage amplifier when the stage is operated between a
low -impedance source and a high- impedance load. The voltage gain in this
application normally is limited to less than 20 dB. Resistors R1 and R2
provide the base bias voltage and, in conjunction with RE, stabilize the
transistor operating point. Capacitor CE (when used) is small to compensate for the limited bandwidth. When amplifiers of this type are cascaded, the RECE time constants normally are staggered to obtain the desired
response. (This arrangement is not often used.)
Feedback pairs (Figs. 6 -26B and 6 -26C) provide improved frequency
response, linearity, and ac and dc stability. This circuit eliminates the need
for peaking circuits. With the addition of positive feedback (Fig. 6 -26D) ,
voltage amplification is increased as a result of feedback loss compensation.
262
TELEVISION BROADCASTING CAMERA CHAINS
Let us take each circuit in typical applications in a color camera, and do
that you know what to expect in servicing and trouble-
a quick analysis so
shooting.
First consider the single emitter-compensated stage (Fig. 6 -26A) . Assume an input stage is being fed from a 75 -ohm line. The line is terminated at the input, so you would expect a bridging impedance to 75 ohms.
(A) Emitter -compensated.
(B) Current feedback pair.
+10v
+10V
Feedback
(Bootstrap)
Capacitor
Source
50051
r- -i
QI
50Sí
2500
R1
2500
6200
R1
6200
10 V
(C) Typical feedback pair.
1
R2250
r
10 V
(D) Bootstrapped feedback pair.
Fig. 6 -26. Typical video -amplifier circuits.
Using the values of Fig. 6 -27A, perform the dc analysis first. The R1 -R2
voltage divider totals 50k, so the current through the divider is 20/50k =
0.4 mA. This 0.4 mA through the 5k resistor gives +2 volts at the base.
Now redraw the circuit as in Fig. 6 -27B so that RB is the equivalent resistance of R1 and R2 in parallel, and indicate the base voltage of +2 volts
(from base to ground) .
If the transistor is germanium, a difference of about 0.2 volt exists between base and emitter (if it is silicon, a difference of 0.6 to 0.7 volt
exists) . So indicate + 1.8 volts at the emitter as shown.
263
VIDEO PROCESSING
The dynamic small-signal emitter resistance (re) is approximately
26/IE, where IE is the emitter current in milliamperes. The dynamic base emitter resistance (REB) can be assumed to be 4 ohms for all practical
purposes in quick analysis. The transresistance (rtr) is the sum of these
resistances and any unbypassed external emitter resistance (RE)
:
rtr =re
+ REB +RE
When RE is rather high relative to ree and REB, these quantities can be
ignored for quick analysis. So in this example, rtr = 1000 ohms, the value
of resistor RE.
Since the emitter is at 1.8 volts, the emitter current is 1.8/1000 ampere,
or 1.8 mA (Fig. 6 -27C) . Ignore the transistor alpha ( which is about 0.98
or 0.99) , and assume the collector current (Ir) is the same as the emitter
current. The voltage across the 5k collector load is, therefore, 0.0018 X
5000, or 9, volts, so you should measure 20 9, or +11, volts at the
collector.
-
+2o v
+20
V
(B) Equivalent circuit.
(A) Circuit diagram.
+20 V
5
1
Times
Input
+2 V
Input at Base
Output at Collector
(C) Emitter current.
(D) Signal waveforms.
Fig. 6 -27. Analysis of amplifier stage.
264
TELEVISION BROADCASTING CAMERA CHAINS
The voltage gain is equal to RL/rtr. In this case, the gain is 5000/1000,
5. Thus, the signal at the collector will have a peak -to -peak value about
5 times that of the signal at the base. The ac axis of the output waveform
will be at approximately +11 volts (Fig. 6 -27D)
It has been assumed that the single stage is feeding a high impedance.
In practice, the collector signal load is lowered somewhat by the parallel
input impedance of the next stage. Therefore, the gain of the stage under
consideration is somewhat less than 5.
When compensating capacitor CE is used (Fig. 6 -26A) , the value to
provide a nominally flat response curve is:
or
.
1
Cr'- 27rfRE
where f is the 3 -dB frequency for the uncompensated stage. Thus, in the
example of Fig. 6 -27A, if the transistors used and the circuit impedances
are such that the response at the top of the intended passband (assume
8 MHz) is down 3 dB, then:
CF
(6.28) (8
1
X
106) (1000)
- 20 pF
(
approx )
Next, we will examine the feedback pair with emitter - follower output.
Because of the effects of transistor and circuit capacitances, reducing the
input and output impedance improves the high- frequency response, but at
the expense of gain. When low impedances are obtained by feedback,
inverse signal feedback provides another means to improve frequency
response and signal linearity. The basic circuit of Fig. 6 -26B shows the
feedback resistor (R1). Note that since the output impedance is coupled
back to the input, the output and input impedances of the circuit the highly
interdependent.
The voltage -gain (Ay) relationship in a feedback pair of this type is:
A,.
Rf
= Rs
where,
Rf is the feedback resistance,
R. is the source (previous stage, line, or generator) resistance.
This type of circuit is common in modern color cameras, particularly as
line drivers for 50- and 75 -ohm lines. Fig.6 -26C shows a typical circuit.
For the analysis, assume the npn transistors are silicon. Since the emitter
of Q1 is grounded, the base of Q1 can be assumed to be at +0.7 volt with
respect to ground. The difference between the -10 -volt supply and the
+0.7 -volt junction is +10.7 volts. So the current in R1 is 10.7/6200
ampere, or 1.72 mA.
VIDEO PROCESSING
265
The same 1.72 mA is present in the emitter return of Q2 and in the
2500 -ohm feedback resistor, Rf. This results in a voltage rise in Rf of 1.72
mA X 2500 ohms, or 4.3 volts. Adding this to the starting -point voltage of
0.7 volt gives a Q2 emitter voltage of +0.7 +4.3, or +5 volts.
Since the Q2 emitter voltage is +5 volts, there should be about +5.7
volts at the Q2 base. This is the same as the Q1 collector voltage. With this
information, the voltage across Re and then the current through Re
( 5 mA) can be calculated. You now know that the Q1 collector current is
about 5 mA, and the Q2 current is about 1.72 mA. (This is incidental
information, however, and not pertinent to troubleshooting techniques.)
Assuming the sending ( source) impedance is 500 ohms, the expected
voltage amplification is 2500/500, or 5. The output signal current is superimposed on the 1.72 -mA quiescent operating current of Q2, so the maximum peak -to -peak signal swing without clipping is 2 X 1.72, or 3.44 mA.
If RL is 500 ohms, this current swing results in a signal voltage of 0.00344
X 500, or 1.72 volts peak to peak. If Rt, is 50 ohms, the peak -to -peak signal
swing is 0.00344 x 50, or 0.172 volt. This is the maximum capability of
the circuit in Fig. 6 -26C.
Fig. 6 -26D shows the previous circuit with minor modifications that
drastically affect the operation. The Q1 collector load has been split into
two equal resistors that total the original 860 ohms. This provides a tap for
the bootstrap capacitor, which provides positive feedback to overcome the
signal loss resulting from negative feedback through Rf. Since there is now
a greater signal-current swing because of the bootstrapping, an additional
current must be supplied to the Q2 emitter. This is done through R2, a
250 -ohm resistor. The difference between the -10 -volt supply voltage and
the +5 volts at the Q2 emitter is 15 volts. Therefore, an additional 15/250
ampere, or 60 mA, is added to the 1.72 mA already in Q2 for a total of
61.72 mA quiescent operating current. The output signal current can now
swing -*61.72 mA for a peak -to -peak swing of approximately 120 mA
without clipping. This develops a signal of 6 volts peak to peak across a
50 -ohm load, or a voltage -gain capability over 30 times that of the circuit
in Fig. 6 -26C. Note that there is no difference between the voltage readings
to ground at the transistor junctions for quiescent operating conditions in
the two circuits.
There are two fundamental tests you can make in troubleshooting these
circuits; both can be done quickly with the scope. One example will suffice;
assume you are scoping the Q2 emitter. First, you know the dc operating
point should be about +5 volts. Calibrate the scope for a sensitivity of 1
volt /cm and center the trace at the bottom graticule line with the probe
touching a grounded point and the scope set on dc input. Then apply the
probe to the Q2 emitter and observe whether the trace moves upward 5 cm.
In the case of the circuit of Fig. 6 -26C, the signal superimposed on the
scope trace will be very small, so to check the signal swing go to ac operation and increase the scope gain to observe the signal. You now have the
TELEVISION BROADCASTING CAMERA CHAINS
266
complete functional story of conditions at the emitter of Q2. In the case
of the circuit of Fig. 6 -26D, you probably could observe the signal swing
without changing the scope settings. This swing should have its ac axis at
the +5 -volt level.
Fig. 6 -28A shows a typical gain -controlled video stage employing a
feedback pair similar to that of Fig. 6 -26C. The difference is that the feedback resistor (Rf) is shunted by a network that includes photoconductive
resistance devices R1 and R2. The filament brightness of the small lamps
determines the resistance of their associated cadmium -sulphide cells, and
therefore the total value of feedback impedance. Voltage is supplied to the
lamps from the master white -level control at the remote -control panel.
If this control should fail, you can, in an emergency, substitute a fixed
resistor of about 1600 ohms for the defective cell. You can shunt the cell
+10 V
Video
Input
120 of
Video Output
-10 V
1502 Coax)
White-LevelControl DC
(A) Photoconductive control element.
Video
Input
Gain
On Remote Panel
Video Output
(B) Variable emitter -current source.
Fig. 6 -28. Typical remote video -gain controls.
VIDEO PROCESSING
267
temporarily with the resistor for a quick check to determine whether this
is the source of the trouble. If you do not have a direct replacement, it
may be necessary to experiment with the value of the substitute resistor to
obtain proper control of the output level without overloading a stage prior
to the final output -level control in the control room. The temporary resistor
should have about one-half the value of the feedback resistor used.
All plug -in modules are susceptible to plug and receptacle contact problems. Intermittent deflection or fluctuations in video black or white level
are often the result of dirty or otherwise faulty contacts. If the equipment
manufacturer recommends a particular cleaner for these contacts, by all
means use it on a regular basis. Otherwise, a good tuner cleaner ordinarily
is satisfactory.
Another popular type of remote video -gain control is shown in Fig.
6 -28B. This simple and stable circuit depends on the principle that the
small -signal emitter resistance (re) is approximately equal to 26 /IE, where
I,; is the emitter current in milliamperes. By control of the current in the
constant -current supply (Q1) the currents in Q2 and Q3 are controlled;
hence, the value of re in these transistors (and therefore the stage gain)
is varied.
Video Sweeping the Processing Amplifier
Modern solid -state processing amplifiers seldom need the application of
video sweep testing except in cases of replacement of critical components.
Tube-type processing amplifiers require more frequent sweep alignment.
The basic video -sweep alignment technique was covered for preamplifiers
in the previous chapter. The same technique holds for processing amplifiers
except for special cases, which we will examine now.
Some amplifier in the system employs circuits in which blanking and sync
signals are injected and in which clamping is employed. These stages require a special testing procedure. Fig. 6 -29 presents a simplified diagram
of a sync- insertion circuit feeding a cathode -follower output stage that has
a clamped grid. This arrangement might be in a mixer -amplifier unit following the switcher stage or a line-output amplifier in which sync insertion
takes place. Point 1 in the clamper stage is considered first. From the
inherent nature of clamping tubes, considerable capacitance is added to
the circuit at this point. Therefore, these tubes cannot be removed without
upsetting the circuit constants, which would seriously affect the operation
of the output stage. Neither can the tubes be left in unless a keyed test
signal is employed as described later. This is true because the resultant
clamping pulses would give spurious response in the output, since the
unkeyed sweep generator contains no blanking pulses and the clamper
normally operates on these pulses. For this reason, it is necessary to replace
the clamper with a tube of the same type, but with the heater circuit opened
by cutting off the heater pins. These °dummy" tubes should be plainly
marked in some fashion (such as with paint or fingernail polish) so that
TELEVISION BROADCASTING CAMERA CHAINS
268
they will not be left in place inadvertantly after testing. When the clamp
is immobilized in this way, the grid of the output stage is left "floating."
It is then necessary to insert temporarily a grid resistor at point 2 in the
circuit (Fig. 6 -29)
The sweep generator may be connected at point 4 for the purpose of
checking this stage and aligning shunt -peaking coil L1. In this stage, another condition also must be considered. Sync pulses usually are inserted
(as shown) by an amplifier that shares a common plate load with the
.
Output Voltage
Video Amp
Clamper
6AG7
6AK5
Video
Input
Detector
To
Sweep
Generator
Scope
Sync
Amp
R
6AK5
Sync
Input
Keying Pulses For Clamper
470k for Tubes Such as
Remove in Previous Stage
LI
6AG7, 6ÁC7, 6AK5
Fig. 6-29. Typical sync- injection and clamping circuit.
video amplifier. Aside from the capacitive effects of the sync amplifier on
the video amplifier, the video amplifier steady -state plate voltage is dependent on the load current drawn by the sync -amplifier stage. Sync pulses
must not be injected into the video amplifier, however, since the resultant
patterns would be meaningless for the purpose of this test. If normal circuit
conditions are to be maintained, the sync -amplifier tube obviously cannot
be removed. If the amplifier has a plug -in connection for the composite
sync signal, this may be removed during the test. If the amplifier is rack mounted and receives sync from a distribution bus common to a number
of amplifiers, the signal should be "killed" in the stage prior to point 3 by
use of a dummy tube. Should point 3 obtain its drive directly from a syncdistribution bus, it is necessary to break this connection temporarily.
Often, it is desirable to employ "keyed" test signals phased by the station
sync generator to eliminate the test amplitude during horizontal- and
vertical -blanking intervals. This permits checking the many types of amplifiers incorporating line -to-line clamps, which otherwise need to be
modified if straight test signals are used. Although commercial equipment
is available for keyed sine waves, video sweep, stair -step signals, etc., there
is an apparent scarcity of available units that process a square -wave signal
properly.
269
VIDEO PROCESSING
-9 V at Approx 24 mA
Blanking
R2
33k
Video
Sweep
Input
100
if
12V
R1
2N1143
75
Sync
R3
10k
t-t
Keyed Signal Output
1750 Load)
Fig. 6 -30. Signal keyer.
Multiburst and window test -signal generators that provide for insertion
of an external signal make it possible to insert any test signal over blanking
and/or sync signals. For those stations that do not have this facility, the
following is useful.
Fig. 6 -30 shows a simple transistor circuit devised for this purpose. The
2N1143 transistor is reasonably priced and is effective for video use. Intro-
(A) 60 -Hz square wave.
(C) Sync added to
B.
(B)
1
-MHz sine wave.
(D) Video sweep signal.
(E) Detected video sweep.
Fig. 6 -31. Waveforms for keyed test signals.
270
TELEVISION BROADCASTING CAMERA CHAINS
duction of sync and/or blanking pulses of negative polarity drives the
transistor to cutoff for the input signal, and the amplitude of the pulse as
adjusted by R6 and R7 appears across the output load ( input of system to
be checked).
Fig. 6 -31A illustrates the keyed output when the test signal is a 60 -Hz
square wave. The setup (blanking) level is adjustable by means of R6 to
the desired amount of pedestal. This type of signal results in a clean composite blanking interval, and with the addition of sync no modification is
necessary for units employing clamps.
The keyer also may be used for sine waves, as shown by Fig. 6 -31B with
only blanking inserted. Fig. 6 -31C illustrates the waveform after sync is
inserted. This unit also enables the engineer to feed keyed video sweep to
the system, with the same advantage of being able to leave all clamping
circuits in an active condition, just as for any composite picture signal.
(See Figs. 6 -31D and 6-31E.)
EXERCISES
Q6 -1.
Q6 -2.
Q6 -3.
Q6 -4.
Q6 -5.
Q6 -6.
Q6 -7.
Q6 -8.
Q6 -9.
Q6 -10.
Q6 -11.
Q6 -12.
Q6 -13.
Q6 -14.
What is meant by the term "phase shift" in a video amplifier?
What is primarily responsible for phase shift at low frequencies in
video amplifiers?
Name five factors that would account for low- frequency losses in a
video amplifier.
What detrimental effects can result from trying to employ a coupling
capacitor of too large a value between video amplifier stages?
Why should you carefully check the level of a 60 -Hz square -wave
input to a unit or system being tested?
What is the fundamental frequency of the 2T pulse for a 4 -MHz
system?
What is the repetition rate of the sine pulse?
In a four-tube camera, is cable compensation used in all four channels?
In a four -tube camera, is aperture correction used in all four channels?
In the circuit of Fig. 6 -27A, assume that the collector load is 3.3k,
the emitter resistance is 500 ohms, the transistor is germanium, and
other values are as shown. What dc voltages would you expect to find
at the emitter and the collector? What voltage gain would you expect?
In the circuit of Fig. 6 -27A, suppose you measure a peak-to -peak
input signal of 0.2 volt. What peak -to -peak signal would you expect to
measure at the collector?
In the circuit of Fig. 6 -26C, assume that R1 is 10k, Rt is 3k, and
other values are as shown. What dc voltages (with respect to ground)
would you expect to measure at the base and collector of Q1 and at
the base and emitter of Q2? The transistors are silicon.
With the circuit in Q6 -12, what would be the maximum peak-to -peak
signal voltage developed in a 75 -ohm load?
Name all of the factors that can cause the CRO presentation shown in
Fig. 6 -22A.
CHAPTER
7
Pulse Processing and
Timing Systems
Television camera chains, both monochrome and color, rely heavily
on precise pulse -processing subsystems for proper operation. Therefore,
it is imperative that the technician understand the principles and circuits
involved in these systems.
7 -1.
THE AUTOMATIC TIMING TECHNIQUE
Older camera chains employed vertical- and horizontal -drive pulses to
generate camera deflection and blanking waveforms. These pulses normally
were inserted in the camera control unit (in the control room) and re-layed to the camera head through the camera cable. The composite blanking signal ( the blanking portion of the transmitted signal) also was inserted in the camera control unit and adjusted in level by the pedestal or
blanking control for the proper setup.
It is important to understand why the driving signals to the camera
chains normally are about one -half the width of their respective blanking
pulses. This is particularly important at the horizontal frequency, as is
shown in Fig. 7 -1. Remember that the transmitted composite blanking and
the driving signals for the camera are inserted at the camera control unit.
However, the camera cable may be as long as 1000 feet in some instances,
and allowance must be made for the cable delay, which is roughly 1.5
microseconds per 1000 feet of cable. Since the total path is to and from
the camera head (2000 feet) , allowance must be made for a total delay
of 3 microseconds. It may be observed from Fig. 7 -1 that if the horizontal drive pulses were of the same width as the horizontal-blanking pulses,
camera blanking would not be ended at the start of the active line interval
in cases where long camera cables are employed (unless the drive is regenerated and narrowed) This effect is not so important at the vertical
frequency, since 3 microseconds is negligible compared with a total of
about 1250 microseconds.
.
271
272
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 7 -2 illustrates an example in which there is a 900 -foot difference
in the distances between two control units and the system blanking distribution. In this case, the blanking pulse is delayed approximately 1.5
microseconds to control unit 1, and only 0.15 microsecond to control unit
2. If the front -porch -width control in the sync generator is adjusted to obtain a normal front porch in the camera-1 signal after sync insertion, then
a switch to the camera -2. signal will result in a lengthened front porch
( Fig. 7 -2B) . Since the receiver retrace is triggered by the leading edge of
horizontal sync and the picture is unblanked by the end of horizontal
blanking, a lengthened front porch causes the picture area to shift to the
left. Similarly, if the front porch is adjusted for normal on the camera -2
signal, a switch to the camera-1 signal will result in a narrowed front
porch, and the picture will shift to the right.
Thus, when camera control units are far from each other with respect
to the system blanking distribution, it is necessary to add delay lines for
the nearest control units to make all delays equal to that of the farthest
unit. This is accomplished most conveniently by using the same length of
feed line to every control unit; excess cable can be coiled up when
necessary.
A similar problem can exist even when centralized control units are
used but the lengths of the camera cables are greatly different. Camera
blanking normally is formed from horizontal- and vertical-drive pulses.
( Vertical -rate pulses are of no consequence in this discussion, since the
amount of delay encountered has no bearing on the long vertical -blanking
interval of 1250 microseconds.) Camera blanking must "fit under" the
10.5 µs (Min)
Horiz
Blanking
Transmitted
Fig. 7 -1. Relationship of pulses
¡--6.67 is lMaxl-{
for horizontal drive
and blanking.
Horiz
Drive
Camera blanking
should be ended by
this time (start of
active line interval).
1000' Cable
13 us)
Delay
Start of Camera Blanking Interval
273
PULSE PROCESSING AND TIMING SYSTEMS
Camera
Control
1
Sync
1000'
Distribution
Blanking
Distribution
Amplifier
100'
Camera
Control
Sync Adder
X X
2
Switcher
Composite
Signal Output
(A) Different cable lengths.
1.6115
Camera -1 Blanking
Delayed 1.51s
Receiver Retrace Trigger
--j
Camera -2 Blanking
Delayed 0.1511s
---j
Scan Shifted to Left
(B) Effect on waveform.
Fig. 7 -2. Effect of blanking delay on
front -porch width.
composite-signal blanking inserted in the control unit. For this reason,
the pickup -tube (camera) blanking normally has a duration of 7 to 9
microseconds compared to the 11- microsecond horizontal-blanking interval
transmitted.
If the camera cable should be as much as 1000 feet long, the total camera
blanking delay is 3 microseconds, as stated previously. It may be observed
that if the duration of camera blanking is greater than 8 microseconds, the
interval is not completed by the end of receiver blanking. When this occurs, the front-porch width is not determined by the delay in the sync
274
TELEVISION BROADCASTING CAMERA CHAINS
generator between the leading edge of blanking and the leading edge of
sync, but rather is actually determined by the end of camera blanking.
Again, if one camera -cable length is only 100 feet, and another camera cable length is 1000 feet, switching from one camera to the other will
cause a picture -area shift on the receiver. In this case, horizontal -drive
pulses to the camera units with the shortest cables must be delayed to
compensate for the delays in the longer cables.
It should be noted that the example of Fig. 7 -1 is for noncomposite
switching, in which sync is added following switching. In the case of
composite switching ( sync as well as blanking inserted in each camera
chain) , the front-porch width is fixed in each camera chain. However,
the sync and blanking must again be coincident at the input of the
switching system so that mixing and lap- dissolves may be made properly
without sync -timing error from two sources.
In color- camera chains (except those of recent design) , an additional
delay over and above that of monochrome cameras results from the
color -processing operations. Thus integration of such color equipment
with existing monochrome camera chains requires an extensive timing
procedure during initial installation.
Pulse- distribution amplifiers are used to provide isolated runs of the required pulses to various points in the plant. Fig. 7 -3 shows the basic pulse
distribution for integration of older color systems with monochrome
equipment. Note that since monochrome and color must be integrated in
this case, all monochrome pulse paths include a delay to match that of
the color system. This is necessary to maintain the same front -porch
width in the composite transmitted signal for both color and monochrome
sources. A shift in front -porch width would cause a shift in the raster
(picture) area at the receiver. Since the normal delay through a color
system is about 1.2 microseconds, horizontal- drive, -blanking, and -sync
pulses must be delayed accordingly for monochrome. It may be observed
that with sync pulses inserted after the final switching point in the system,
if the front porch were set for the normal 1.6 microseconds for monochrome, only about 0.4 microsecond of front porch would exist for the
color system.
Most modern camera chains, both monochrome and color, incorporate
an automatic sensing circuit that compensates for system time delay, thus
circumventing timing problems for various pulse-cable and processing
paths. The sensing circuit detects the time delay in the camera cable, the
encoder, and the color -filtering process, and automatically advances an
internally generated horizontal -drive pulse to avoid introduction of delay
in the outgoing video signal.
Fig. 7 -4 presents the relationship of some of the pulses discussed in this
chapter. Before taking up the technique of how the timing pulse is used
to advance the horizontal -scanning waveform in the camera, we must first
be sure of an understanding of the "boxcar" circuit.
275
PULSE PROCESSING AND TIMING SYSTEMS
Horiz Dr (Air),
Cal-Pulse Gen
1
---.-
Cal Pulse
Horiz Drive to Color Termination
LI
Pulse Box
Delay
1
Film Cameras 7,
Pulse Box
Horiz Dr
--(Standby)
9, 11
2
Shop Bench
Pulse Box
3
Film Cameras 6, 8, 10
Spare
Chk Sw Hence to Termination
Vert Dr IAirl
Pulse Box
1
Film Cameras 7,
Pulse Box
Vert Dr
8,-0-(Standby)
-0
Change Over
)
Pulse Box 3
Film Cameras 6, 8, 10
Spare
Check Sw Hence to Termination
Blanking (Air)
Blanking to Color Termination
Pulse Box
Delay
Chk
Pulse Box
Sw
3
Film Cameras 6, 8, 10
-
-w-
Delay
-
Sync
4-4-(Standby)
To
9, 11
2
Shop Bench
Blanking
(Standby )
Sync
IAIrI
1
Film Cameras 7,
Pulse Box
(0-..
11
Shop Bench
Sync Gen
Switch
9,
2
Chk Sw and
Pulse-Cross-Mon Sw
Spare
Check Sw Hence to Termination
Switcher 1 Sync Adder
Monitor Sync
DA Sync Relays
Switcher
2
Sync Adder
Shop Bench
DA Sync Relays
Switcher
e
3
Spare
Check Sw and Pulse- Cross -Mon Sw
-Pulse Amps Fig. 7 -3. Typical integration of monochrome and older color systems.
The Boxcar Circuit
The boxcar circuit is used extensively in modern color-camera pulse
circuitry. In Fig. 7 -5A, the input capacitance and resistance form a short
time constant and therefore provide differentiator action. In the absence
of a pulse, the transistor is held in saturation because the base is returned
to -10 volts and the emitter is grounded. The ratio of base resistance to
collector resistance (RI,) in a boxcar circuit is about 10 to 1, never more
than 20 to 1, so that saturation is assured. In this state, the collector is
276
TELEVISION BROADCASTING CAMERA CHAINS
very nearly at ground potential and is forward biased relative to the base;
this is the condition for saturation (both junctions forward biased). Thus,
we should expect the dc voltages shown in Fig. 7 -5A.
The applied positive pulse (which always has a peak amplitude nearly
equal to the boxcar supply voltage) reverse -biases the base -emitter junction and drives the transistor to cutoff. Capacitor C attempts to charge
toward -10 volts but is clamped to the base-emitter potential, which
again results in transistor saturation.
-
System
Hor iz-B la n king
Interval
Target
Blanking
Peak
- Picture
White
Picture
Black
White Pulse
Timing Pulse
Horiz
Drive
)
-Gate Pulse
Fig. 7 -4. Pulses in horizontal -blanking interval.
For the duration of the time that the transistor is cut off, the collector is
practically at the supply voltage of -10 volts, resulting in a rectangular
output pulse that is narrower than the input pulse. The output -pulse duration is determined by the time constant of RB and C. The actual width
is 0.7 REC. Thus, the output-pulse duration for the values given in Fig.
7 -5A is:
(0.7) (10,000) (430 x 10 -12)
=3µs
The actual dc voltages measured between the transistor electrodes and
ground depend on the duty cycle of the applied pulse-the length of time
the transistor is on relative to the length of time it is off. This is why it is
always desirable to use a dc scope for troubleshooting, as described earlier.
A chain of test pulses, and any other application in which a pulse must
be triggered from the trailing edge of a preceding pulse, requires boxcars
in cascade as in Fig. 7 -5B. This is the "delayed" pulse technique. Note that
the transistors are now npn. When a positive pulse is applied the base
cannot go further positive because of the clamping action of the base-
277
PULSE PROCESSING AND TIMING SYSTEMS
emitter junction. The short time constant causes a large negative signal
excursion at the trailing edge of the applied pulse, and the first transistor,
Q 1, is driven to cutoff. This results in a rectangular, positive pulse at the
QI collector for the time the base is negative relative to the +0.7 -volt
-10 v
1k
0.1
V
Cl
I430
pF
07
10VJ
(A) Single pnp stage.
+9,3V--'
-0.7V -I, ,s.7.I
I
-O,1V-j
`.:10V
II
-9.9 V---ILJ
0.7RC--1 I-+10V
10V
+10 V
+0.7V
-9.3V
+10V
+0.1 V
-a
____
--- -
,.
1
-9,3V
I
+10V
+0,7V T
1
I
+10 V
I
é-{
+O.1V
(B) Cascaded npn stages.
Fig. 7 -5. Boxcar circuits.
potential (see waveforms in Fig. 7 -5B) . The same action then occurs at
the Q2 base, and the leading edge of the Q2 collector pulse occurs at the
trailing edge of the Q1 collector pulse.
You can see that in a circuit using pnp transistors (Fig. 7 -5A), if the
applied pulse is negative -going the output pulse is delayed. If the transistors are npn and the applied pulse is negative-going, an undelayed output
278
TELEVISION BROADCASTING CAMERA CHAINS
pulse results. By using this knowledge, you should be able to know exactly
what to expect, including the pulse duration, in waveform tracing.
Time Advance
See Fig. 7 -6. The basic function of this circuit is to sense the delay
through the individual camera chain (including the luminance delay in
the encoder) , compare it to incoming sync, and derive a dc error voltage
to start camera drives and blanking to compensate for this total delay.
One output of the horizontal -sync separator in the control -room rack
equipment goes down the camera cable to an automatic time -correction
( ATC) multivibrator (Fig. 7 -6A) This multivibrator turns off after
about two- thirds of a line (Fig. 7 -6B) , depending on a dc control voltage
from the comparator. The trailing edge of the multivibrator pulse initiates
operation of a delay boxcar. The trailing edge of the delay pulse starts a
timing -pulse boxcar, the output of which is inserted into the camera video
(during blanking time) and fed back to the encoding equipment in the
control room. This pulse is separated from video and fed to the other
end of the comparator, where the trailing edge is compared to the leading
edge of sync. The resultant dc error signal is the off control for the ATC
multivibrator.
.
Video
Out
Timing Pulse
Separator
Y Delay
Proc
Amp
Error
Delay Time
Gene ator
DC
(Boxcars/
w Video
I
I¡
Drives
a To Horiz
Camera Head
I
n
"Off"
Control
System
SD
nc
In
Volts
Separation
i
ATC
-o-
MV
Regeneration
"On" Drive
Pulse
Camera
Cable
Control -Room Racks
Camera Head.
In Studio
(A) Block diagram.
Horiz Sync Input
at Racks
Camera Drive
Width Adjusted
by DC
Error
--
--I
(B) Pulse waveforms.
Fig. 7 -6. Delay compensation for color camera chain.
'
PULSE PROCESSING AND TIMING SYSTEMS
279
By using the leading edge of the timing pulse to derive horizontal drive
and blanking, the camera signals are advanced relative to incoming sync
timing. The amount of advance depends on the corresponding cable length
and Y delay. In the camera head, the pulse that corresponds to a given
video line is initiated by the sync pulse that corresponds to the preceding
line in time.
If a failure occurs in the timing -pulse loop, the picture will not be
lost but will shift horizontally, and the front porch will not be correct.
Since this timing is a line -to -line function, an intermittent condition can
cause erratic shifts of portions of the lines in the raster. Faulty camera
cables or connectors can cause this condition. If camera horizontal -drive
multivibrators are of the driven type, requiring pulses derived from the
timing section, lack of scanning will activate the pickup -tube protection
circuits and disable the camera (Section 7 -5) .
7 -2. LEVEL -CONTROL PULSES
(MANUAL AND AUTOMATIC)
In modern color- camera chains, pulses are used not only to calibrate
individual amplifier gains (Section 7 -3) but also for manual and automatic video -level control. Before we examine this circuitry more closely,
it is necessary to understand the nonadditive mixing (NAM) technique
that is generally employed.
Nonadditive Mixing
The three- channel camera derives luminance from the three primary color channels. In the four-channel camera, the three color tubes operate
on color brightness only; the luminance channel provides the "true"
luminance information, which can be different from chroma luminance.
Therefore, to obtain proper color balance in camera setup, proper
luminance-to-chrominance ratio, and proper "gain riding" of signal level,
the NAM signals are provided.
Review Chapter 2 in Television Broadcasting: Equipment, Systems, and
Operating Fundamentals. The NAM circuits are fed from a "receiver
matrix" based on the principles discussed in that chapter, with this fundamental difference: the brightness levels of B, R, and G are derived as
follows:
(B -Y') +Y =B
(R -Y') +Y =R
(G -Y') +Y =G
where,
Y'
is the "derived" brightness,
Y is the "true" brightness.
In other words, the color-difference signals are obtained by matrixing from
the I and Q circuitry prior to modulation, and then the "true" luminance
280
TELEVISION BROADCASTING CAMERA CHAINS
from the Y (monochrome) channel is added. When the color chain is
operated in the test position with color bars, the luminance provided is the
"derived" value, since the luminance tube is not contributing to the signal (nor are any of the color tubes) So now the monitoring facilities in
NAM positions will be looking at a signal as the color receiver sees it.
Now study the NAM gated detectors of Fig. 7 -7. The red, green, and
blue signals just described are applied to the detectors as shown in the
diagram. Note that the negative detectors are pnp, whereas the positive
detectors are npn. Assume for the moment that the gates are saturated and
the emitters of all transistors are held at essentially zero (ground) potential. Further note that under this condition positive -going pulses have no
effect on the npn transistors.
Now assume that the red signal momentarily has the highest amplitude.
The negative excursion will turn Q2 on, and the resultant negative pulse
at the emitter (hence across R1) will hold Q1 and Q3 off since their bases
are not as negative as the base of Q2. Thus, only the red negative -going
signal will appear across R1. This is to point out that the largest negative going signal at any instant will appear across R1, holding the remaining
transistors off.
Similarly, the signal with the largest positive -going swing will appear
across R2, holding the other two transistors off. This is the basic principle
.
Negative Detector
G
Positive Detector
Input
1
1.51111-
T
R
Q
Input
irLiliÓ
1
T
2
B
Q5
Input
1
l_n_n_J
T
Q3
Ó
Q
/%/
--N/-
_EL
To NAM
To NAM
Monitors
R2
R3
Q7
R4
Q8 Gate
Gate
Fig. 7 -7. Basic
circuit for nonadditive mixing.
Monitors
0
PULSE PROCESSING AND TIMING SYSTEMS
281
of NAM: only that signal with the largest amplitude will be passed, without being added to the other signals (nonadditive mix) .
Now look at the gates, Q7 and Q8. They are held in saturation by the
base currents through R3 and R4, until negative pulses arrive at the bases
to turn them off. Since these pulses are from opposite sides of the NAM
flip-flop, one transistor is gated on when the other is gated off. The pulse
durations are such that three to four lines of NAM white are on, then
three to four lines of NAM black are on, and so on throughout the entire
field. You can see that which gate is monitoring white and which gate is
monitoring black depends on the video polarity at the point of detection.
The important point is that with NAM, the highest -level signal in any of
the channels at the time of sampling is monitored. This is one form of electronic switch.
The Nature of Automatic and Manual Control Circuitry
Modern color cameras have one thing in common; either automatic
is a more appropriate term) or manual control of
white and black levels is available. You will need an orientation in this
new type of operation, because it is a different concept in manual control
as well as in automatic control.
There are two basic control functions:
( "semiautomatic"
Manual control does not sense any picture information. White level
normally is set by a manually adjusted "white pulse" inserted during
blanking time.
2. Automatic control senses peak white picture information, detects
this for a dc control voltage, and uses as a reference a standard
set by the operator. (NOTE: Automatic control normally is employed
only in film camera chains.)
1.
First, study Fig. 7 -8A. Note that the master white -level control sets a
reference dc level for all four channels simultaneously. In addition, the
master chroma control sets the reference dc voltage for all three color
tubes. Note carefully that when the switch is in the "auto" position, the
master white control still supplies a reference dc, in this case through the
NAM detector circuitry. Remember that the NAM detector supplies only
one output at a time, from the channel with the largest instantaneous peak
amplitude, and in this instance we are concerned with the peak white
detector.
NOTE: In some cases, the master controls mentioned above are simply
termed "white" and "chroma." The individual balance controls are then
designated "mono," "red," "green," and "blue."
Now study Fig. 7 -8B for manual control of white level. A white pilot
pulse is inserted into the video; this pulse is timed from the trailing edge
of a master timing pulse that occurs during horizontal-sync time (Fig.
TELEVISION BROADCASTING CAMERA CHAINS
282
. Notice that the Y- channel control voltage also feeds the white -pulse
amplitude modulator stage. This control voltage sets the amplitude of the
white pulse at this point.
The gain -controlled amplifier usually contains a pair of transistors with
a feedback resistor. Around this feedback resistor a Raysistor is used.
A Raysistor is a light- controlled resistance device consisting of photocells
illuminated by lamps. AS the lamp current is changed, the resistance changes
accordingly. The lamp current is provided by a dc amplifier that receives
the output of the peak detector.
Thus far, all we have is a white pulse that occurs during sync intervals
and that is controlled in amplitude by the master white control on the
remote panel. This combined video and pulse signal is sent through more
amplification (not shown) and returned as the sample video shown in
7 -4)
White Balance Controls
-12.5
V
Man
NAM Det
Master Chroma
-o
Auto
Mas er
White Level
To R
To G
To B
Master
AWL,
To Y
(A) Control -panel circuits.
AWL
Gating
-i
Auto
AWL Ref Amp
Man
Peak Det
Amp
w-
Sample
Video
(Auto or Man)
Horiz Drive
Ref Volts
DC
AWL Override
From
Auto
Y Feed
in
Fig.7-8A
Standard
White
Pulse
o
Man
White Pulse
Amplitude
Modulator
During Horiz Sync 1
DC Amp
P
lot
Raysistor Filament Current Control
Rays'stor
Video
Gain-Controlled
Input
Amp
(B) Amplifiers in camera head.
Fig. 7 -8. Control of white level.
Output
PULSE PROCESSING AND TIMING SYSTEMS
283
Fig. 7 -8B. Now the closed -loop nature of just the simple manual white level control begins to become apparent. It is all in the interest of providing automatic level control, as we will see shortly.
We are still concerned with manual control at this point. Note that for
the manual position of the operating switch the automatic white level
( AWL) amplifier is receiving horizontal -drive pulses, which occur during the first part of the horizontal -blanking time ( hence sync interval) .
Therefore, the sampling of the entire video interval being fed back to one
side of the peak detector is only during horizontal-blanking time; hence,
only the pilot white pulse is sampled. The resultant dc depends on the
amplitude determined by the manual adjustment of the white -level control. This direct current is fed to the lamps of the Raysistor in the gain controlled stage, determining the amplitude of the video output.
What we actually have done is to provide a convenient means of going
to automatic control, using the same circuitry except for the addition of
a reference from peak white video. In the automatic mode, the AWL
reference amplifier is gated by a wide pulse that rejects the blanking
interval and samples the video interval. If the peak white video being fed
back to the peak detector is higher than the reference, the dc error signal
that results has a direction to decrease the gain of the controlled stage.
The error signal is in the opposite direction when the peak video is lower
than the reference. A bias voltage known as AWL override is fed into
the white -pulse amplitude modulator; this voltage prevents the gain of
the controlled stage from increasing on complete fades to black. Such a
gain increase would increase the noise level.
Manual and automatic black -level control works essentially in the same
way as white -level controls. Remember the description of NAM circuitry,
and the basic function should be clear. For manual control, a black pulse
is added to the video during the horizontal -blanking interval. The amplitude of this pulse is governed by the control -panel black -level control.
For automatic black control, the output of the NAM circuit responding
to the negative -going signals (signals approaching black) is peak detected.
Just as in automatic white control, the black -level control adjusts the
ABL detector reference voltage for black. This emphasizes the fact that
the white- and black -level controls on the panel are just as functional in
the automatic mode as in the manual mode. In the automatic mode, these
controls are the source of the reference voltage for the associated automatic
level detector.
When optical black is provided, the black-level detector is gated so
that it responds to an unilluminated portion of the pickup tubes (this
area is provided by an optical mask) . Thus, when the automatic black
level for each channel is adjusted in initial setup, the black levels of the
individual channels are held at a reference regardless of dark- current
variation caused by temperature excursions or other factors. This feature
normally is used only in nonbroadcast applications such as closed-circuit
284
TELEVISION BROADCASTING CAMERA CHAINS
installations, since the optical -mask black portion must be in the active
picture area.
7 -3.
LEVEL CALIBRATION AND TEST PULSES
Nearly all modern camera chains employ calibration and test pulses for
setting amplifier gain and checking amplifier continuity. First, we will consider typical techniques used in film camera chains.
All channel gains, white levels, and black levels must be standardized,
or matched, before color -balancing procedures are carried out. Modern
color chains (both film and studio) provide for this by inserting reference
pulses equivalent to pickup -tube beam currents at the preamplifier inputs.
Fig. 7 -9 shows the basic principle of amplifier calibration. Assume
a calibration pulse (pulse No. 1) is injected at the output of the camera
head. Further assume this calibration pulse is equivalent to the proper
peak output signal of the camera, for example 0.7 volt peak to peak.
Then a reference pulse (pulse No. 2) is inserted into the pickup-tube
target load; this pulse is equivalent to peak high -light target current
( normally 0.3 p,A for 1 -inch vidicons or 0.6 /LA for 11 -inch vidicons
used in the luminance channel). Now if we adjust the camera amplifier gain and black -level controls for the channel involved so that this pulse
matches the calibration pulse, we have "standardized" the channel gains.
If the pickup tube has the electrode voltages applied (but the lens
capped), the dark current is present for proper setting of the black level.
Fig. 7 -10 shows the usual positioning of pulses in the four- channel film
camera. The luminance channel is given the pulses of Fig. 7 -9 (the refCamera
Cable
Camera
Control
Composite Output Level
Gain
Proc Amps
& Encoder
Proc & Video
Amps
Output Amps
Monitoring Facilities
Ref
Cal PulseW
Pulsel2)
Cal
Pulselll
Ref
J
Pulse(2)
White Level Ref
}
Black Levels
Active Video Region
Fig. 7 -9. Principles of amplifier calibration.
285
PULSE PROCESSING AND TIMING SYSTEMS
erence pulse is now pulse No. 3) plus a pulse timed with the green -channel calibration pulse (pulse No. 2) . Pulse 2 in the luminance channel is
at 0.59 of unity level (the luminance level of green) for proper matrixing
( see Fig. 7 -11) . The blue and red channels receive only the reference
pulse (No. 3) . (Non: this last pulse might be designated No. 5, as in
the RCA system, because of spacing.) Such an arrangement makes it possible to carry out balancing procedures with test pulses merely by looking
at the picture monitor (on NAM function) . When black levels are
matched, no "stripes" appear in the No. 1 white pulse on the monitor.
1
2
Ref
0.59M
-J-
Blk level
G
Fig. 7 -10. Pulse positions for
four -vidicon monitoring setup.
B&R
1
11
2
3
Raster
A Gray
Black
When white levels are matched, no stripes appear in the No.
3 (or No.
5) white pulse on the monitor. A CRO presentation will show all white
and black levels identical. White-balance controls are simply amplifier
gain controls. Black- balance controls usually set a clipped reference voltage
for black-level control.
NOTE: Color-camera setup techniques are covered thoroughly in Chapter
10 of Television Broadcasting: Equipment, Systems, and Operating Fundamentals.
After the channel gains are calibrated, we have a working standard
against which to "color balance" the pickup -tube operating parameters.
The basic requirement (assuming the encoding system has been optimized
on color bars) is that zero subcarrier must occur for black, white, and all
shades of gray. This, in turn, demands that the pickup -tube outputs be
TELEVISION BROADCASTING CAMERA CHAINS
286
identical in amplitude (at white, and all shades of gray to black) when
the optical system is looking at a monochrome gray scale.
In this case, a logarithmic transmission -chart slide (for film chains)
is used, and the scope is connected to monitor the camera output signal.
The slide -projector lamp must be operated at a voltage near the normal
line voltage to obtain proper color temperature. Bear in mind also that
in this adjustment compensation is made for any film -base color temperature that might exist, and that this can be different for different film.
The basic initial adjustments are:
Target Voltage: Adjust to obtain the required peak white level.
Black -level adjustment: This adjustment is required since any change
in target voltage can change the dark current.
NOTE: If the camera chain employs "optical black" circuitry, the vidicon
dark -current level is held automatically at a constant pedestal regardless
of dark-current variations under target adjustments.
The procedure is simply to adjust the target voltage of one channel at
obtain the reference white level for the white chips. Then readjust (if necessary) the black level for each channel to obtain reference
black. When these steps are completed, you are ready for the more time consuming procedure of obtaining tracking at all steps of the gray -scale
signal, so that all channel output signals are superimposed on the individual steps from black to white.
a time to
Cal Pulse
0.596
i
Ref Pulse
Timing Pulse
(A) Pulses in camera -head monochrome channel.
Cal Pulse
Encoded Green
Ref Pulse
-Ref White
-Ref
Black
(B) Encoded composite monochrome -green signal.
Fig. 7 -11. Test pulses in RCA color
camera.
PULSE PROCESSING AND TIMING SYSTEMS
287
After encoding, the presence of subcarrier in the gray areas but not in
the white and black areas indicates mistracking. There are two major
causes of lack of tracking: (1) gamma- correction circuitry and (2) spectral sensitivity of the optical paths.
Cameras employing adjustable gamma circuits in the luminance and
chrominance channels are readily corrected in case of gamma mismatch.
Sometimes only the luminance channel is adjustable, with either fixed
gamma correction or no gamma correction in the chrominance channels.
In other cases, a fixed gamma correction of 0.7, 0.5, or unity is selectable
by means of a switch. Whatever method is used in a particular camera,
the best possible tracking should be obtained by experiment before attention is given to the second cause listed above. The second condition must
be corrected by using neutral- density discs in front of one or more of the
chrominance -tube lenses.
If you have a properly adjusted color monitor (one that produces a
truly black and white monochrome picture when color burst is present) ,
you can tell immediately where to start in selecting a proper neutral-density
filter to be inserted in front of the pickup -tube lens. For example, if grays
are greenish ( this is the most common condition) , start with a neutral
filter of about 0.2 in front of the lens of the green tube. You must, of
course, readjust the target voltage of this tube to bring white level back
to reference white. Since the target voltage has been readjusted, check
for any change in green black level, and adjust if necessary. If grays have
now gone "minus green" (purple) , the neutral- density filter is too dense,
and a 0.1 filter should be tried. Once the optical paths are balanced, it
normally is not necessary to change filters when vidicons are changed.
When the film camera has been properly registered and tracked, it is
imperative, in practical film operation, to balance whites and blacks of each
camera channel for the particular film base and color processing method.
For example, a film base can be slightly blue in color temperature. If it is
possible to still -frame the color projector, do this on the first frame that
has a white and a black area. If the base is blue, the blue channel will
have higher setup in the black region than other channels. Balance black
and white gains. This is the only way you can keep the flesh -tone concept
in telecasting color film. It requires some amount of rehearsal time for
previewing the film, but if you are critical in operations, the time is
well spent.
A good example of what can happen is illustrated by the following
specific example. Assume two studio color cameras have been well balanced
to one another. The master color monitor shows good flesh tones for the
live pickups. Now stop and think a moment. You have balanced and
tracked the camera on the scene ( essentially) by balancing on a chip chart under the studio lights for the scenes to be used. You "roll film"
for a film clip or commercial; the flesh-tones turn greenish or bluish on
the monitor. What has happened? Simply this: Many stations balance on
TELEVISION BROADCASTING CAMERA CHAINS
288
the gray -scale slide, but this is no guarantee that flesh tones will be correct on the film clip, which may have a base with a different color
temperature. All encoders can be showing proper phase setup on color
bars, but flesh tones may not be correct from all color sources. This is
the result of camera color balance. Such a condition can exist between
studio cameras not properly color balanced, or between live and film
sources not properly color balanced.
Another basic cause of this trouble is lack of proper color balance between slide projectors and film projectors. You should have the gray -scale
slide mentioned, and also film loops of the same gray scale for the film
projectors. You should balance the film projectors (if necessary) to each
other by slight adjustment of projection -lamp voltages. Normally, voltages
below about 105 volts cannot be used on projection lamps; less voltage
causes the color temperature to go toward red.
You should obtain two Wratten GL filters, one CC10B and one CC20B.
Place one of these filters behind the lens of the slide projector and start
with a lamp voltage of 100 volts. Adjust the lamp voltage and try the two
filters one at a time until the camera stays in balance between the slide projector and film -projector gray scales. Once you get good balance in this
way, a minimum of adjustments should be necessary in going from one
film to another.
Some manufacturers now supply specific types of heat filters that colorcorrect slide -projector light sources to specific film projectors. Always investigate current manufacturers' recommendations in color balancing of
separate light sources for slide and film projectors.
Fig. 7 -12A shows the major difference between the setup of a studio
camera and that of a film camera. When a camera has an I.O. and three
vidicons, a major difference in signal outputs requires a modification in
reference -pulse use. Because of the electron- multiplier output of the I.O.,
the signal output per lumen is greater for this tube than for the vidicon.
I.O.
0.6
Video
Amp
FT-
Video
Amp
Preamp
I
4-
Video
Preamp
Amp
Pole
-
Comp
I
Ht-g-
Preamp
I
I
Video
Amp
Injection of
0.3
Ref Pulse Here
Injection
of
Ref Pulse Here
(A) Camera with 1.0. and vidicons.
(B) Camera with Plumbicon tubes.
Fig. 7 -12. Injection points for test pulses.
289
PULSE PROCESSING AND TIMING SYSTEMS
The 41/2-inch I.O. normally is operated at about 0.6 to 0.7 pA, which is
about a 2 -pA peak-to -peak current swing on high lights. Therefore, the
reference level is at the preamplifier outputs for the vidicons, but at the
first amplifier stage for the I.O.
When the camera is placed in the test -pulse mode of operation, continuity of circuit function is complete except for the vidicon preamplifiers.
Therefore, variable or switchable gain controls are included in the vidicon
preamplifiers to obtain the desired reference signal voltage from all
channels.
Plumbicons generally are operated at about 0.3 p,A peak white current,
the same as for vidicons. In a camera employing three or four of these
tubes (Fig. 7 -12B) , all outputs are made equal. The additional control
(or fixed circuit) to be found here is the pole compensation to accommodate the target capacitance of the Plumbicon, which requires compensation around 100 kHz in some camera tubes.
The initial level setups for the live ( studio) camera are the same as
those described for the film camera. When you set up first on test pulses,
you have removed the variables provided by the control panel, and you
have standardized white and black levels for all channels. From this point
on, the techniques are somewhat different, and these differences will be
covered in the next few paragraphs.
The first step in cameras employing the I.O. is to find the knee. It does
not matter how much light the camera "sees"; the iris is opened until
the steps of the chip chart, as observed on the CRO, start to compress on
the top white wedges. This statement assumes the proper neutral-density
filter has been installed in front of the I.O. luminance tube ( when used
with three vidicons in the chrominance channels) For example, the instruction book may call for the knee to occur at exactly f/8 with 250
foot -candles of incident illumination on the chip chart. The proper
neutral- density filter must be installed for this to occur.
Regardless of what you may have heard about color- camera operation
employing an I.O. and vidicons, you are going to operate the I.O. at
about one -half or even one f/ stop over the knee. If you attempt to operate
just below the knee, the picture on both monochrome and color receivers
will be "washed out." Remember that the vidicon does not have a
"natural knee "; the chroma signals keep on increasing as the light is increased or as the iris is opened. "Artificial knees" (a form of signal
limiting over a given amplitude) for vidicons have been tried, but, although these help to some extent, you still must use considerable judgement in setup procedures and operations.
You can understand that it is easier to track three or four tubes of the
same type than it is to track three vidicons with an I.O. tube. But it can
be done satisfactorily if you understand what you are doing.
When you have found the I.O. knee, double check the target -voltage
setting for 2.3 to 2.5 volts (or whatever voltage is specified) above cut.
TELEVISION BROADCASTING CAMERA CHAINS
290
off. This procedure normally involves a switch that, when thrown to the
target-set position, places a negative bias of 2.3 to 2.5 volts on the target.
The target voltage is adjusted until the two or three lightest chips can be
observed with this bias applied. Then the switch is returned to the normal
operating position. Remember that every time you readjust target voltage
on any tube (except the Plumbicon) you must recheck the black-level
setting.
The next step is to measure output signal level of the I.O. channel.
This is normally 0.7 volt peak to peak at the camera output, which matches
the calibration pulse observable at this point. If the peak white signal is
higher or lower than this (operating at 1/2 stop over the knee) , adjust the
dynode gain for proper peak white output. It is very important now to
recheck the monochrome black level. If necessary, adjust both the automatic black -level (ABL) and manual black -level (MBL) controls (both
are involved, even in automatic operation) for proper black levels in
automatic and manual operating modes.
(A) Black -level shading, excessive
(B) Minimum subcarrier, even
chroma level.
black levels.
Fig.
7 -13.
Video waveforms for crossed gray scale (wideband scope).
At this time, while you are observing the crossed gray scale of the
monochrome channel (only) , remember that the scope should show linear
steps from the logarithmic gray scale. Adjust the monochrome gamma
circuits for the best amplitude linearity from black to white. If the black
levels at the two extremes of the scale do not match (Fig. 7 -13), adjust
the shading slightly for proper balance. (Shading is adjusted initially for
flat black levels.) NOTE: If you observe excessive "grass" at black level
and even in the blacker-than -black region, either the multiplier -focus
voltage is wrong or the vertical sweep of the I.O. tube is not properly
centered. The multiplier -focus voltage should be adjusted for maximum
output level with the best shading. The I.O. sweep centering should be
checked with the size gauge normally provided. This gauge consists of a
transparent ring to be inserted temporarily in front of the I.O. tube. The
sweeps are centered and sized so that the inner circle just meets the top
and bottom of the raster blanking and the outer circle just meets both
sides of the raster blanking. This assures proper aspect ratio and size.
PULSE PROCESSING AND TIMING SYSTEMS
291
IMPORTANT: After resetting the multiplier focus or shading, recheck
the black -level adjustments.
Before attempting to track the chroma tubes with the I.O., always be
sure the sweeps for each tube are centered and sized properly on the mask
of the chroma field lens.
Because of the many variables, the tracking procedure for the I.0.-vidicon combination is a little more involved than that described for the film
chain. When vidicon target- current meters are provided, simply adjust
the target voltage for 0.3 µA peak high -light current, and then match
levels with the I.O. by means of the adjustable vidicon -preamplifier gains.
When target-current meters are not provided, the preamplifier gain must
be made as low as possible consistent with reasonable target voltage required to match the I.O. output signal level. If the black level for the particular channel involved becomes so high that it is beyond the range of
the ABL and MBL controls, excessive preamplifier gain is indicated. Except for this complication, the target -balancing procedures are the same
as those described for the film -camera setup.
When the targets in the camera have been balanced, return the test
switches to the operate position and adjust the white- and black -balance
controls on the remote control panel in the control room. The first step
is to select the monochrome channel and adjust for a half stop over the
knee to check for proper output signal level. Then select the green -channel
signal, and adjust the master chroma control to get the same output level
as for monochrome (using the chip chart) . Then (normally) go to NAM,
and balance the black and white levels for red and blue on the picture
monitor as was done for the film chain. Remember that this is proper
"color balance "; the subcarrier must be as nearly zero as possible for all
shades of gray (Fig. 7 -13). The matching of color cameras to one another
on an actual color scene will then involve only minor adjustments.
To check "color balance" between two or more cameras, a common,
switchable color monitor is used to observe all camera outputs. First of
all, what "color" does this monitor have when "looking at" a color camera
operating in the monochrome mode (color subcarrier removed) ? The picture should be strictly black and white; if it is not, and you do not have
time to adjust the monitor, remember the color, and fix it in your mind as
your black-and -white reference.
Next, bear this in mind: Even though you have "balanced" all color
cameras on a gray scale, there may be a different skin tone from each
camera. Why? This results from the difference in gamma correction if
the fault is in luminance levels around the skin -tone reflectance area. This
is to say that the gray scale must be the same on all cameras as well as
being balanced for minimum subcarrier; i.e., all gray -scale steps must be
at the same level relative to capped level. To achieve this, it might be
necessary to make minor adjustments to gamma and/or target voltages and
I
TELEVISION BROADCASTING CAMERA CHAINS
292
then rebalance for black and white levels. This problem is common to all
types of color cameras, including the three- or four -Plumbicon camera.
However, only the gamma- correction circuitry is involved in the latter
cameras, since target voltage on the Plumbicon has a minimum effect on
sensitivity or gray -scale tracking.
It should be obvious that all cameras should be color balanced on the
same gray scale under the same light conditions. Then, when you switch
between cameras on the common color monitor, slightly different colors
of "gray" can be compensated for by slight adjustments of the black -balance
controls. The final check is to observe a live model under the lighting to
be used ( which should be the same as for the gray -scale setup) and select
the camera with the most pleasing skin tones as the "standard."
When the studio camera employs amplitude controls in aperture -correction circuitry, the camera picture monitor can be fed from the strobe
output of a suitable waveform monitor (such as the Tektronix 529) with
the delayed sweep adjusted to the 300 -line point of the EIA resolution
chart (Fig. 7-14). In most cases, with well designed cameras, this amplitude control can be adjusted to obtain a 100 -percent response relative to
picture white at 300 lines resolution. Remember that the amount of picture noise introduced is the limiting factor. The vidicon and the Plumbicon
can stand more aperture correction than can the image orthicon for a
given noise level at the output.
NOTE: In Fig. 7 -14, the scope time base was adjusted to five TV lines,
rather than a single line, so that the brightening pulse would be readily
apparent in the photo.
-
Fig. 7 -14. Picture monitor fed from
line- strobe oscilloscope.
\IÌIIIINhI`- -`"I
7 -4. CAMERA DEFLECTION CIRCUITRY
Deflection -yoke assemblies were covered in Chapter 4. This section will
describe the pulse formation and circuitry necessary to drive sawtooth currents through the yoke assemblies.
Sawtooth generators of the vacuum -tube type generally incorporate one
of the two basic circuits, a blocking oscillator (BO) or a multivibrator
(MV) . The sequence of operation for the blocking oscillator (Fig. 7-15)
is as follows:
293
PULSE PROCESSING AND TIMING SYSTEMS
When the grid of VIA swings positive, plate current increases, and
plate voltage decreases (because of the drop across RL) The increased
voltage drop across the transformer primary is coupled into the secondary
winding with such polarity as to reinforce the positive swing of the grid.
This is known as a feedback cycle, which drives the grid positive to a
point at which no further increase in the plate current is obtained. At this
time, transformer feedback ceases (no current change through the inductance) , and the grid voltage falls off rapidly. Therefore, the plate
current decreases rapidly, and the plate voltage increases. The new change
in plate current causes a voltage of opposite polarity to appear across
the primary, initiating a negative voltage on the grid and, therefore, a
new feedback cycle of opposite polarity. At this time the tube is biased
well beyond cutoff. The large negative charge on grid capacitor Cl (a result of the previous grid current) begins to flow through grid resistor R1
while the tube remains non -conducting. When the capacitor discharges,
the grid again reaches a potential at which the tube starts to conduct.
The time allowed for the capacitive charge to leak off is determined by
the time constant of the RC combination. This time constant is made
somewhat longer than the interval between drive pulses to that the
pulse triggers exert positive control over the timing of the sawtooth
waveform.
The second section of the tube (V 1B in Fig. 7-15 ) , is the discharge
section. The grids of both sections are tied together and therefore receive
the same voltages simultaneously. When the oscillator section is conduct.
Blocking -Osc Section
Discharge Section
Charging
Section
B+
R3
Saw Amplitude
R2
Sawtooth
Output
VIA
T1
V1B
Cl
Pulse
Saw
Input
-Forming
Capacitor
I
Rl
-Trace --1
Retrace
Across C2
Cutoff Level
U
Fig. 7 -15. Basic blocking -oscillator circuit.
-
Waveform With
C2 Open
294
TELEVISION BROADCASTING CAMERA CHAINS
ing (grid positive), the discharge section also conducts because its grid
is also positive. When the grids become negative, capacitor C2 in the
plate circuit of the discharge section slowly charges through resistors R2
(adjustable) and R3. This generates the trace portion of the sawtooth
waveform. When the tube is triggered into conduction by the driving
pulse, the capacitor rapidly discharges, generating the retrace portion of
the waveform.
Variable resistor R2 is used to adjust the voltage toward which C2 can
charge; hence it determines the sawtooth amplitude. If the circuit is for
vertical deflection, the control is termed "height." If the circuit is for
horizontal deflection, the control is a width control.
Fig. 7 -16 illustrates a basic multivibrator circuit. Essentially, such a
circuit provides feedback action between two tubes (usually a single duo triode type) so that one tube conducts while the other is nonconducting,
then vice -versa on the succeeding alternation. The similarity of circuit
action to that of the blocking oscillator described above will become
obvious in the following discussion.
----/\/\
Drive
Pulse
R6
Amplitude
Control
R7
)__h__
B,
Fig. 7 -16. Basic
multivibrator circuit.
Assume for the moment that the grid of V2 is swinging in the positive
direction. The plate current of V2 will increase, and since this current
passes through common cathode resistor R2, V1 receives a negative signal
(grid more negative with respect to cathode because of increased voltage
drop across R2) Thus the plate current of V1 decreases, and its plate
voltage increases. As the V1 plate voltage increases, the grid voltage of
V2 increases still further in the positive direction, thus reinforcing the
initial increase in this voltage. Here it may be seen that tube V1 is serving
as the feedback tube, similar in action to the transformer in the blocking
oscillator described above. The grid voltage of V2 increases in the positive
direction until no further increase of V2 plate current is able to take place.
At this time, since no further change of plate current occurs, feedback
ceases, the V2 grid voltage begins to decrease, and the feedback cycle is
.
PULSE PROCESSING AND TIMING SYSTEMS
295
reversed. Since the negative grid voltage on VI is now decreased, the V1
plate current increases (tube starts conducting) , and the resulting reduced
plate voltage on V1 drives the grid of V2 below cutoff. Then V2 remains
nonconducting until capacitor Cl discharges sufficiently through R4 and
RS. During this interval, capacitor C2 is charging through resistors R6
and R7, and the sweep of the sawtooth is formed as shown in the diagram.
When V2 starts conducting, capacitor C2 rapidly discharges through the
tube, and the sawtooth waveform returns rapidly to zero as shown. Since
115 is a variable resistance in the grid circuit of V2, it provides a means
of determining the rate of discharge of C1 and, hence, the frequency of
the sawtooth wave. Resistor R7 in the plate circuit of V2 determines
the amount of charge placed on C2 while the tube is not conducting;
hence, it provides a means of adjusting the amplitude of the sawtooth
waveform.
+10V
D
Drive Pulse
Fig. 7 -17. Basic solid -state sawtooth generator.
There are many variations, but Fig. 7 -17 illustrates the basic solid -state
sawtooth generator. In the interval between drive pulses, capacitor C
charges toward the supply voltage, since the transistor is cut off (switch
open). This forms the trace portion of the sawtooth. At pulse time, the
transistor is saturated (switch closed) to discharge C rapidly. This is the
retrace portion of the waveform. The time constant, RC, is made long
compared to the interval between pulses so that only a small part of the
exponential capacitor charge is used. This assures excellent linearity of the
trace portion of the sawtooth.
Deflection -Coil Driving Circuitry
Fig. 7 -18 shows a basic tube -type horizontal -deflection output circuit.
The sawtooth voltage wave is applied to the grid of V1, termed the driver
tube. It may be seen from the diagram that the top of deflection transformer T1 is connected to the top of the horizontal- deflection coil (across
the damper tube). When the driver tube is conducting, current in the
horizontal- deflection coil is increasing, as shown in Fig. 7 -19. When the
296
TELEVISION BROADCASTING CAMERA CHAINS
H
Horiz Deflection
Peak
Coil
Driver
VI
From
Horiz
Sawtooth
Horiz
Center
Fig. 7-18. Basic tube -type horizontal- output circuit.
driver tube rapidly decreases conduction, the coil current also rapidly
decreases, as shown in Fig. 7 -19. This interval is the flyback time. It is well
known that a rapid change of current through an inductance creates a
voltage surge that is dependent on the rate of change of the current and
the self- inductance of the coil. The rapid decrease in conduction is sometimes used as the source of voltage for the high -voltage rectifier (flyback
type) The yoke inductance and its distributed capacitance are said to
ring in a damped oscillatory fashion for about a half cycle. The tube which
now comes into operation is V2, the damper tube. This tube conducts and
causes current through the coil in the direction shown by Fig. 7 -19. Thus,
it may be seen that the scanning (sawtooth) current through the horizontal- deflection coil is supplied alternately by the driver and the damper.
In practice, it is found that the driver tube supplies sweep for the right
side of the picture, and the damper tube supplies sweep for the left portion
of the picture.
The control (labeled "Horiz Lin C" in Fig. 7 -18) in the driver-tube
cathode, in conjunction with the control labeled "H Peak," adjusts the
linearity of the sweep. These controls are largely effective only on the
center to right side of the raster (during driver -tube conduction)
.
.
Left Side of Raster
Driver-Tube
Conduction
Damper-Tube
Flyback
Conduction
I
Right Side of Raster
Driver -Tube
Conduction
Period
Fig. 7 -19. Current waveform for horizontal sweep.
PULSE PROCESSING AND TIMING SYSTEMS
297
Vert -Deft
Coils in Yoke
From
Vert-Sawtooth
Gen
Fig. 7 -20. Basic tube -type vertical -deflection circuit.
Two linearity controls are shown in the damper -tube circuit, one in the
grid circuit and one in the cathode circuit. These control the phase and
extent of conduction of the damper tube, and affect the linearity of the
left side of the picture. The horizontal -centering control adjusts the
amount of an externally applied direct current through the deflection coil,
serving to center the sweep. Another pickup -head control, in the grid
circuit of the driver tube, is marked "H Peak." This is a feedback adjustment to prevent the driver tube from conducting too soon, which would
upset the proper transition of current from the damper tube to the driver.
A vertical -deflection circuit is shown in the diagram of Fig. 7 -20. This
circuit operates on the same principle as that described above, but at a
much lower frequency (60 Hz as compared to 15,750 Hz) ; therefore,
details of the circuit are somewhat different. The coil of the vertical deflection yoke is largely resistive at 60 Hz, and no large surge of voltage
occurs during retrace. Thus, there is no ring, or oscillation, to be damped,
and no damper tube is employed.
Feedback Linearity Correction
Modern deflection circuitry, whether tube type or solid state, employs
some form of automatic linearity control of the sawtooth wave for both
horizontal and vertical deflection. The basic form is shown in Fig. 7 -21.
We will analyze this circuit, which, with minor variations, is rather
common in all types of cameras.
In this specific circuit, VI operates between cutoff and conduction. Upon
the arrival of the horizontal -drive pulse at the grid, the tube conducts,
rapidly charging the saw -forming capacitor to the supply voltage (waveform 2) . At the end of the pulse, the tube is cut off, and the capacitor
discharges in an exponential manner. During the on period, a portion of
the voltage is developed across R1, and this action adds a pulse to the
sawtooth waveshape (waveform 2) . Linearity control R1 allows variation
of the amplitude of this pulse, providing vernier control of sweep
linearity.
298
TELEVISION BROADCASTING CAMERA CHAINS
Because the load presented to the plate of V3 by the horizontal -yoke
windings and output transformer TI is largely reactive at the horizontal
frequency, the voltage waveform at the V3 plate resembles a pulse more
than a sawtooth (waveform 3) This is necessary to drive a sawtooth
current waveform through the yoke. A sampling resistance of small value
(R2) in series with the deflection coils produces a voltage waveform that
is identical to the yoke -current waveform ( waveform 4) This provides
a source of negative feedback voltage to the comparison amplifier (V2)
and to the saw- former capacitor.
.
.
Arriz Drive
sawtootn
Voltage
-
-- Resistive
2
_ _ Component
Voltage Pulse
at
11-
Vi Putte
Sawtomn
Current in
nerator
Output
Saw C
3
d
Sampling
Resistor
Hari:-Output
Comparison
Nair
Amp
Output
V2
Transfo nier
Horiz-0tll Coils
n
V3
Horiz -Drive
Pulse
Negative Feeeoack
Hone Size
Fig. 7-21. Automatic linearity control.
The comparison amplifier compares the original sawtooth waveform at
the grid with the feedback signal at the cathode. Any deviation from a
truly sawtooth current through the yoke causes a correction voltage to
drive the V2 cathode in such a direction as to compensate for the nonlinearity of the original sawtooth applied to the grid. In addition, since
the original capacitor charging curve is exponential, capacitor Cl couples
the error, or feedback, voltage to the saw- former "charging" capacitor
( through resistor R3) with such polarity that it tends to make the exponential curve more linear.
NOTE: The same horizontal- and vertical -drive pulses that energize the
deflection circuitry are employed to generate target-blanking pulses.
These blanking pulses normally are produced by conventional multi vibrator circuits.
299
PULSE PROCESSING AND TIMING SYSTEMS
7 -5.
PICKUP-TUBE PROTECTION
Failure of either the horizontal- or vertical-deflection circuitry would
limit pickup -tube scanning in the respective dimension to a very narrow
path (single horizontal or vertical line) with resultant burning of the
overbombarded target area. Thus all cameras employ some form of pickup tube protection. Regardless of the type of circuit used, the basic idea is
common to all: a dc voltage related to the deflection voltage waveforms
is used to activate or deactivate the beam current in the pickup tube.
See Fig. 7 -22A. Transistors Q1 and Q2 sample the outputs of the vertical- and horizontal -deflection yokes respectively, and control the voltage
appearing at the pickup -tube cathode. The biasing arrangement shown
causes conduction of both transistors in the absence of a base signal. Under
this condition ( transistors saturated) , the common collector voltage goes
to the emitter voltage of +20 volts, cutting off the pickup -tube beam.
+20V
Horiz Coils
Vert
Coils
-10 V
-10
V
Pickup-Tube Cathode
Normal -10 V Loss of Sweep: +20
To
V
(A) Solid state.
K1
RC
Filter
Horiz Output
0
u
Relay Coil
2
i °1
y
Cathode of
Pickup Tube
From Vert Output
RC
Filter
(B) Tube type.
Fig. 7 -22. Protection circuits for pickup tube.
300
TELEVISION BROADCASTING CAMERA CHAINS
When both deflection signals are present, both transistors cut off and the
common collector voltage goes to -10 volts. Since this point is connected
to the pickup-tube cathode, beam current can exist.
In the absence of either deflection signal, the respective transistor conducts (saturates), and the common collector goes from -10 to +20 volts,
again cutting off beam current. This specific circuit treats retrace or flyback
duration as absence of deflection signals and triggers the beam to cutoff,
thus providing blanking. In this case, blanking and pickup-tube protection are provided by the same circuit.
A common tube -type protection circuit is shown in Fig. 7 -22B. Filtered
vertical-circuit pulses feed the control grid, and filtered horizontal- circuit
pulses feed the suppressor grid. A hold -off resistor (R1) applies a positive
Dark Area
Shading -Correction Waveform,
Horiz Rate
(A) Dark
area at left.
Dark Area
OIM
Shading -Correction Waveform,
Horiz Rate
(B) Dark area at right.
Fig. 7 -23.
Waveforms
PULSE PROCESSING AND TIMING SYSTEMS
301
Dark Area
Shading- Correction Waveform,
Vert Rate
(C) Dark area at top.
Dark Area
Shading- Correction Waveform,
Vert Rate
(D) Dark area at bottom.
Dark Area
Shading -Correction Waveform,
Horiz Rate (Parabola)
(E) Dark area in center.
for shading correction.
TELEVISION BROADCASTING CAMERA CHAINS
302
potential to the cathode such that the tube does not conduct in the
absence of either the horizontal- or vertical -deflection waveform. Under
this condition, the relay coil (K1) in the plate circuit is not energized,
and relay contacts 1 and 3 are closed. This applies a positive potential to
the pickup -tube cathode, preventing beam current. When both deflection
voltages are present, the tube conducts, energizing relay K1. This closes
contacts 1 and 2, returning the pickup -tube cathode to ground.
+12 5V
12. 5 V
0--4w-..
POS.
VOLTAGE
DECOUPLER
0 -26
105V
EMITTER
FOLLOWER
O
HORIZ SAW
CONT
Ó_y
HORIZ
TRAPEZOID
-7
PHASE
SLITTER
0 -6
GEN.
0 -5
EMITTER
FOLLOWER
0-8
NEG
HOR K
CLAMP PULSE'
HORIZ
SAWTOOTH
HORIZ PARAR
GEN
O
CONI
-3
EMIT TER
FOL LOWER
0 -4
INTEGRATOR
( PARAR I
0
-I/0-2
EMITTER
FOLLOWER
0-20
VERT SAW
CONT
Z
VERT
SAW TOOTH
PHASE
SLITTER
0 -19
GEN
0
18
EMITTER
FOLLOWER
NEG VERT
DR
0 -21
OR
VERT
VERT PARAR.
CONT
D
SAW TOOTH
GEN
EMITTER
FOLLOWER
0 -17
0 -16
-II
INTEGRATOR
(PAR AB .1
0- 14/0 -15
10
2V
-12 5V
IO
5V
12 5v
Fig. 7 -24. Shading generator
303
PULSE PROCESSING AND TIMING SYSTEMS
7 -6. SHADING -SIGNAL FORMATION
In spite of the coarse and fine adjustments provided for pickup tubes,
shading sometimes occurs at various areas of the raster, as illustrated in
Fig. 7 -23. Correction of a dark area at left (Fig. 7 -23A) or right (Fig.
7 -23B) of raster requires a horizontal -rate sawtooth of opposite polarity.
A dark area at the top (Fig. 7 -23C) or bottom (Fig. 7 -23D) requires
A/B
GREEN
M
OUTPUT
0- 11/012
u
LO
1/2
/111111111SS
iSiis
HORIZ
$AW
HI
F/H
REO
OUTPUT
LO
6/1
0- 9/0-10
111111HORa
S0Sfs0
VERT
.
SAW
BLUE
OUTPUT
0- 22/0-23
T/U
LO
O 16/IT
Ï"1..
>ts111111
VERT PARAB
Hi
O
W/X
MONO
OUTPUT
0- 20/0 -25
LO
0 19/20
Courtesy RCA
for RCA TK -27 color camera.
TELEVISION BROADCASTING CAMERA CHAINS
304
opposite polarity. A vertical dark area in the
middle of the raster (Fig. 7 -23E) requires a parabolic waveform at the
horizontal rate. If the dark areas are at both sides, a parabolic waveform
of polarity opposite to that shown would be required. A horizontal dark
area in the middle of the raster requires vertical-rate parabolic pulses.
A block diagram of one model of shading generator is illustrated in
Fig. 7 -24. The functional description of this unit is as follows:
The horizontal -trapezoid generator develops a sawtooth waveform, during active scan, that rides on a pulse that occurs during camera blanking.
The amplitude and polarity of this waveform are proportioal to the incoming horizontal -saw dc control voltage. The pulse is added to the saw tooth to correct for nonlinearities of the output transformer.
The horizontal -sawtooth generator produces a sawtooth waveform at a
horizontal rate. The amplitude and polarity of the waveform are proportional to the horizontal -parabola dc control voltage.
a vertical -rate sawtooth of
Feedback Pair
Parabola
Emitter
Follower
Phase
Splitter
Q1
Outputs
Sawtooth In
Cl
Integrator
Gdì
Capacitor
Fig. 7 -25. Circuit for generating parabolic waveform.
The vertical- sawtooth generators provide sawtooth waveforms at a vertical rate. The amplitude and polarity of the sawtooth are controlled in one
case by the vertical-saw dc control voltage, and in the other by the vertical parabola dc control voltage.
The phase splitters provide equal positive -going and negative -going
waveforms, which are fed to potentiometers that allow control of the
amplitude and polarity.
The integrators (parabolic) integrate a sawtooth input to produce a
parabolic waveform. They provide equal positive and negative outputs to
be fed to potentiometers that allow control of the amplitude and polarity.
The four output amplifiers provide a low input impedance for signal
mixing. Also, they amplify the mixed signals and provide a low driving
impedance to the isolation transformer.
PULSE PROCESSING AND TIMING SYSTEMS
305
Original Video (Hotu Ratel
Unkeyed Vertical Shade
To Be Added to
Video
Original Video With
Vert Saw Superimposed
Original Video Restored ny
Clamp Action in Processing Amp
Fig. 7 -26. Vertica -shading waveforms without keying signal.
The voltage decouplers isolate the dc supply voltages from the rack
wiring and other modules.
We have already studied the formation of a sawtooth waveform from
pulses. It remains to see how the parabolic waveform is shaped. See Fig.
7 -25. A sawtooth signal from emitter follower Q1 is fed into the feedback
pair Q2 -Q3. This waveform is integrated through the action of capacitor
Cl in the feedback path. Then the waveform across C2 is the integral
of the sawtooth and is therefore, parabolic in shape. Equal loads are
placed in the emitter and collector circuits of output transistor Q3 to ob-
Original Video IHoriz Ratel
Vert Shading With Keying Signal
Original Video With Added Keying
Signal Varying at Vert Rate
(Shading is retained after clamping.)
Fig. 7 -27.
Vertical- shading waveforms with keying signal.
TELEVISION BROADCASTING CAMERA CHAINS
306
tain equal and opposite parabolic waveforms for the parabola control potentiometers (Fig. 7 -24) .
The vertical component superimposed on the video of any signal in the
camera will be "clamped out" in the following processing amplifier ( see
Fig. 7 -26) It is necessary therefore to key the vertical- shading signal at
a horizontal rate, with the tips of the keying pulses returned to a fixed
potential. Since the processing amplifier clamps to the tips of horizontal
blanking, the vertical component of the signal will be restored as shown
in Fig. 7 -27.
.
Video Output
of Pickup TuDe
Pr ea mp
L
Mixed Horiz & Vert
Shading Innut
-1
I
Switch
Cl
Q1
-1.--I
Clamping Pulse
IHoriz Ratel
L
Vert Rate
-
S
./
rrated Shading
Holes' provide hori ontal- clamping points to axis.
Fig. 7 -28. Circuit for keying vertical- shading signal.
Fig. 7 -28 shows a typical keying circuit for this purpose. A negative
clamp pulse on the base of clamp- reference transistor Q1 charges Cl to
the emitter potential (ground in this example) of Q1. With the tips of
the pulse clamped at this potential, Q1 cuts off following the negative
pulse and is held cut off until the next pulse appears. At cutoff, Q1 is an
open circuit and has no effect on the shading signal, but during each
pulse Q1 is saturated and switches the signal to ground.
In this circuit, the only collector -to- emitter supply voltage is the shading
signal, and the only signal appearing at the collector is the serrated shading
signal. When there is a zero shading component, there is also zero output
at the Q1 collector. In this circuit, the collector signal is coupled to the
preamplifier and added to the pickup -tube video signal.
EXERCISES
Q7 -1.
Q7 -2.
Why is the horizontal -drive width made approximately half the horizontal- blanking width?
Do all boxcar circuits narrow the input pulse?
PULSE PROCESSING AND TIMING SYSTEMS
307
How do you know what output -pulse width to expect in a typical
boxcar circuit?
Q7 -4. If the input pulse to a pnp boxcar is positive, is the boxcar output
delayed or undelayed?
Q7 -5. If the input pulse to an npn boxcar is negative, is the boxcar output
delayed or undelayed?
Q7 -6. Can you change Plumbicon sensitivity by target adjustment?
Q7 -7. In the automatic mode of operation, are the iris control and black
and white level controls on the control panel still operable?
Q7 -8. Do automatic level controls affect levels in manual operation?
Q7 -9. What are the variables in color- camera balance?
Q7 -10. What must be rechecked when the I.O. target voltage is readjusted?
Q7 -11. What must be rechecked when the vidicon target voltage is readjusted?
Q7 -3.
CHAPTER
8
Cancera Control and
Setup Circuitry
This chapter covers the remote operating camera -control panel and
associated setup controls, which may be physically located at the camera
head, on the control panel, or in rack-mounted gear. We also will discuss
the usual types of waveform monitors associated with setup and control.
8 -1.
THE RCA TK -60 MONOCHROME CAMERA CHAIN
Obviously, the simplest form of camera chain is the single -tube monochrome type. We are justified in covering such a camera chain not only
because there are many still in use, but also because the color camera is
exactly the same, except for the added signal channels and additional associated controls. These will be covered in their proper place later in this
chapter.
Fig. 8 -1 illustrates the entire TK -60 chain with the exception of the
rack -mounted processor unit and the console-mounted picture and waveform monitor. Fig. 8 -2 illustrates the normal hookup of the complete
installation.
Iris control is achieved by a transistor servo amplifier system. The irises
of all four lenses are adjusted simultaneously from either the rear of
the camera or the remote -control position. In this manner, the iris is
always preset as lenses are changed.
Voltage regulation for the image- orthicon tube is achieved by using
corona discharge tubes. Where utmost precision is desirable, these are enclosed in a temperature- controlled oven. Other regulating devices that are
utilized include zener diodes and Victoreens.
The nuvistor type of triode tube is used exclusively in the video preamplifier, and in a number of other circuits associated with blanking and
deflection. The signal -to -noise performance and freedom from micro phonics of this tube make it particularly useful in preamplifier service.
308
CAMERA CONTROL AND SETUP CIRCUITRY
309
(A) Rear view of camera.
(B) Interior of camera.
(C) Operating control panel.
Cajrtesy RCA
Fig. 8 -1. RCA TK -60 monochrome camera.
TELEVISION BROADCASTING CAMERA CHAINS
310
This system includes two separate intercom circuits. These may be interconnected or operated independently. A transistor amplifier and volume
control are provided at each point where a headset is plugged into the
system.
A 41/2 -inch I.O. tube is used in the camera. Circuits in the processor
provide aperture response to produce 100 -percent response at 400 TV
lines, phase corrected. These circuits also provide up to 13 dB of aperture
correction peaked at 8.0 MHz, with continuously variable amplitude adjustment to compensate for tolerances in pickup tubes.
Video equalization to compensate for different lengths of camera cable
is made available on a tap switch. Positions on the switch correspond to
100 -foot increments of cable length, up to a maximum of 1000 feet.
A completely solid -state power supply, Type WP -16B, provides all
regulated voltages for the complete TK -60 chain. It occupies 7 inches of
rack space.
Either of two modes of operation can be selected. In a clamp -on-black
mode, exposure is governed by the iris control, and the black-to -white
level (gray scale) is dependent on settings of the contrast and brightness
controls. This is similar to the operating techniques used with standard
3 -inch image- orthicon cameras. In a clamp -on -white mode, brightness becomes a setup adjustment to establish the desired peak white level. The iris
control is adjusted to maintain white level of the scene at the desired setting over the knee of the I.O. characteristic. The contrast control is used
to compensate for reflectance variations in the darkest areas of the scene.
Use of the remote -control panel permits a single video operator to
handle as many as six TK -60 cameras. In normal operation, only two
controls -IRIS (exposure) and CONTRAST -may require attention. For
convenience, a brightness setup control, remote lens -cap switch and tally,
on -air tally, intercom phone ack, and headset-level control are included.
TK-60 Processor
50' Camera Cable
TM 6 Master
o
0000
o
óó000ó
0000
o o0ó
&a
Monitor
o0
Remote
Control Panel
0
Console Housing
F
TK-60 Camera
WP -16 Power Supply
Interconnecting Cables
Courtesy RCA
Fig. 8 -2. TK -60 monochrome camera chain.
CAMERA CONTROL AND SETUP CIRCUITRY
311
The rear panel of the camera (Fig. 8 -1A) contains the main operating
controls and jacks for the cameraman, as follows:
Lens cap
Orbitor on -off switch
Viewfinder selector switch: effects preview, camera, or effects (only)
Normal and reverse vertical deflection
Engineering phone and cue volume controls
Engineering phone and cue jacks
The items listed above are all on the left side of the rear panel. The
following controls and jacks are on the right side:
Viewfinder contrast, peaking, and brightness controls
Iris remote and local switch
Iris control
Normal and reverse horizontal -deflection switch
Production phone and cue volume controls
Production phone and cue jacks
In the center is the turret -control handle for a four -position lens complement. Immediately above this handle is a six -position filter control to
select any of six different neutral- density filters. Thus, if a given depth
of field must be retained under varying lighting conditions (dependent
on f stop of iris) , the exposure for the I.O. can be controlled by this knob.
The setup control panel (lower right corner inside camera in Fig. 9 -1B)
contains the following controls:
Orth focus
Wall focus
Switch with align, calibrate, set, operate, and B+ off positions
Target voltage
Beam current
Viewfinder focus
Viewfinder contrast
Viewfinder brightness
Beam align No. 1 and No. 2 controls
Video gain
Multiplier focus
Image focus
Once the camera is properly set up, the above controls require no further
adjustment under normal operations.
There is only one part of the setup procedure for this particular camera
that differs from the steps followed for other cameras: the adjustment of
the VIDEO GAIN control on the camera preset panel. The RCA procedure
is as follows: Set the selector switch on the preset control panel to the
"cal" position. Block off the scene during the white portion of the calibration pulse by inserting the edge of the neutral- density filter disc into the
TELEVISION BROADCASTING CAMERA CHAINS
312
optical path, or, preferably, by placing a black material of very low reflectance (velvet or equivalent) over the proper portion of the scene. Adjust the VIDEO GAIN control for a match of the video white to the white of
the calibration pulse. This adjustment will be more accurate if the viewfinder brightness level is set near the kinescope cutoff point. Return the
viewfinder brightness to its normal level. Remove the edge of the filter
disc or other obstruction from the optical path, and set the selector switch
to the "oper" position.
The normal operating procedure for this camera as given by RCA is
presented below. Comparable controls in all cameras normally are adjusted by this basic procedure.
Warm -Up
The image orthicon is the determining factor in the camera -chain
warm -up time for on -air operation. It is recommended that approximately
20 minutes be allowed for the image orthicon to reach optimum operating
temperature. The camera circuitry is sufficiently stablized after 2 minutes
for of -air operation, but the image orthicon will be slightly sticky until
it reaches operating temperature. However, the orbiter permits immediate
use without serious burn -in.
Lighting
The TK -60 camera is capable of handling wide light -level ranges
through the combined action of the IRIS control and neutral- density filters
on the filter disc. Except for special effects, the contrast ratio of a specific
scene should not be more than 20 to 1. Contrast ratio is best controlled by
flat -lighting the scene, which also can reduce undesirable shading effects
caused by shadows and unbalanced lighting conditions. It will become
more evident from experience that the application of proper lighting techniques contributes substantially to the ease with which camera -chain
operation is carried out.
Modes of Operation
A clamp -on -black or clamp -on -white mode of operation is possible for
the TK -60 camera chain. Either mode may be selected by a jumper arrangement in the camera and processing amplifier. The following paragraphs further describe the two modes relative to actual camera -chain
operation and the specific controls involved.
the clamp -on -black mode, changing either the iris
Clamp -on- Black
the
setting or
brightness or contrast levels causes a change in all video
levels (gray to white) with respect to black. Therefore, depending on
the lighting and special effects desired, the proper combination of control adjustments must be made to maintain a constant output level. The
following information may be used as a guide for the control procedures
required to accommodate various scene conditions.
-In
CAMERA CONTROL AND SETUP CIRCUITRY
313
The IRIS control must be adjusted to maintain scene whites at the
proper setting over the knee of the image- orthicon characteristic. This
setting is therefore a function of scene white level. During operation, the
iris -delegate switch on the camera is in the remote position, since iris
control is one of the main control operations at the camera-control position ( remote-control panel) .
If scene whites are maintained at a constant level, the remote CONTRAST
and BRIGHTNESS controls are used to make corrections for contrast -ratio
changes. The combined operation of these controls sets the dark areas of
a scene to the desired level while maintaining a 0.7 -volt output level.
When both the scene white level and the contrast ratio changes, the
CONTRAST, BRIGHTNESS, and IRIS controls must be adjusted to maintain
the desired black level, output level, and operation of the image orthicon
relative to the knee of its characteristics.
Special -Effects Operation Combinations of the control settings described above may be used to obtain the special effects possible as the result of black clipping or operating the iris below the knee of the imageorthicon characteristic. When the foregoing abnormal settings are to be
employed, the normal settings should be recorded to expedite a return to
the original operating condition.
NOTE: Under conditions of extreme contrast, such as outdoor sunlight
to shade, the gamma -corrector circuit will allow the system to compensate at the sacrifice of true gray -scale rendition. A 0.7 -gamma position is
provided in the processor for this purpose.
Clamp -on- White -In the clamp -on -white mode of operation, the video
white level remains fixed (clamped) relative to black, for scene contrast
changes or contrast -control variations, as long as the iris sets the scene
whites over the knee of the image- orthicon characteristic. Therefore, a
constant peak-to -peak output is maintained without the use of the brightness control. The constant output level facilitates operational simplicity
for most operating conditions. This will be evident from the following
control-setting procedures given for possible scene conditions.
The IRIS control must be adjusted to keep scene whites at the proper
setting over the knee of the image- orthicon characteristic. If scene contrast ratio is maintained at the constant level, the IRIS control is then the
only operating control necessary.
The CONTRAST control serves to maintain the effective contrast ratio
by compensating for reflectance variations in the darkest areas of a scene.
Therefore, if scene white level remains fixed, the CONTRAST control is
the only control required.
The use of both the CONTRAST and IRIS controls is necessary when scene
white -level and contrast -ratio changes occur simultaneously.
Under normal operating conditions, the BRIGHTNESS control is not
used. However, it may be required in some instances as described for
314
TELEVISION BROADCASTING CAMERA CHAINS
special effects. If this control is so employed, the normal setting should be
recorded to expedite a return to the original operating condition.
A combination of the foregoing control settings may be employed to
obtain special effects such as those resulting from black clipping or operation of the iris below the knee of the image-orthicon characteristic. If
the iris is set below the knee, the peak -to -peak output level changes,
and this may be compensated for by resetting the brightness level. The
note under "Special- Effects Operation" for the clamp -on -black mode of
operation also applies to the clamp -on-white mode.
Target Voltage Effect on Operation
With the low target voltage (2.3 volts above cutoff) , the knee of the
image-orthicon characteristic is rounded. This condition makes operation
over the knee less critical and permits the iris to be opened further to
lift scene blacks with relatively small loss of white detail. The result of
this condition is an improved signal -to -noise ratio. Other advantages of
low- target -voltage operation are extended tube life, reduced target flicker,
and reduced microphonics.
With a high target voltage (3.0 volts above cutoff) , the knee is sharp,
and all scene contents are below the knee while the scene whites are at
the knee.
Stopping over the knee can cause a loss of white detail, resulting in
a chalky appearance. With the proper iris setting, a good gray -scale rendition is possible (operation is more linear)
.
Camera- Position Operating Controls
The optical -focus and lens -selection controls are the only operational
controls required for on -air operation that are located at the camera.
For normal operation, the IRIS control is switched to the remote position.
Local operation of this control serves mainly to facilitate camera -setup
procedures.
The neutral- density-filter disc normally is kept in the open position so
that the lens may be stopped down for greatest depth of field and the light
requirements can be kept to a minimum. In bright sunlight, if the iris
cannot control the high lights below the knee of the image- orthicon
transfer characteristic, neutral- density filters must be inserted in the light
path. In cases where the director calls for a given depth of field, the iris
must be set, and the high lights must be attenuated to the knee with
neutral-density filters.
The orbiter switch normally is placed in the immobilize position. As
the image orthicon approaches the end of its useful life, it may develop
spots or burns on the dynode. With the switch in the immobilize position, these blemishes will appear to orbit as the picture remains stationary.
If this becomes distracting, the orbit position should be used. This causes
the blemishes to remain fixed while the picture orbits so slightly as to be
CAMERA CONTROL AND SETUP CIRCUITRY
315
unnoticeable. The orbit position cannot be used when absolutely stationary
images are required, as in a superimposition or centering a title card.
By use of the deflection-reversal controls (on the camera rear control
panels), the image- orthicon horizontal and vertical scans may be reversed
for certain special-effects applications.
The LENS CAP switch tends to place the camera in a standby condition by capping the lens electronically and inserting a reduced-amplitude
calibration pulse to indicate proper operation.
The viewfinder input is obtained from the output of the processor and
is therefore identical to the signal going out on the line. Viewfinder peaking is available as an aid in obtaining a rapid, well- defined focus point
without overshooting the correct setting.
The viewfinder picture -selector switch (on the camera rear control
panel) permits the cameraman to view his picture as it is inserted or
superimposed with other signals coming from the special -effects switcher.
When no effects switcher is programmed, the viewfinder picture -selector
switch is kept in the camera position, and the cameraman always sees the
output of his processor and may set the viewfinder brightness and contrast levels to his individual preference.
Anytime the cameraman desires to see the output of the effects
switcher, such as when he is preparing his camera for a superimposing
operation with another camera, he may switch to effects preview. Then he
can see the output of the effects switcher, regardless of how his camera
may be switched at the time.
When the viewfinder is switched to effects, a relay ( energized by the
effects tally circuit) switches the viewfinder input to the effects switcher
only when the camera is punched up on the effects switcher. In this manner, the camera pickup is always on the viewfinder screen together with
any other picture superimposed on it by the effects switcher. The camera
picture is not disturbed by the effects switcher if it is not one of those
being superimposed.
The overscan switch is located on the camera deflection chassis. This
switch may be used to prevent raster burn -in when the camera is left on
for long periods of time without capping the lens.
The foregoing description has been presented to emphasize the interdependence of controls assigned to the camera and the camera-control
operator.
8 -2.
THE MARCONI MARK VII FOUR -PLUMBICON
COLOR CAMERA
Fig. 8 -3A illustrates the rear control panel of the Mark VII camera.
Fig. 8 -3B shows the camera -control position as installed at WBBM -TV,
and Fig. 8 -3C shows the monitor -alignment rack installation at the same
station.
316
TELEVISION BROADCASTING CAMERA CHAINS
Courtesy Ampex Corporation
(A) Rear view of camera.
Courtesy WBBM -TV
Courtesy WBBM -TV
(B) Camera -control console.
(C) Monitor/ alignment rack.
Fig. 8 -3. Installation using Marconi Mork
VII color camera.
CAMERA CONTROL AND SETUP CIRCUITRY
317
Individual color-channel gain controls normally are not required to be
adjusted during a telecast. However, some organizations prefer to have
the controls available at the operational position (as in Fig. 8 -3B) for
correction of color errors in the scene. One example is when light reflected from some colored object in the set falls on the face of an artist.
The block diagrams in Fig. 8 -4 show the sequence of the main video
signal -processing functions in the camera (Fig. 8 -4A) and the camera
control unit, or CCU (Fig. 8 -4B) . The positions in the chain of the
major operational and preset video controls, test- signal injection points,
and monitoring and bridging points are shown also. The camera has four
separate video chains; the block diagram is that of the luminance channel.
The differences between the video circuits for the luminance and color
channels are shown by dash lines in Fig. 8 -4B.
Refer to Fig. 8 -4A. The signal current from the pickup tube is amplified by the head amplifier, which is mounted directly on the deflection
yoke. The output of the head amplifier is fed to another amplifier, which
includes a preset gain control for setting the sensitivity of the channel.
Each of the following two stages contains a special remote -gain -control
arrangement by means of which the gain of the unit can be controlled from
the CCU. The first of these is for the main operational control of gain by
means of the master gain control, which is connected to the corresponding
unit in all four channels. Operation of the master gain control thus
simultaneously varies the gain in the four channels. The second remote
gain control is for individual control of the channel gain.
The variation of gain within the remote-gain -control stage is by means
of a photosensitive resistor that forms part of a video attenuator. Current
for the lamp associated with this resistor is derived from a dc control signal originating from the remotely located gain -control potentiometer.
Because the photosensitive resistor is also extremely temperature sensitive, a technique of dc feedback stabilization is employed to ensure stable
and linear control of video gains. In the case of the master gain control,
this also ensures that the same gain variation is produced in all four
channels.
The second remote -gain -control stage is followed by a first clamping
stage, which establishes the signal dc level necessary for proper operation
of the limiter stage that follows. The clamper also removes spurious low frequency components of the video signal that might otherwise cause
overloading or intermodulation in later stages. The limiter stage also
serves to protect later stages from signal components of excessive amplitude. Such a limiter is especially necessary when Plumbicon tubes are
employed, since this type of tube does not have the inherent self -limiting
characteristic of the image orthicon and therefore can produce large amplitude signals from scene high lights.
The final video stage in the camera is an amplifier that feeds the 75 -ohm
video coaxial section in the camera cable. Note that the operational control
318
TELEVISION BROADCASTING CAMERA CHAINS
E
a
(A)
Camera.
Fig. 8 -4. Video processing
19
CAMERA CONTROL AND SETUP CIRCUITRY
as To
L
Cl
r
âá
_rÑ
c
H
a
kna
É
3
3
E
ca o
(B) CCU.
Courtesy Ampex Corporation
in Mark VII system.
320
TELEVISION BROADCASTING CAMERA CHAINS
of gain occurs at an early point in the video chain. This arrangement has
the advantage that all the following stages operate at a substantially constant signal level, and no compromise is necessary in the choice between
a high level to override spurious signals and a low level to avoid overloading. Thus, the signals fed from the camera to the CCU are at a
constant level of 0.7 volt peak to peak.
The first of several waveform monitoring points is at the CCU video input from the camera cable (Fig. 8 -4B) . Signals from these monitoring
points may be fed to the external waveform monitor under control of the
waveform -monitor push- button selector switch on the CCU control panel.
Following the monitoring point is the changeover arrangement for injection of test signals. This enables the performance of the CCU to be
checked independently of that of the camera.
The first CCU stage is the camera -cable corrector, which corrects for
the frequency -amplitude characteristic of the cable, and is adjustable to
suit the length of cable in use. This is followed by the aperture corrector,
which gives an adjustable and substantially phaseless accentuation of the
higher video frequencies. (Such correction is necessary to compensate for
the finite scanning aperture of the pickup tube.) This stage is not included in the color channels, since the aperture loss within the narrower
chrominance band is negligible.
The next block is an amplifier that includes a preset gain control. The
purpose of this stage is to provide the correct signal level for the linearizer
stage. The linearizer is a nonlinear amplifier that can be adjusted to have
an amplitude transfer characteristic complementary to that of the pickup
tube; thus, the overall transfer characteristic can be made linear. This
facility is included for two reasons. The first is the need to provide the
linear camera-channel output that is necessary for certain proposed methods
of processing four-tube camera signals for transmission. The linear signal
is available at the special output socket, for which an output amplifier of
75 -ohm impedance is provided. The second reason for including the
linearizer is that the linear signal makes it possible to use a particularly
convenient type of gamma corrector (described later) .
The next two circuit blocks are included only in the luminance channel. First is a negative -picture amplifier that provides the facility of reversing picture polarity; it is intended for use in black-and -white operation of the camera. The negative -picture amplifier is followed by a remote- gain -control stage of the type already described. The purpose of this
stage is to vary the video gain as the master black control is adjusted, so
as to maintain the total video signal excursion constant. This is achieved
by deriving the dc control signal from the master black control potentiometer. The remote -gain -control stage is required only in the luminance
channel because the master black control adjusts the black level of the
luminance channel only. Because of the small range of black -level adjustment that is required, and the dominating influence of the black level
!°1
CAMERA CONTROL AND SETUP CIRCUITRY
321
of the luminance signal in the reproduced picture, it is not necessary to
vary the black level of the color channels simultaneously.
The next amplifier stage in the chain has a preset gain control for setting the correct video level for the following stages. This amplifier drives
the peak white clipper, which includes the main clamp and serves the
usual purpose of preventing the signal excursions from exceeding peak
video level. This is followed by the blanking mixer, in which the correct
black level is established and the system standard blanking is introduced.
The final processing stage is the gamma corrector, a nonlinear amplifier
that gives the outgoing video signal the desired gamma characteristic.
The special circuit arrangement employed permits the gamma exponent
to be varied continuously by means of a single control while the peak video
signal amplitude remains constant. For greater precision, the nonlinear
characteristic is approximated by four linear segments. A switch on the
CCU control panel provides a choice between two different gamma characteristics. A preset control is provided for each characteristic, one having
a gamma range of 0.4 to 0.6 and the other a range of 0.6 to 1.0.
The output amplifiers in each of the four channels provide three outputs at standard level into a 75 -ohm impedance.
8 -3. THE RCA
TK -44A CAMERA CONTROL
Fig. 8 -5A illustrates the RCA TK -44A camera -control unit with the
normal operating controls exposed and the cover for the setup -panel
adjustments closed. Fig. 8 -5B shows the setup -panel cover open to expose
the setup controls. The TK -44A camera is a three -tube design employing lead -oxide photoconductive tubes and contour -enhancement circuits (Chapter 6)
While there are considerable differences in physical location and (sometimes) nomenclature of controls in various camera systems, all are the
same in general function. The following description serves as emphasis
and review of adjustments and operation.
.
Setup Controls
Setup controls are primarily concerned with operating parameters of the
pickup tube or tubes. The image orthicon requires the greatest number of
adjustments. (The following information is presented through the courtesy
of RCA.)
Proper Scanning (Horizontal and Vertical Controls) -Full -size scanning
of the target should always be used during operation. Full-size scanning
can be assured by first adjusting the deflection circuits to overscan the target
sufficiently to cause the edge of the target ring to be visible in the corners
of the picture, and then reducing the scanning until the edge of the target
ring just disappears. In this way, the maximum signal -to -noise ratio and
maximum resolution can be obtained. If the camera employs an orbiter, a
TELEVISION BROADCASTING CAMERA CHAINS
322
(A) Setup panel covered.
(B) Setup panel exposed.
Courtesy
Fig. 8 -5.
RCA TK -44A camera
control.
RCA
CAMERA CONTROL AND SETUP CIRCUITRY
323
scanning -size adjustment jig should be used to assure the proper size of the
scanned area.
(This paragraph does not apply to image orthicons employing electronically conducting glass targets [Chapter 4).) Underscanning the target,
i.e., scanning less than the proper area of the target, should never be permitted. Underscanning produces a larger- than -normal picture on the
monitor. If the target is underscanned for any length of time, a permanent
change in target -cutoff voltage of the underscanned area takes place, and
the underscanned area then is visible in the picture when full -size scanning
is used.
A mask having a diagonal or diameter of 1.8 inches ( 1.6 inches for
types) should be used in front of the photocathode to set limits
for the maximum size of scan, and to reduce the amount of light reaching
41/2 -inch
unused parts of the photocathode.
Alignment of Beam -Proper alignment of the beam in an image orthicon
is one of the most important steps in obtaining a good picture. Proper
alignment for a non -field -mesh image orthicon is obtained when the small
white dynode spot does not move when the beam -focus control (grid -4
voltage) is varied, but simply goes in and out of focus. For tubes that have
a field mesh, the alignment currents are adjusted so that picture response is
maximum and the center of the picture does not move when the beam-focus
control is varied, but simply goes in and out of focus. Auto -alignment devices are useful for determining the exact setting of alignment -coil current.
Setup Differences Between Field -Mesh and Non -Field -Mesh 1.0.'sThere are two families of image-orthicon camera tubes, one with field mesh
and the other without field mesh. Proper setup of any image orthicon is
best assured by observing the procedure outlined in the technical bulletin
for the individual type. This procedure, however, can be simplified greatly
by noting the three principal differences between the setup procedures for
non -field -mesh image orthicons and field -mesh image orthicons.
For non -field -mesh types, the technical bulletin explains that a ".. .
preliminary alignment adjustment can be easily made by adjusting the
alignment current to produce a maximum signal output when the tube is
focused on a test pattern. Final adjustment is achieved by regulating the
alignment -coil current so that the small white dynode spot appearing on
the monitor does not move when the beam -focus control (grid No. 4) is
varied, but simply goes in and out of focus."
In the case of field -mesh image orthicons, where no white dynode spot
is involved, correct adjustment of alignment -coil current is obtained by
regulating this current "... so that the center region of the picture does
not move when the beam -focus control (grid No. 4) is varied, but simply
goes in and out of focus." Note that this generally occurs at the point where
the center of the picture is brightest.
The second difference in setup procedures for the two types of image
orthicons lies in proper application of the dc operating voltages. In non-
324
TELEVISION BROADCASTING CAMERA CHAINS
field -mesh types, the dc voltages may be applied with the lens capped.
If the lens is capped, however, it should be uncapped momentarily while
the grid -1 voltage is adjusted to setting that provides a slight amount of
beam current.
Insofar as field -mesh orthicons are concerned, under no circumstances
should the lens be capped during application of dc operating voltages. The
lens always must be uncapped and the lens iris opened to allow light to
fall on the photocathode before application of the dc voltages.
The third difference to remember is that field -mesh types generally
operate properly at only one particular mode of focus of the scanning beam.
If a large, coarse -mesh background is evident either with the lens capped
or in the low lights of the scene, grid 4 probably is being operated at the
wrong mode of focus. For three- inch -diameter field -mesh image orthicons,
the proper mode of focus generally occurs within the range of 140 to 180
volts on grid 4.
Beam Current (Beam Control)-During alignment of the beam, and
also during operation of the tube, always keep the beam current as low as
possible to give the best picture quality and also to prevent excessive noise.
Lack of sufficient beam results in improper resolution of high lights.
Target Voltage (Target Control) -Focus the camera on a test pattern.
Then adjust the target voltage to the point at which a reproduction of the
test pattern is just discernible on the monitor. This value of target voltage
is known as the target -cutoff voltage. The target voltage should then be
increased to the recommended value above cutoff for standard lime -glass
target types and to the selected target voltage ( according to operating
needs) for electronically conducting glass target types. The beam -current
control should then be adjusted to give just sufficient beam current to discharge the high lights. The interrelationship among tube sensitivity, signal to -noise ratio, and resolution may be used to obtain optimum camera performance for different lighting conditions. The determining parameter is
target voltage. At high target voltages, signal -to -noise ratio is enhanced at
the expense of resolution. As the target voltage is reduced, this relationship
reverses. For a given telecasting session, it is practical and advisable to
maintain the target voltage at a single value because such voltage constrains video gain, gamma correction, and other adjustments in the camera
chain. Furthermore, it generally is advisable to employ the same target voltage value for all cameras telecasting a given scene.
The target -voltage control should not be used primarily as an operating
control to match pictures from two different cameras. Matching should be
accomplished first by individual adjustment of the lens -iris openings. Small
changes in target voltage then may be used to produce picture matching.
The target -control voltage calibration should be checked periodically to
assure that the target -voltage adjustment is correct.
Resolution (Image {Photocathode]- Focus, Orth {Beaml-Focus, and
Aperture- Correction Controls)- Adjust the lens to produce best optical
CAMERA CONTROL AND SETUP CIRCUITRY
325
focus, and adjust the voltages on the image-focus electrode, grid 6, and grid
4 to produce the sharpest picture.
A loss in resolution can be caused by operation of the tube at too high
a temperature. Sometimes a loss in resolution can be traced to a dirty lens
or dirty tube faceplate; both the lens and the faceplate should be cleaned
periodically. (Be sure not to scratch the optical surfaces.) Make sure the
camera and turret are closed to prevent light leakage, which will "wash
out" the picture.
The aperture- correction amplitude control generally can be adjusted to
obtain 100 -percent response ( relative to 100 TV lines) at 300 lines resolution for a 3 -inch I.O., or 400 lines for a 41/2-inch I.O. The limiting factor
is excessive noise in the picture (Chapter 6 ) .
Color-Camera Registration Controls -To superimpose exactly the three
or four images on the raster, the following controls are involved: horizontal
and vertical size, horizontal and vertical linearity, horizontal and vertical
centering, and skew. The registration test chart is employed to permit these
adjustments. For proper color balance, the gamma controls must be set
to match all channels to one another in light -to- signal transfer characteristics.
Wall Focus (41/2-Inch I.O. Only) -The wall focus is adjusted so that no
mesh pattern is observed in low -light areas of the picture.
Photoconductive Tubes-Setup controls for the vidicon and lead -oxide
tubes consist mainly of the target, beam, and alignment controls. The same
color- camera registration controls are necessary. These tubes do not have
a knee, except as may be provided in video -processing amplifiers.
Operating Controls
Operating adjustments consist largely of maintaining proper pickup tube exposure, and small adjustments of black -level and gain (paint-pot)
controls for color chains.
Proper Exposure (Lighting, Iris, and/or Neutral - Density Controls)Proper exposure of the image orthicon is required at all times for consistent
production of high -quality pictures. The most common error in lighting
and exposure control is to overexpose the image orthicon to "bring up"
information in the low lights of the scene. A much better picture can be
obtained by filling in the low -light areas of the scene with fill light rather
than by opening the lens and overexposing the image orthicon.
In general, as the light level incident on the image orthicon is increased
and the signal output reaches the knee of the light transfer characteristic,
picture quality is improved because of an increase in resolution, signal -tonoise ratio, and contrast range. Signal -to-noise ratio and contrast range are
directly proportional to the square root of the illumination on the faceplate
of the image orthicon, and they increase until the high lights reach the
knee of the light transfer characteristic. Any further increase in light level
does not materially improve the signal -to -noise ratio but does increase
TELEVISION BROADCASTING CAMERA CHAINS
326
resolution slightly. Operation of the tube with the high lights substantially
above the knee allows it to handle a wider contrast range, because the
whites are compressed without loss of detail and the blacks are raised out
of the noise.
Remove the lens cap and focus the camera on a neutral ( black -andwhite) test pattern consisting of progressive tonal steps from black to
white. Open the lens iris just to the point at which the high lights (highest
step) of the test pattern do not rise as fast as the low lights (lower steps)
when viewed on a video -waveform oscilloscope. This operating point is
the knee of the light transfer characteristic.
For black- and -white operation, the camera lens then should be opened
approximately one to two stops above the knee for each individual scene.
This operating point assures maximum signal, good gray scale, freedom
from black borders, the sharpest picture, and the most natural appearance
of televised subjects or scenes.
The camera lens should be adjusted continuously to maintain this
operating point as the illumination in each scene changes. Operation at
this point is especially important for studio pickup in order to obtain the
best gray scale in the picture and to reduce the possibility of image retention.
For outdoor and other scenes in which a wide range of illumination may
be encountered, the camera should be panned across the scene that has the
least amount of illumination, and the lens iris should be adjusted so that
the high lights in that area are just above the knee. The camera then will
be able to handle all scenes having higher illumination without requiring
lens -stop adjustments. When the camera is to be shifted rapidly from a
scene of low brightness to a scene of high brightness, or vice versa, as may
take place during panning, the camera always should be set for the dark
scene.
For color cameras, the lens setting usually is adjusted so that operation
takes place with the high lights just barely over the knee. The lens setting
should be adjusted continuously as the scene changes so that the exposure
will not result in operation substantially above the knee, which causes color
dilution and contamination.
Table 8 -1. Illumination of Outdoor Scenes
Lighting Conditions
Direct Sunlight
Full Daylight*
Overcast Day
Very Dark Day
Twilight
Deep Twilight
*Not Direct Sunlight
Scene Illumination
(Lumens /Ft2)
10,000- 12,000
1000 2000
-
100
10
1
0.1
327
CAMERA CONTROL AND SETUP CIRCUITRY
Scene Illumination-The image-orthicon camera serves as an exposure
meter and is the final judge of scene illumination and lens opening. However, before an attempt is made to televise a particular scene, it is good
practice to check the incident illumination with a light meter to determine
whether the light level is adequate for a picture of good quality. In general,
the illumination should be measured with the light meter pointing toward
the camera.
Scene illumination on the camera-tube face may be calculated from the
following formula:
E
4Ef2
0.8R
where,
E. is the scene illumination in lumens/ft2 (foot -candles) ,
E is the tube -face illumination in lumens/ft2 (foot -candles) ,
f is the f number of the lens,
R is the reflectance of the scene.
For outdoor scenes, Table 8 -1 can be used as a guide in determining the
approximate scene illumination.
Because of the high sensitivity of the image orthicon, it may not be possible on very bright days to stop the lens down far enough to reduce the
high -light illumination on the photocathode to a value near the knee of
the signal -output curve. When such a condition is encountered, the use
of a Wratten neutral-density filter selected to give the required reduction
in illumination is recommended (Table 8 -2) . Ordinarily, two filters -one
having 1- percent transmission and the other 10- percent transmission-will
give sufficient choice. Such filters with lens- adapter rings can be obtained
at photographic -supply stores.
Under almost all conditions, the use of a lens shade is beneficial.
8 -4. THE CAMERA WAVEFORM MONITOR
The camera waveform monitor (CRO) always incorporates a calibration
pulse for proper gain adjustment, and a selector switch for wideband or
Table 8 -2. Neutral- Density Filters for Exposure Control
Filter
Density
0.30
0.60
0.90
1.00
2.00
3.00
4.00
Transmission
Percentage
50.0
25.0
13.0
10.0
1.0
0.10
0.010
Equivalent
Number of
Lens Stops
1
2
3
3.3
6.6
10
13.2
328
TELEVISION BROADCASTING CAMERA CHAINS
IRE response. Fig. 8 -6 illustrates the Tektronix Type 528 CRO often incorporated in camera control consoles. A somewhat more elaborate waveform monitor (Tektronix Type 529) is visible in Fig. 8 -3B. The Type 529
is more useful as a line master monitor since it incorporates vertical- interval test (VIT) signal observation.
The Type 528 television waveform monitor provides video waveform
displays on a 5 -inch CRT and occupies 51/4 inches of height and 1/2 rack
width. All-solid -state circuitry provides low power consumption and longterm reliability.
Either of two video inputs, selectable from the front panel, may be displayed. The displayed video signal also is provided at a video- output jack
for viewing on a picture monitor. Calibrated 1 -volt and 4 -volt full -scale
(140 IRE units) sensitivities are provided for displaying common video
and sync signal levels. A variable sensitivity control permits uncalibrated
displays from 0.25 volt to 4.0 volts full scale. The built -in 1 -volt calibration signal may be switched on to check vertical -sensitivity calibration.
Flat, IRE, chroma, and differential -gain frequency -response positions permit observation of various signal characteristics.
Horizontal sweep selection provides 2H ( two -line) , 1 p,s /div (expanded two -line) , 2V ( two-field) , and 2V magnified (expanded two-field)
sweeps. Displays of RGB and YRGB waveforms from color -processing
amplifiers are provided for by means of interconnection through a rear panel nine -pin receptacle. A dc restorer maintains the back porch at an
essentially constant level despite changes in signal amplitude, APL, and
color burst. This function may be turned off when not needed.
A basic function of modern camera waveform monitors is in the form of
clamping used. It will be recalled from Chapter 6 that there are two general
Courtesy Tektronix, Inc.
Fig. 8 -6. Tektronix Type 528 waveform monitor.
CAMERA CONTROL AND SETUP CIRCUITRY
329
types of keyed clamping circuitry: the "fast- acting" clamp used in processing amplifiers to remove all low -frequency ( sine -wave) disturbances,
and the "slow- acting" clamp useful in waveform monitors.
All camera waveform CRO's have a switch that allows operation with
the dc restorer off or on. However, since the restorer switch normally is
left in the on position, actual hum components in the waveform might be
ignored by the operator if he is unaware of other indicating devices which
would show this defect. Therefore, modern circuitry of this nature incorporates the slow -acting clamp so that, although the waveform is held
constant in position on the CRO with changes in APL, any hum component is visible, although reduced in amplitude. This alerts the operator to
turn the dc restorer off so that actual low- frequency characteristics may be
observed.
The fast-acting keyed clamp was covered in Chapter 6. Fig. 8 -7 illustrates
a basic slow- acting clamp typical of modern waveform monitors. Such a
circuit has three fundamental sections:
1.
A comparator that measures any difference between a fixed (or variable) reference voltage and the amplifier output voltage (E0 )
Diode Gate
Composite
Video Input
CI
A=40
Vertical
Positioning
(A) Keyed dc- feedback restorer.
(B) Equivalent circuit,
gate closed.
Fig. 8 -7. Slow- acting clamp for waveform monitor.
TELEVISION BROADCASTING CAMERA CHAINS
330
duration of
the clamping pulses
3. A memory circuit to remember the reference point between samples,
when the diode gate is open
2. A line -to -line keyed gate to close the diode gate for the
The line -to -line keyed clamp is a fast -operating type of dc restorer. However, note from Fig. 8 -7B what occurs during the time the gate is closed.
At this time, Cl is in parallel with R1, forming a low -pass filter in the
feedback loop. Thus, insufficient feedback current is available to charge Cl
to the correct value in one sample (one TV line). The equivalent effect
on the overall system is that of a "slow" restorer.
The dc- feedback restorer of Fig. 8 -7A has two basic functions: to establish dc stability (for drift -free operation) and to establish a selected portion
100
11
1111111
191
11,111111 III'
1I!11
11.1011111
VII'ïlill1111111111111!III.
11
11111111
111111
I
11111111111111111
1111
IlaliNI11R1
P
II
L,
1i61!1'9
1111111111n
ï111a
111111111111i1
1111111
l'
111111111
h.Illlflll 1I11!Iblilll
II
'111111111111111111111111111
!1!!!1il;Ii¡1i^'l'lü
flIIII1.I11111111i
1111111111.1111.1111111111
II1
'I'M ;'I'I1I'
Limits - I nner: K - 2% Outer: K - 4%
Window and 0.125 us T-Pulse Response
Courtesy Tektronix, Inc.
(A) Flat position.
Ir
-
u,
w
Kin
III
1111111111
_,
I!
1161111
nil
-zo
-
20
1
1
1
Limits- Inner: K - 2% Outer: K - 4%
Window and 0.125 us T-Pulse Response
Courtesy Tektronix, Inc.
(B) IEEE position.
Fig. 8 -8. Multiburst response
CAMERA CONTROL AND SETUP CIRCUITRY
331
Lim its-Inner: K 2% Outer: 1(6%
Window and 0.125 se T-Pulse Response
Courtesy Tektronix, Inc.
(C) Loin -pass position.
F-
--+-
--
I
- --
-20
_
Limits -Inner: K. 2%Outer:K4%
Window and 0,125 us T -Pulse Response
Courtesy Tektronix, Inc.
(D) High -pass position.
White
urst Burst Burst Burst Burst Burst
1
2
3
a
5
6
100 IEEE
0.5
1.5
2.0
3.0
3.6
A. 2
MHz
MHz
MHz
MHz
MHz
MHz
(E) Applied signal.
of waveform monitors.
7.5 IEEE
TELEVISION BROADCASTING CAMERA CHAINS
332
of the composite video waveform (sync tip or back porch) as a reference
point so that the CRO trace does not change vertical position under
varying APL.
Under ideal quiescent conditions with pulses applied to the diode
bridge, +Eo should equal -Ea. If amplifier drift causes +E0 and -Eo to
be unequal, the difference voltage applied to the comparator (differential
amplifier) is amplified. The amplified error voltage charges the memory
capacitor during the next pulse that closes the diode gate. Although the
current available through Rf is limited, amplifier drift occurs over a much
longer period than the time required to charge Cl through Rf. Thus, the
amplifier is made virtually drift -free.
Now consider the composite video waveform. The feedback diode gate
is closed during either sync -tip or back -porch time. The absolute voltage
level of the sampled part of this waveform is compared to the reference
voltage at the center arm of the position control. The difference voltage
between the point of sampled video and the positioning- control dc level is
amplified and applied to the memory. Because of the low -pass filter formed
by Cl and Rf, the error must exist continuously for about 1 millisecond
( typical time constant of the low -pass filter) before the error will be completely corrected. This time constant prevents complete removal of any
60 -Hz hum that may exist in the composite video waveform.
In monitoring the video waveform for the purpose of "riding level," it is
important that the operator use the IRE (IEEE) response position of the
1 O
__
EIRE
-_
194%to97.5 %at0.35MHz
EMI ,
o
__
70%to80%at 1MHz
i
60
r
,0
1
31.2% to 42.5% at 2 MHz
- -T
Rolloff must fall between the two curves.
20
7% to
0
i
0.35
2
1
Frequency
'C
3.6
1MHz1
Fig. 8 -9. IEEE response curve (1958 standard 235 -1).
14%at3.6MHz
1
CAMERA CONTROL AND SETUP CIRCUITRY
333
selector switch. This removes the possibility of excessive "gain riding"
with changes in high- frequency content of a scene. The multiburst response
of the Types 528 and 529 waveform monitors for various response- switch
positions is shown in Fig. 8 -8. Fig. 8 -9 illustrates the standard IEEE response curve for all modern waveform monitors when used as level -monitoring devices.
It is very important that the maintenance department check this response
with either video -sweep or single -frequency sine-wave response runs at
periodic intervals. Fig. 8 -10 gives typical response- switch circuitry and
representative maintenance adjustments provided.
Response Switch
i
i
o
Flat
Chroma
Video
Signal
Fig. 8 -10. Typical response- switch circuit.
8-5.
CAMERA INTERPHONE
A quite important link between the camera -control position and the
camera operator is the interphone system. This makes possible conversations
among the camera and control operators and the director. A schematic
diagram of the interphone for the RCA TK-60 camera chain is shown in
Fig. 8 -11A.
These units are connected internally and may be connected externally to
similar units as desired. Each small unit includes an amplifier, a bridge
rectifier, and a sidetone- compensation bridge. Sidetone is automatically
maintained at a level approximately equal to the received signal level for
any number of connected stations.
Each station employs a single -stage transistor amplifier. An unbypassed
resistor in series with the emitter of the transistor determines the gain by
controlling the amount of inverse feedback. Power to operate the amplifier
is derived from the microphone- energizing direct current from the common supply circuit to which the intercom unit is connected.
In this unit, a germanium -diode bridge rectifier is interposed between
the line and the amplifier to maintain the correct polarity at the amplifier
334
TELEVISION BROADCASTING CAMERA CHAINS
regardless of the polarity of the intercom supply voltage. Two diodes are
biased to a conducting state, and two are biased to a nonconducting state
by the direct current on the intercom circuit. Received and transmitted
voice frequencies, superimposed as they are on the direct current, pass
unimpeded through the conducting diodes to the amplifier input or to the
intercom line.
(A) Intercom unit.
Connecting Diodes
Impedance Interphone Bus
and all Other Connected Stations
Mic
(8) Sidetone bridge.
Courtesy
Fig. 8 -11. Interphone for camera chain.
RCA
CAMERA CONTROL AND SETUP CIRCUITRY
335
In order to maintain the desired relationship between sidetone level and
received level, voice frequencies must be kept out of the power supplied
to the amplifier. A choke -input filter provides the required decoupling for
this purpose.
In the interphone unit, a resistance bridge circuit (see Fig. 8 -11B for
equivalent circuit) is employed. The intercom line and all other stations
connected to it make up a portion of one side of the bridge. A pair of fixed
resistors and the forward resistances of the two conducting diodes in series
with the line complete this side of the bridge. The remaining three sides of
the bridge are provided by three fixed resistors.
The local microphone is connected across one diagonal of the bridge, and
the input to the transistor amplifier is connected across the other diagonal.
If the bridge were to be perfectly balanced, sidetone would be completely
eliminated because the amplifier input would be at the null point of the
bridge for signals from the local microphone.
When only two stations are connected to an intercom circuit, the level
received at each station is relatively high; however, the sidetone- compensation bridge at each station is substantially unbalanced, permitting a high
sidetone level to be applied to the amplifier input. When additional stations
are added to the intercom line, the received level decreases, and the impedance of the intercom circuit decreases as well. As the net impedance
of the intercom decreases, the bridge approaches, but never reaches, a condition of balance. The sidetone level decreases accordingly. Resistance
values in the bridge have been selected to hold the sidetone level to within
2 dB of the received level for any number of conference -connected stations
up to 32.
EXERCISES
NOTE: These exercises are largely review of setup and operations techniques covered in previous study. For background information, see
Harold E. Ennes, Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc.,
1971.)
Q8 -1.
Q8 -2.
Q8 -3.
Q8 -4.
Q8 -5.
Q8 -6.
Q8 -7.
What
is the maximum brightness contrast range for optimum control
of the camera, and how is this determined?
What is the maximum amount the waveform on an unclamped CRO
should shift from scene to scene?
Name the factors that can affect color -picture sharpness.
What is the final check for registration?
When a new color set is used, what is the first thing to check?
Is it possible to have good color balance on a wide shot, then not
have good balance on a tight spot in the same scene and with the
same camera?
What are the factors pertinent to background effects on skin tone?
336
Q8 -8.
Q8 -9.
Q8 -10.
Q8 -11.
Q8 -12.
Q8 -13.
Q8 -14.
Q8 -15.
Q8 -16.
Q8 -17.
TELEVISION BROADCASTING CAMERA CHAINS
How much color -temperature change can occur in lighting before
skin tones show error, and what line -voltage change does this
represent?
Are light dimmers ever used in color studios?
For color lighting, give the optimum ratio, relative to base light, for
(A) back light and (B) key, or modeling, light.
In what part of the visible -light region does the spectral response
of the average camera pickup tube peak?
Does incandescent lighting help or hinder the spectral response
of the average pickup tube?
Are the deflection yokes for all channels in a color camera connected
in series or in parallel?
Are the focus coils in a three -channel image -orthicon camera connected in series or in parallel?
What is skew, and how is it corrected?
What should be the reflectance value of reference white in a color
scene?
If there are controls marked "Q" on a color camera, what function
do they perform?
CHAPTER
9
The Subcarrier and
Encoding System
Older color systems used an external master color- subcarrier generator,
which incorporated count-down circuitry for locking the station sync generator to the frequency (color) standard. More recent systems employ sync
generators that include the color -subcarrier generator as an integral part.
This also includes the burst -flag generator (burst keyer) , which, in older
systems, was an additional rack -mounted unit.
NOTE: It is imperative at this point for the reader to review NTSC color
fundamentals. This subject is covered extensively in Harold E. Ennes,
Television Broadcasting: Equipment, Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc., 1971), Chapter 2. A
portion of that chapter deals with the choice of a color -subcarrier frequency that allows minimum interference between luminance and chrominance picture information.
9 -1. THE DOT STRUCTURE
IN NTSC COLOR
You are probably already aware of the fact that a dot pattern is noticeable on the picture tube (either monochrome or color) when a color signal is viewed. This pattern is particularly noticeable at the edges of vertical transitions, and is more pronounced on a wideband studio monitor
than on a home receiver. You may correctly assume that the process of
interleaving color information with monochrome information is less than
perfect.
From the review mentioned in the note above, we know the reason for
the choice of a subcarrier frequency at the 455th harmonic of one -half the
line- scanning frequency:
455 X 15,74.26
337
3.579545 MHz
338
TELEVISION BROADCASTING CAMERA CHAINS
This frequency results in minimum interference, as is emphasized by Fig.
9 -1 and the following analysis:
Fig. 9 -lA represents the modulated subcarrier signal. The first scan
represents scanning of a given line in the picture; this line is not scanned
again until the first scan in the next frame ( interlaced scanning) The
peaks of the subcarrier component cause a "dot pattern" to be laid down,
as in Fig. 9 -1B. The dots are actually quite small because the subcarrier
frequency is high. Also, since the frequency is an odd multiple of one-half
the line frequency (as well as being an odd multiple of one-half the frame
frequency) , the spurious dots on each line in one frame fall midway between those on the lines immediately above and below. This gives the
checkerboard pattern shown, rather than a line pattern with rows and
columns of dots. The latter would be much more noticeable.
.
Two
Fields
Two Frames
Fig. 9 -1. Principle of dot interlace.
Since the subcarrier is an odd multiple of one -half the frame frequency
(as well as an odd multiple of one-half the line frequency) , the subcarrier
component along a given line reverses in polarity between successive scans.
Remember that "successive scans" means a given field in the next frame.
The subcarrier reverses in polarity between these two scans because it
passes through some whole number of cycles plus one -half cycle (180° )
during each frame period. This is shown by the dash sine wave of Fig. 9 -1A.
The net result is shown by Fig. 9 -1C. The peaks of the subcarrier component that caused a dot pattern in one frame are cancelled in the next
frame. This process is termed cancellation interlace, or simply dot interlace.
The above would be entirely true if we had a perfect system. However,
in practice, the signals are applied to kinescopes that are inherently nonlinear even above cutoff, and certainly below cutoff (negative light) . Perfect cancellation is impossible at the present stage of development. Therefore, the cancellation of spurious signal components by dot interlacing is
not quite as good as Fig. 9 -1C indicates. It is satisfactory, however, at normal viewing distances. When you are very close to a monitor tube (either
monochrome or color) that is displaying a color signal you can see the
"dot crawl" that occurs from bottom to top of the raster. On a wideband
studio picture monitor, this is particularly noticeable at transitions between
color bars.
339
THE SUBCARRIER AND ENCODING SYSTEM
Fig. 9 -2 represents a part of a scanning pattern in the presence of a disturbing frequency that is an odd multiple of one -half the line frequency.
The letters in the circles have the following significance:
(A) Dots produced by the disturbing signal
(
subcarrier) in the first
field of scanning
(B) Dots produced in the second field
(C) Dots produced in the third field
(D) Dots produced in the fourth
field (two frames of picture)
From this we can see that a complete cycle of the dot pattern takes up two
full frame scans. Also observe that the dot patterns are opposite each
other in successive lines and in successive frames.
i
Fig. 9 -2. Complete dot -pattern cycle.
9 -3. Dot crawl.
Fig. 9-3.
Dot crawl, the apparent upward motion of the dot pattern in the picture resulting from the sequence in which dots are laid down, is illustrated
by Fig. 9 -3. The circles represent points of maximum intensity, and the
numbers within them indicate the field to which they correspond. The
time sequence (1- 2 -3 -4) , which the eye tends to follow, corresponds to a
continuous upward motion. In effect, the dot crawls one line per field.
There are 60 fields per second, so the dot crawls 60 lines per second. There
are about 480 active lines displayed on the raster (525 lines minus vertical blanking lines) Thus, 480/60, or 8, seconds are required for the dot to
crawl from the bottom to the top of the raster.
Actually, the field rate for the color standards is not quite 60 per second,
but is 59.94 per second. As a result, slightly more than 8 seconds are required for a dot to crawl from the bottom to the top line of the raster. If
you concentrate on a dot on the bottom line and follow it to the top line
(using a stop watch) , and if the elapsed time is between 8 and 8.1 seconds,
the subcarrier frequency is correct. In an emergency, you can use this
method to set the subcarrier frequency rather accurately. Naturally, you
must use an underswept monitor so that all active lines are visible.
.
9 -2. POWER -LINE
CRAWL ON COLOR STANDARDS
Since there is a difference (slip) between the color field rate and the
60 -Hz power -line frequency, a "crawling" hum bar results on receivers or
monitors if hum is present either at the transmission or receiving end.
340
TELEVISION BROADCASTING CAMERA CHAINS
At the receiver, this is no different from the condition that existed for
monochrome operation when the receiver was powered from a source not
synchronized with the source of transmitter power. Thus, when interference, such as from operation of power tools or rotating machinery in the
neighborhood, exists on the power line, a floating "band" of noise flashes
can drift from bottom to top of the receiver picture tube.
You can observe this slip by connecting a scope at the composite -sync
output of the sync generator (locked to color standards) and adjusting
the time base to give one field in 10 centimeters on the graticule. Use a
60 -Hz trigger on the scope. Since the difference between 60 and 59.94 is
0.06, there is a shift of 0.06 field per second, and the vertical interval will
require about 16 seconds to travel the 10 centimeters on the scope. But do
not try to set the actual subcarrier frequency by this method. Because of
the large countdown, the error could be considerable.
9 -3.
ADJUSTING THE SUBCARRIER COUNTDOWN
For color operation, the master oscillator in the sync generator is no
longer the frequency standard of the system. To maintain the proper harmonic relationship between the subcarrier frequency and the scanning
frequencies, the subcarrier generator becomes the frequency standard for
the system, and is so termed by most manufacturers.
The subcarrier- frequency oscillator is normally a crystal -controlled type
with the crystal in a precision temperature -controlled oven. A trimmer
allows adjustment of the frequency with reference to an external standard.
See Fig. 9 -4. The crystal oscillator feeds a conventional buffer amplifier
peaked at the subcarrier frequency. Note the low -value resistor (2 ohms)
across the secondary of Ti. This provides a very low sending -end impedance for conventional 75 -ohm distribution lines.
Tube -type subcarrier generators almost universally employ locked oscillators for counters. This type of oscillator operates at the desired output
-
Locked Osc
V3
Buffer Amo
V2
Low
Internal Impedance
T1
Distribution
TCounter
To Next
Fig. 9 -4. Use of locked oscillator.
75
3-11
THE SUBCARRIER AND ENCODING SYSTEM
r
2
r
r
n
ff
3
5
5
(B) Expanded cycle.
(A) Fundamental.
All "Shoulders" Are Even
2
I_Detail
in
(C) Locked waveform.
D
3
5
5
I
(D) One
cycle of
(C).
Fig. 9 -5. Frequency- divider waveforms.
frequency, with provision for injecting a small voltage into its grid circuit
to lock the output at a subharmonic of the injected frequency. A typical
circuit is that of V3 of Fig. 9 -4. This is a Colpitts circuit, in which the
cathode and the midpoint of the tank -circuit capacitors are at ground potential. (The Hartley circuit also may be used in this application. The
action is the same.) The tank circuit is adjusted so that, in the absence of
the 3.58 -MHz signal, V3 oscillates at a frequency very close to the required counter frequency. If this is a 5 -to-1 counter, the frequency is
715.909 kHz. The waveform looks something like that in Fig. 9 -5A.
Application of the 3.58 -MHz signal to the V3 grid should lock the oscillator in precise phase and maintain a stable frequency division of (in this
example) 5 to 1. Such will be the case if the peaking of T1 is correct and
the tank coil is properly tuned. The trick is knowing how to recognize the
proper locked frequency. Fig. 9 -5B shows one lower- frequency (715.909
kHz) cycle with the higher- frequency subcarrier "lock." This is on an
expanded time base to show details. Count all the negative -going peaks to
check actual count. Trigger the scope on the negative slope of the signal
so that the sweep starts at the end of the previous negative peak. We can
see here that the count is five.
Unfortunately, a locked oscillator can produce a nonintegral count, for
example 4.5 to 1. You should always use a time base that will display seven
or more complete lower- frequency cycles, as illustrated in Fig. 9 -5C. A
properly locked 5 -to -1 (or 7 -to -1) waveform has identical symmetry over
all cycles. A gradual change over the cycles reveals a nonintegral countdown.
342
TELEVISION BROADCASTING CAMERA CHAINS
Locking action is a positive function and will occur with a sharp increase
of "crispness" as lock -in is achieved. A smeared presentation on the scope
indicates that the stage is not synchronized in the presence of the higher frequency input. ( NOTE: Always check the stability of the scope trigger
function.)
Fig. 9 -5D shows how to interpret the count on the longer time base of
Fig. 9 -5C. Remember to count the negative peaks, as shown. On this time
base, the negative excursion looks more like a slight notch.
The 7 -to -1 locked oscillator receives the 5 -to -1 countdown and is usually
the same type of circuit. Again, it can lock to a nonintegral count such as
6.50 to 1 or even 6.75 to 1. The best way to check this is to observe the
waveform of the 5 -to -1 counter and adjust the scope time base to obtain
exactly seven of the low- frequency cycles in 10 centimeters on the scope
graticule. (If you prefer, you can obtain seven complete cycles in 7 centimeters.) Then without changing the scope time base, observe the test
10 cm
7
1
Cycles of
5
-to -1 Counter
Cycle of
7
-to-1 Counter
(A) Method of determining count.
1st Cycle
6th Cycle
(8) Double -check for exact
count.
Fig. 9-6. Waveform checks of divider count.
THE SUBCARRIER AND ENCODING SYSTEM
2
Fig.
9 -7.
343
Cycles at Input
8 Cycles at Output
Check
of multiplier action.
7 -to -1 counter. There should be exactly one
low- frequency cycle in the same space on the scope graticule (Fig. 9 -6A) .
Remember to use the trigger on the negative slope so that the sweep start "dot" occurs on the last negative peak of the previous cycle.
Fig. 9 -6B shows the importance of double checking the frequency by
readjusting the scope time base to display seven or more cycles of the
lower- frequency component. In the example shown, the oscillator is locking to a 6.75 -to-1 count instead of 7 to 1. Always adjust the oscillator
tank coil to a point midrange between proper counts on the lower and
upper limits. A number of consecutive cycles should appear the same.
The four -times multiplier is usually a conventional circuit with the plate
circuit resonated at four times the input frequency. There is really only
one suitable method for adjusting this stage, and this is as follows:
point at the output of the
1.
While observing the 7 -to -1 counter, adjust the scope time base to
get exactly two cycles of the low- frequency component in exactly 10
centimeters on the scope graticule. Trigger the scope on the negative
2.
Without changing the scope time base, observe the output of the
four -times multiplier. Adjust it to obtain exactly eight cycles in the
slope of the signal.
10 centimeters (Fig. 9 -7) Adjust the frequency to the center of the
range between the upper and lower limits of the proper count. In
most cases, it is only necessary to adjust the tank circuit for maximum
output. The stage is usually noncritical, being capable of tuning only
to the proper frequency of four times the input signal.
.
NOTE: The exact sequence in counting can be different. Therefore, the
input frequency to a given counter (or the multiplier) may differ from the
above examples. The thing to remember is to trigger the sweep with the
counter preceding the one to be checked, and adjust the time base for the
number of cycles that equals the division of the counter to be checked.
You should use the same procedure in checking the operation of the sync generator counters.
Counters with ratios higher than 7 to 1 (such as 13 to 1) are usually
of more straightforward design, such as a conventional cathode -coupled
344
TELEVISION BROADCASTING CAMERA CHAINS
(A) Four -cycle envelope.
(B) One cycle expanded.
Fig. 9 -8. Waveforms of 13 -to -1 counter.
THE SUBCARRIER AND ENCODING SYSTEM
345
oscillator. The higher- frequency component is injected into the grid to
obtain the locking action. Fig. 9 -8A shows a four-cycle "envelope" of the
output of a 13 -to-1 counter. Although it is a rather difficult task, you are
assured of better accuracy of adjustment if you count the high- frequency
peaks occurring over the four low- frequency cycles. There should be
13 X 4, or 52, peaks. Fig. 9 -8B illustrates one cycle on an expanded time
base.
9 -4. SOLID -STATE COUNTERS
In the interest of completely "automatic" circuitry, modern solid -state
subcarrier counter circuits often incorporate no adjustments. It is therefore necessary for the maintenance engineer to be able to analyze such
circuitry in the event of miscounting or complete failure.
Input
J
Output
L
Timing
--I
Rest
r- Measure -t-
Recover
I
Rest
-
Fig. 9 -9. Timing chart of typical solid -state divider.
A counter that is gaining great popularity is a special form of monostable or bistable multivibrator circuit employing noncritical timing elements. It has three basic states of operation -rest, measure, and recover
as shown for a typical 5 -to -1 counter in Fig. 9 -9. Note that the measure
ramp is initiated by one positive -going trigger (trailing edge of one sync
pulse) and is terminated by the next positive -going trigger (trailing edge
of next sync pulse) . The rate (slope) of the recover period is such that
it automatically terminates about midway between the last input trigger
to be rejected and the next one to count. The remaining time is the rest
period.
Fig. 9 -10A shows a typical 7 -to -1 divider. Transistors Q1 and Q3 form
a bistable multivibrator with diode trigger steering (X2 and X5) . Transistor Q2 is the ramp generator with a collector- catching network (to prevent saturation) and a bias network controlled by switching transistor Q4.
-
TELEVISION BROADCASTING CAMERA CHAINS
346
U
en
1
U
in
X
S
NA.A.,C
(A) Circuit diagram.
Fig. 9-10. Typical transistor
)-}7
THE SUBCARRIER AND ENCODING SYSTEM
Waveform
C
V
cm
cm
5.0
0.5
O1NIIIIM;
i,'f
T
11111111111111111111111111
111111111111111111.1111111111
E
F
5.0
1.0
riri
0.5
0.5
MMEI^VIE
MANIMIATMOS
110i111111I1/1111111
1,.1OIN'S
G
5.0
0.5
D
5.0
0.5
2
ASAIMINIMIKANI
MENIIMMINIME
11111111111111111111111
11r111111r1111111111r11ri11
1
t2
t3-
` t4. t1+
To
IT0 Output Period)
(B) Waveforms.
frequency divider.
TELEVISION BROADCASTING CAMERA CHAINS
348
Although you should already have a good basic background in transistor
circuitry, the analysis of this circuit can be tricky, so we will study it
briefly.
Since a bistable multivibrator has two stable conditions, assume the initial state to be just prior to t1. Transistors Q1 and Q4 are on, and Q2 and
Q3 are off. The negative input trigger at t1 (waveform C, Fig. 9 -10B) , has
no immediate effect at .X2 since the anode of this diode is at a negative
potential through the saturated Ql. But the coupling through C5 causes X5
to pass the negative trigger. This drives Q1 and Q4 off, reversing Q2 and
Q3 to the on condition. A negative -going ramp develops at the Q2 collector (waveform F) so that at t2, when the next trigger appears, X5 blocks
it but X2 passes it. Diode X5 is not affected because its anode is negative
through Q3. But Q1 is off (anode of X2 is positive) , so X2 passes the
negative trigger. This drives Q2 and Q3 off, reversing the operation of
Q1 to the on condition. But note this: Q4, which is controlled by the collector of Q2, remains off, preventing further triggering through X5. A
positive -going ramp develops at the Q2 collector (waveform F) , and,
after the time corresponding to the predetermined number of triggers, this
ramp causes Q4 to go on at t3. Since X5 will now pass the next trigger,
the cycle is repeated.
NOTE: Observe through the above anlysis that negative -going triggers are
used. These are npn transistors. The example of Fig. 9 -9 applies if pnp
transistors are employed.
In the collector circuit of Q2, R1, X1, and X4 prevent Ql and Q4 from
being off and Q2 and Q3 from being in saturation upon application of
supply voltages. This collector- catching network prevents Q2 from saturating and assures a starting condition with Q1 and Q4 on. If Q2 is saturated, an otherwise normal pulse passed by X2 might not be sufficient to
trigger the circuit out of conduction.
9 -5.
FINAL SUBCARRIER GENERATOR COUNTDOWN CHECK
The following steps constitute a final check on the countdown of the
subcarrier generator:
Connect the vertical input of the scope to the 3.58 -MHz generator
output.
2. Use external trigger for the scope horizontal sweep and trigger from
the 31.468 -kHz generator countdown.
3. The pattern on the scope should lock on a steady trace with no
tendency to drift or jump. Temporarily remove the external trigger
to be certain the scope is under external trigger control. Critically
adjust the trigger -amplitude control on the scope for a "clean" trace.
4. Reversing the above connections to the scope should also produce
1.
a steady trace.
THE SUBCARRIER AND ENCODING SYSTEM
349
NOTE: Some of the more recent generators (solid- state) count down as
follows:
Divide by 5 to 715.909 kHz
Divide by 7 to 102.3 kHz
Divide by 13 to 7867 Hz
This sequence gives one -half the line frequency, or a total division of
455. This signal then locks a 31.468 -kHz master oscillator in the sync
generator by means of an automatic phase- frequency control (APFC)
using a voltage -controlled circuit. In this case, Step 2 above simply involves taking the scope external trigger from the master oscillator in the
sync generator.
9 -6.
FINAL CHECK ON SYNC-GENERATOR COLOR LOCK
For a final check on color lock of the sync generator, these steps may be
followed:
1.
2.
3.
4.
5.
Place the sync generator on internal crystal control, and check all
counters in the generator for proper centering.
Place the sync generator on free -run operation. Observe any vertical
output (composite sync, blanking, or vertical drive) with the scope
on 60 -Hz trigger. Adjust the master -oscillator frequency for a very
slow drift of the trace from left to right on the scope. Check the syncgenerator AFC operation with 60 -Hz line lock.
Place the sync generator under external control with the color -subcarrier countdown (31.468 kHz) feeding the external input of the
sync generator.
Connect the scope vertical input to the 3.58 -MHz output of the color subcarrier generator. Trigger the scope horizontal sweep with horizontal drive from the sync generator. The resulting pattern should
be perfectly stationary with no erratic jumps.
If this condition does not exist, recheck the color-subcarrier generator
as in Section 9 -5. If the subcarrier generator is not at fault, the trouble is in the AFC (or APFC) circuitry of the sync generator.
9 -7. SETTING THE COLOR- SUBCARRIER FREQUENCY
There are four quite satisfactory methods for setting the subcarrier frequency. Of these, only Method 1 gives positive assurance that the FCC
frequency tolerance is being satisfied by the station.
Method 1: Frequency Standard and Primary Measuring Service
With this method, the station may or may not have a color -subcarrier
frequency meter. The procedure is to contact the frequency- measuring
service by telephone and transmit a composite color signal for purposes of
measuring the subcarrier burst frequency. If the station has a subcarrier-
350
TELEVISION BROADCASTING CAMERA CHAINS
frequency meter, adjust it to agree with the measuring- service reading,
and then adjust the subcarrier frequency for zero reading. Or, the service
can "talk you in" to the proper frequency adjustment.
Method 2: Vectorscope
If the station is affiliated with a network, or if the output of a good
color receiver can be fed to the alternate input of the vectorscope, you can
make use of a rapid and convenient way of adjusting the color -subcarrier
frequency. In this method you must assume, of course, that the network or
the received station is using the proper subcarrier frequency. Feed the external composite color signal to the B input of the vectorscope, and feed
the local composite color signal to the A input. Place the vectorscope on
external 3.58 -MHz lock from the subcarrier generator (normal operation
in station use). Place the input switch in the "A shared with B" position.
The local burst, locked to the local subcarrier generator, will be stationary,
and the signal being used as a standard will have its burst rotating at a
rate dependent on the frequency difference. Adjust the local subcarrier frequency until the "standard" burst is as nearly stationary as possible. You
cannot obtain a steady lock unless you have color genlock facilities.
This is an extremely quick and convenient procedure if a signal -selector
switch is incorporated with the B vectorscope input, so that you can
"punch up" the reference frequency at a moment's notice.
Method 3: Oscilloscope
If you do not have a vectorscope, the next best procedure is to feed the
reference composite color signal to the external trigger input of an oscilloscope. Trigger the scope at the horizontal frequency. Observe 10 to 12
cycles of the local 3.58 -MHz subcarrier on the scope triggered by the reference signal. Adjust the subcarrier frequency to obtain a stable trace.
Method 4: Dot Crawl
This method should be used only in an emergency when you do not have
any other means of checking the subcarrier frcquency. With a good stop
watch, time the travel of a single dot from the bottom to the top of the
raster on an underscanned monitor. The dots are quite apparent at color -bar
transitions. Adjust the subcarrier frequency until this time is between 8
and 8.1 seconds. (Review Section 9-1.)
9 -8. THE COLOR -SYNC
TIMING SYSTEM
The burst keyer supplies a gating pulse (burst flag) to the encoder
(colorplexer) for the purpose of keying on a burst of the subcarrier sine
wave. This burst (used as a frequency and phase reference in color monitors and receivers) consists of 8 to 10 cycles of the 3.579545 -MHz sub carrier, delayed 0.39 to 0.64 µs from the trailing edge of horizontal sync.
THE SUBCARRIER AND ENCODING SYSTEM
Fig. 9-11. Tube -type burst -key generator.
351
TELEVISION BROADCASTING CAMERA CHAINS
352
The burst therefore occurs on the back porch of horizontal blanking. Furthermore, this gating pulse must be defeated during the 9H interval of
equalizing and vertical -sync pulses.
Fig. 9 -11A is a simplified functional schematic diagram of a typical
tube -type burst -key generator. The vertical- and horizontal -pulse inputs
may come from composite station sync by way of a sync separator, or they
may be the sync -generator drive-pulse outputs. In some cases, horizontal
pulses are derived from composite sync, but vertical pulses are supplied
from sync -generator vertical drive.
The circuit of V1 is a typical cathode -coupled multivibrator; in this case
it is used as the burst- eliminate pulse generator. The network composed of
Cl and R1 differentiates the pulse, and diodes X1 and X2 pass positive
triggers to the grid of V 1A. Note that this grid is returned to ground
through the diodes, but the grid of V1B has a positive potential. Therefore, V 1B is normally on, while VIA is cut off by the positive potential at
its cathode (conventional cathode -coupled arrangement) The positive
trigger applied at the grid of VIA drives this section to conduction;
the resulting negative pulse at the plate drives V1B to cutoff. The charge on
C2 holds this state for a time determined by the grid potential of V1B,
which is adjusted by the eliminate -width control. This control normally is
adjusted for a pulse width of 9H.
This stage usually is followed by a conventional clipper (V2) to get a
good, flat top over the 9 -line duration. The resulting negative pulse of 9H
duration is fed to the suppressor grid of the mixer tube and must be of
sufficient amplitude to hold this tube nonconducting during the 9H interval.
The horizontal circuitry is a duplicate ( except for time constants) of that
described above. In this case, two multivibrators must be used: One generates a delayed trigger for the actual burst -width multivibrator so that the
color -sync burst will be gated on at the appropriate time following the trailing edge of horizontal sync. The timing waveforms are shown by number
in Fig. 9 -11B. The burst -delay multivibrator is adjusted so that the trailing edge triggers the burst -width multivibrator approximately 0.5 µs
following the trailing edge of horizontal sync. The burst -width multi vibrator is adjusted to allow 8 or 9 cycles of the subcarrier to appear in
the burst. Since one cycle at 3.579545 MHz has a duration of 0.28 µs,
then:
.
= (8) (0.28) = 2.24 µs (minimum)
cycles = (9) (0.28) = 2.52 p.s nominal
8 cycles
9
The resultant positive pulse is applied to the control grid of the mixer
tube and is passed except during the time the tube is held at cutoff by the
burst -eliminate (negative) pulse on the suppressor grid.
Some burst keyers employ a tapped delay line and a bistable multi vibrator. The first tap on the delay line goes to an on- trigger tube, and the
THE SUBCARRIER AND ENCODING SYSTEM
353
second tap goes to an off -trigger tube. These triggers control the two sections of the bistable multivibrator so that the duration of the pulse is set
by the delay line rather than multivibrator time constants.
Fig. 9 -12 presents typical circuitry in transistor burst -key generators.
The horizontal pulse is amplified and inverted by Q1. The inverted pulse
is integrated by the combination of R1, R2, and Cl. Note that the Q2
-20V
-10
-1
Clipper
Horiz Pulse
Boxcar
I
Q3
C2
Q2
-U-
Burst
-Width
J
R3
To
L
Gate
1111
(A) Diagram.
,-,
i
Q2 Base
-10 V
I
I
I
I
-
-20V
(B) Waveforms.
I
Q2
Collector
II
0
I
L
f--{
Fig. 9 -12. Solid -state
10 V
Delay Range (Breezeway)
burst-key generator.
emitter is returned to -10 volts, and the voltage at the Q1 collector holds
Q2 in the off condition until the integrated pulse at the Q2 base becomes
more positive than the Q2 emitter voltage. When this occurs, Q2 saturates,
and its collector goes essentially to -10 volts. So the Q2 collector is at
zero volts prior to the pulse (cutoff) and swings to -10 volts for that
portion of the base pulse more positive than -10 volts. The delay control
(R2) adjusts the rise time of the integration (Fig. 9 -12B) Therefore the
output pulse from the Q2 collector is delayed the proper amount for the
.
breezeway interval.
Transistor Q3 is connected in a "boxcar" circuit, in which the clamped
base pulse width (and therefore the output pulse width) is determined
TELEVISION BROADCASTING CAMERA CHAINS
354
by the C2R3 product. Width control R3 is adjusted to obtain the proper
number of cycles in the burst interval.
A typical transistor gate, which performs the same function as V3 in
Fig. 9 -11, is shown in Fig. 9 -13. Note that when either transistor is saturated, the common collector is at ground (zero) potential. Note also that
both transistors must be cut off for the collector to reach the supply
voltage. In the absence of input pulses, Q1 is saturated and Q2 is cut off.
Burst Width
Ql Base
Ql
9H
Output
91i Elim
Pulse
(B) Waveforms.
(A) Diagram.
Fig. 9 -13. Burst -key gate.
The collector is at zero (ground) potential. When the positive burst -width
(burst-key) pulse arrives at the Q1 base, this transistor is cut off. Since
Q2 is already at cutoff, the collector rises to the negative supply voltage.
Therefore an inverted key pulse is passed to the output across RL. This
action continues until the 9H eliminate pulse arrives at the Q2 base. For
the duration of this pulse, Q2 is saturated and the collector voltage remains
at ground, eliminating the burst -key pulses at Ql.
NOTE: Recall that all synchronizing generators inherently employ a 9H
gating pulse in the formation of composite sync. In modern color -sync
circuits, you will find the burst flag is generated in the main portion of the
sync generator without need of additional equipment. In older equipment,
it is formed in a separate unit.
9 -9.
ADJUSTMENT OF BURST-KEY GENERATOR WITH
ENCODED SIGNAL
An encoded signal may be used in the adjustment of the burst -key generator according to the following procedure:
Observe the composite color signal (sync and blanking added) at
some point after encoding. Trigger the scope externally with horizontal drive to obtain a steady trace.
2. Adjust the burst -key delay control for proper breezeway (Fig. 9 -14).
1.
THE SUBCARRIER AND ENCODING SYSTEM
355
11.1µs
0.56us
1. 59
is
4.76us
2. 24
us--1. %us
0.90P
0.9to1.1S
0.I1P
O.1S
B
eeteway
Nominal
Microseconds
Blanking
1
1
1
Tolerance
Microseconds
+0.3
-0.6
Sync
4.76
±0.32
Front Porch
1.59
+0.13
Back Porch
4.76
+0.96
-0.61
Sync to Burst
0.56
+0.08
Burst
2.24
Blanking to Burst'
6.91
Sync
&
Burst
Sync & Back Porch
-0.32
-0.17
+0.27
-0
+0.08
-0.17
7.56
+0.38
9.54
±0.32
-0.49
'Blanking -to -burst tolerances apply only to signal before addition of sync.
Fig. 9 -14. Time intervals for horizontal sync and burst.
3.
Adjust the burst -key width control for 8 or 9 complete cycles. Do not
count cycles of less than 50 percent of the nominal peak -to-peak
value (Fig. 9 -15A) If the first cycle starts with a positive -going
alternation, count the negative peaks. If the first cycle starts with a
negative -going alternation, count the positive peaks (Fig. 9 -15B).
.
NOTE: If you use straightforward internal triggering of the scope instead
of an external trigger, you can obtain the "interlaced" burst pattern of
Fig. 9 -15C. You should count 16 peaks for 8 cycles of noninterlaced
burst.
TELEVISION BROADCASTING CAMERA CHAINS
356
4. Trigger the scope externally with vertical drive, and use the delayed
sweep on the scope while observing the vertical interval for one field.
The time base should be such as to permit observation of the 9H
interval plus a few horizontal -sync pulses following the trailing equal-
izing pulses. Adjust the burst -eliminate control to eliminate all
bursts in the 9H interval. Operate the field -shift key on the scope, and
observe the alternate field. If necessary, readjust the burst- eliminate
control to obtain the 9H "key out" on this field. Go back and forth
between the two fields and adjust until the 9H elimination is correct
for both fields (Fig. 9 -16)
5. In certain tube -type key generators, the tubes for the delay, burst width, and burst -eliminate multivibrators must be selected for best
stability of adjustment. Initially, check these adjustments daily. Experience with the particular equipment will then indicate whether
these checks can be made on a weekly or monthly schedule.
.
Count
8
1
Do
not count.
(A) Peak amplitudes.
Count
1
(B) Polarity of peaks.
Count 16 peaks for
8
cycles.
- WT000=00000>
(C) Interlaced display.
Fig. 9 -15. Counting of cycles in burst.
9 -10.
ADJUSTMENT OF BURST -KEY GENERATOR BY ITSELF
In some instances, after servicing it might be necessary to adjust the
burst -key generator as a unit before placing it in service with the encoder
unit. The procedure is as follows:
1.
Always check first for proper output level. This is usually 4 volts
peak -to -peak across a 75 -ohm load.
THE SUBCARRIER AND ENCODING SYSTEM
Fig. 9-16. Adjustment of burst -eliminate control.
357
TELEVISION BROADCASTING CAMERA CHAINS
358
output of the burst -key generator. You can
do this quite simply by using a coax "T" on the scope; insert sync in
one input, and connect the scope probe to the 75 -ohm termination on
the key -generator output.
3. Trigger the scope externally with horizontal drive. Adjust the delay
control for a breezeway (in this case, the interval between the trailing edge of horizontal sync and the leading edge of the key pulse) of
2. Mix station sync with the
about 0.5 µs.
4. Adjust the burst -width control for a key -pulse (flag pulse) width of
2.4 µs.
5. Set the scope time base to vertical rate and trigger with vertical
drive. Adjust the burst -eliminate control so that no flag pulses appear
during the 9H vertical interval, but pulses start immediately after
the first horizontal sync pulse following the last trailing equalizing
pulse in both fields.
9 -11. THE ENCODING PROCESS
Operations performed by the encoder (colorplexer) are:
(A) Matrixing of R, G, and
(B)
(C)
(D)
(E)
(F)
(G)
(H)
B video signals from the camera processing amplifiers to produce luminance and chrominance signals. (In
four-tube cameras, Y, R, G, and B signals are involved.)
Filtering of the chrominance signals to the required bandwidth.
Delay compensation in the Y and I channels to correct for the
delay of the Q signal (narrowest bandwidth, hence most delayed) .
Modulation of the 3.58 -MHz carrier by the chrominance signals.
Insertion of the color-sync burst.
Aperture compensation of the luminance (Y) signal.
Mixing of the Y, I, and Q signals to form the complete color
signal.
Insertion of composite sync (optional)
.
Fig. 9 -17 is a functional block diagram of the encoding system. We will
elaborate each of the blocks as we go along. For a review of the mathematics of the encoding system, see Harold E. Ennes, Television Broadcasting; Equipment, Systems, and Operating Fundamentals (Indianapolis:
Howard W. Sams & Co., Inc., 1971) Chapter 2. It is important to understand at the start that any signal inserted at the red input of the encoder
will activate the red gun in the color picture tube. The same is true for
the green and blue inputs of the encoder and their respective guns in the
color picture -tube.
Fig. 9 -18A illustrates the Cohu color video encoder. The operation of
the color encoder is discussed in general terms with reference to Fig. 9 -18B.
This diagram shows in block form the various sections of the encoder,
except the power supply, and indicates the signal flow between sections.
,
359
THE SUBCARRIER AND ENCODING SYSTEM
hE'
E
o
L
U
ác
a
E
.
3
"8
E
/
h,1,
,
'
,i
enwe3
Fig. 9 -17. Basic encoding system.
360
TELEVISION BROADCASTING CAMERA CHAINS
The function of the encoder is to process and combine individual inputs to produce a compatible color signal. In addition, the encoder has
integral full -bar and split -bar generators, and various other test and operational features. It also contains circuits for optional relays for the remote
control of certain functions.
The signal, which is provided at three separate outputs of the encoder,
is suitable for use in color and monochrome systems that include video display devices, recorders, and transmitters. The encoder requires a 3.58 -MHz
subcarrier, sync, blanking, and RGB or YRGB video inputs.
The encoder pulse and subcarrier inputs pass through isolation circuits
on the input board to the matrix, modulator, and processor boards. The
video inputs pass through isolation and switching circuits on the input
board to the matrix board. When the internal bar generators are being
used, the video signals are blocked at the input board. The switching circuits are controlled from the matrix board.
At the matrix board, the red, green, and blue video signals from the
input board (or from the bar generators) are matrixed and amplified to
provide Y, I, and Q video signals. The blanking - signal input of the matrix
board is used in the generation of full -bar or split -bar outputs. The I and
Q video signals are filtered and fed to the modulator board. Before reaching the modulator board, the I signal is delayed to compensate for the
greater delay of the Q signal, which has a narrower bandwidth.
If RGB inputs or the bar generators are used, the Y signal, which is derived from matrixed RGB video, provides the luminance. The Y signal
may or may not pass through a 3.58 -MHz notch filter on the matrix board,
depending on the connection of a patch cord. If YRGB inputs and luminance correction are used, the luminance signal is derived from a network
that uses the Y signal from the matrix and the monochrome input. Without
luminance correction, the monochrome input is the luminance signal and
is amplified and fed to the Y output of the matrix board.
The luminance signal, which is fed to the processor board, also passes
through a delay line to bring it into time coincidence with the Q signal.
The luminance line has a termination adjustment (R1), which, together
with the two delay lines, is on the interconnection board.
(A) Photograph.
Fig. 9 -18. Cohu
361
THE SUBCARRIER AND ENCODING SYSTEM
o
3
m
a
E
U
C
C
C
m
CJ
U
1
C
j
m
C
N
i
C
C
_
m
(B) Block diagram.
Courtesy Cohu Electronics, Inc.
color video encoder.
362
TELEVISION BROADCASTING CAMERA CHAINS
In addition to the matrixing circuits, the bar -generator circuits, the
luminance- correction network, and the notch filter, the matrix board contains the input, mode, and various other switching circuits including ( when
installed) the relays for remote control. Also included are switching circuits for disabling the subcarrier and burst -flag circuits on the modulator
board when monochrome -without -burst outputs are required from the
encoder.
At the modulator board, the subcarrier signal is amplified and limited
to obtain constant amplitude. The sync input of the modulator board passes
through a limiting circuit, and is delayed to bring it into coincidence with
the I, Q, and luminance signals. The output of a burst -flag generator, which
also is fed to the processor board, adds burst -flag pulses to the I and Q
signals, which are amplified and clamped at the black reference level
(blanking) by the sync signal.
The combined I video and burst -flag signal modulates a subcarrier signal, and the combined Q video and burst -flag signal modulates another
subcarrier signal that has been passed through a 90° phase-shifter network.
The suppressed- carrier outputs of the modulators are passed through
tuned amplifiers, combined, and fed as the chroma signal to the processor
board through the chroma- switching circuit on the input board.
The modulator board also has a 360° phase- shifter circuit that includes
coarse and fine phase- shifting controls for matching subcarriers to two or
more encoders.
At the processor board, the luminance signal is amplified, clamped, and
white -peak clipped before passing to a circuit in which aperture correction
can be performed. After further amplification, the luminance signal is
again clamped, and blanking is then added. The blanking signal, which
comes from an amplifier and delay circuit, is also combined with the burst flag pulses after they have passed through a delay and pulse- shaping circuit.
The Matrix
Fig. 9 -19 illustrates one example of a matrix (cross -connected voltage
divider) for proportioning the red, green, and blue video inputs to obtain
Y, I, and Q signals. The matrix consists of three sets of three resistors. The
matrix input signals, R, -R, B, G, and -G, are applied to the resistors as
indicated in Fig. 9 -19, with each resistor receiving one input. The output
of each set of resistors is capacitively coupled to an amplifier (Y, I, or Q) .
The resistance values in each set of resistors are proportioned so that the
currents in the resistors are summed at the amplifier inputs in the following
standard proportions:
+ 59%G + 11%B
-I = -60%R + 28%G + 32%B
Y = 30%R
Q=21%R-52%G+31%B
363
THE SUBCARRIER AND ENCODING SYSTEM
a
o
Rl
Red
Y
30%
R +
59%
G +
11% B
R2
R3
R4
I
to
Bandwidth
Filter
-60%
G
Green
R +
28%
G +
32% B
-Q to
R7
Bandwidth
Filter
21%
R
- 52%
G +
31%
B
-21%
R
+ 52% G - 31% B
Blue -4.
Fig. 9 -19. Example of matrix circuitry.
The Y output of the matrix results from the addition of the red, green,
and blue signals developed across R1, R2, and R3. The Y signal has component amplitudes, proportioned as indicated in the Y equation, corresponding to the brightness factors of the colors red, green, and blue relative to the 100 -percent brightness characteristic of white. Similarly, the I
and Q signals have component amplitudes proportioned as indicated in
the respective equations. The vector sum of I and Q represents the color.
The magnitude of the vector carries the saturation information, and the
angle carries the hue information. Note that with equal red, green, and blue
inputs (white or gray) the algebraic sum of I and Q is zero.
See Fig. 9 -20. These are more detailed drawings of the specific I ( Fig.
9 -20A) and Q ( Fig. 9 -20B) networks. The I phase inverter inverts the
phase of the red signal injected at the grid; since blue and green are injected at the cathode, no phase inversion of these signals takes place. Hence,
the video signal at the plate is -I.
Obviously, the gains for blue and green are different from the gain for
red. Remember that a white or gray signal means that all inputs are of
identical amplitude. Thus, when R = G = B, the white -balance control (actually a red -signal gain control) is adjusted so that all subcarrier is cancelled in the I output.
Note also that since the I -test pulse is injected at this stage, it becomes
a
-test signal. Since this pulse occurs either on a split field (without RGB
present at the same time) or by itself ( also without RGB present) , and
since it is not simultaneously matrixed with the luminance channel, it has
zero luminance and is inserted on the blanking pedestal. It is matrixed such
that with normal input level, its peak-to -peak amplitude at the encoded
output is the same as the amplitude of a properly adjusted sync burst. It is
-I
TELEVISION BROADCASTING CAMERA CHAINS
364
filled with subcarrier just as are all of the actual color bars; only white
(equal input amplitudes) contains no subcarrier.
You now should be able to correlate the signal processing just described
with that of the Q phase inverter (Fig. 9 -20B) , and note that the output
here is a +Q signal. In this circuit, green is the inverted signal and is
adjusted for Q white balance.
Bandwidth Limiting and Delay Compensation
The equivalent bandwidths prior to modulation assigned to the I and
Q signals are as follows:
Q- Channel Bandwidth
At 400 kHz less than 2 dB down
At 500 kHz less than 6 dB down
At 600 kHz at least 6 dB down
I- Channel Bandwidth
At 1.3 MHz less than 2 dB down
At 3.6 MHz at least 20 dB down
Corresponding to these band limits are three ranges of pictorial detail.
Fine details corresponding to video frequencies above 1.3 MHz are reproduced in monochrome; larger areas corresponding to frequencies between
0.5 MHz and 1.3 MHz are reproduced in a two -color orange -cyan system;
and still larger areas, corresponding to frequencies below 0.5 MHz, are
reproduced in a three -color red-green -blue ( full -color) system. This particular division of color reproduction was found to represent a proper comI
Phase Inv
This is
a
Q
-I signal.
Phase Inv
This is
G
R+ G+ B
+Q signal.
Green
Red
I
I
Gain
Q
White Bal
Q
Green
Red
Blue
Blue
I
a
+R +B
Q
Test
(A) I channel.
White Bal
Test
(B)
Q
channel.
Fig. 9 -20. Phase inverters and matrix balance for white.
Gain
THE SUBCARRIER AND ENCODING SYSTEM
365
promise to assure adequate color fidelity under the bandwidth limitations
of the system.
NTSC standards state that the luminance and chrominance signals ( the
respective carrier envelopes as radiated) shall match each other in time
within about half the duration of a picture element, or 0.05 microsecond.
Since the luminance signal and chrominance signals pass through circuits
of different bandwidth, delay circuits are required in the wideband circuits
to bring the respective signals into time coincidence. The time -coincidence
standard has the effect of setting a tolerance in these delay circuits.
Fig. 9 -21A illustrates the relative bandwidths and necessary delay compensation at the encoder. Since the Q channel has the narrowest bandwidth,
it has the most signal delay. Hence the Y and I video channels must be
delayed the appropriate amounts with respect to their assigned bandwidths
to obtain time coincidence with the Q channel.
Fig. 9 -21B shows the bandwidths of the luminance and chrominance signals prior to modulation at the transmitter. Fig. 9 -21C illustrates the resultant signal radiated from the transmitting antenna.
Burst Amplitude and Phase
See Fig. 9 -22A. The amplitude of the burst -key (flag) pulse is set by
the burst -gain control. This pulse is divided into I and Q pulses by the
burst -phase control. The I pulse goes to the stage containing +I video, and
the Q pulse goes to the stage containing -Q video. The Q modulator receives the color subcarrier delayed 90° from the subcarrier feeding the
I modulator. Fig. 9 -22B shows the effect on the modulators. If the amplitude of the pulse is increased while the ratio of amplitudes remains the
same, the resultant vector amplitude (burst amplitude) increases. If the
amplitude remains fixed while the ratio is varied, the phase of the vector
sum (burst phase) changes according to the ratio change. When the ratio
of burst I pulse to burst Q pulse is -tan 33 °, the resultant burst phase is
correctly placed along the -x axis.
Since the keying pulse (burst -flag pulse) is timed on the horizontal back
porch, no video is present, and only the burst of subcarrier appears at the
modulator outputs.
The Modulation Process
The reader should be familiar with the fundamentals of the entire encoding process and the modulation technique from previous study. It remains to examine further the advances in the state of the art with respect
to stable video balance and automatic carrier balance. Such modern processing normally includes closed loops between subcarrier, driver (or processing) circuitry, and the modulator.
Fig. 9 -23 illustrates the paths to be discussed in the RCA encoding system. The path prior to modulation is shown in Fig. 9 -23A. For the color -bar
test position, the camera inputs are blocked by loading relays. (In the Cohu
TELEVISION BROADCASTING CAMERA CHAINS
366
18 MHz
Delay Line
Y
r-- -x,1.3
1.2 µs
MHz
Low-Pass
Delay Line
Filter
0.2µs
0.5 MHz
Low-Pass
Filter
(A) Bandwidths and delays at encoder.
2
0
2
8
4
EY
EQ
6
4
8
1.0
0
2.0
3.0
4.0
5.0
6.0
Video Frequency IMHz)
(B) Prior to modulation at transmitter.
I
;
1
1
3.579545 MHz
.25 MHz
1.0
L
t
c 0.8
d
i
-1.3 MHz-:
S
0. 6
I
F
_
0. 4
0.5+0,5-{
v
Q
I
1
2.0
3.0
I
MHz
g
_
Iti
i
'c &1
ú
1.0
I
iI
MHz
0.2
0
1
4.0
Radiated Signal Frequency (MHz Above Channel Edge)
(C) Signal radiated from transmitter.
Fig. 9 -21. Bandwidths of color signals.
N
5.0
60
367
THE SUBCARRIER AND ENCODING SYSTEM
encoder, a dc bias blocks the camera video.) Fig. 9 -23B illustrates the basic
modulation process in the RCA encoder.
The red, green, and blue video signals that are matrixed into I and Q information may originate either from the camera vidicon channels or from
the bar generator. The monochrome information from the bar signals is
obtained in the matrix module by properly matrixing the red, green, and
blue video signals. In normal film- camera operation, the monochrome information is obtained from the 11/2 -inch vidicon (luminance tube) , and
the monochrome signal is switched straight through the matrix module.
In three -tube camera operation, the matrixing of the red, green, and blue
into monochrome is done at the camera in the control module, and the
encoder treats this monochrome signal the same as when it originates from
the luminance channel.
In the RCA TK -42 studio camera, luminance information is supplied
by the 41/2-inch image orthicon, and chrominance information by three
vidicons. This live camera, unlike the film camera (TK -27) cannot be
operated in the three -vidicon mode. The RCA TK -44A studio camera
employs only three channels, using lead -oxide vidicons.
The I and Q signals are limited in frequency response in accordance
with the FCC transmission specifications. The low -pass filters in the driver
module limit the response of the I signal to approximately 1.5 MHz and
of the Q signal to approximately 0.5 MHz. A delay line in the driver module is used to match the delay of the I signal to that of the Q filter. The
Burst Key (Flag) to +I Modulator Video
I
Burst Phase
Burst Gain
(Adjusts Amplitude Ratio)
(Adjusts Amplitude)
Q
-
Burst Key (Flag) to -Q Modulator Video
L
Burst -Key Pulse
(A)
1
and Q key pulses.
+
+l
Pulse
(B) Vector diagram.
'Burst
Phase
+180°
-Q
Pulse
Fig. 9 -22. Method of generating burst.
330
1
00
TELEVISION BROADCASTING CAMERA CHAINS
368
monochrome signal is passed through the Y -delay module in order to
match the delay of the chroma signals ( Fig. 9 -23A)
The system always provides a monochrome camera signal from the
monochrome -blanker module, even when bars are "punched up" at the
color control panel or when the monitor- module test switch is not on
"operate." In these cases, the monochrome signal from the camera is
switched directly to the monochrome -blanker module, while the monochrome signal matrixed from the color bar goes through the Y-delay
module. Since the timing at the camera is adjusted to take care of all delays, a timing shift will occur at the camera when the monochrome signal
.
Bar Gen
Blanking
RI GI
BI
From {.
Camera
G-r
Y-
I
& Q
Burst
Matrix
B
(Derived)
(True)
Y
Y
H
1
Y
Delay
Q
-y---
Driver
R
-0-
Filters
Flag Regen
Delay Trimmer
1
Input When in Bars or Mono Only
Blanking Input
-.{
Burst -Flag
Input
¡Normal Input For Color Operation
M
Monochrome (Only) Output
Blanke
(A) Processing prior to modulation.
Chroma From
Subcarrier
Modulator
In
Clamp
I
Q
--
Y-11.
I
n
Chroma Loop
Subcarrier Module
360° Phase Shifter
Squaring Ckts
-i
Quadrature
Sampling Pulses
Composite
Color Output
(Luminance Plus
Chrominance)
O
Chroma
Q
I
Driver
Burst
Q Bu
Color
Modulator
Burst
st
Gate
Blanker
Burst -Flag
Input
Y
(True) or Y' (Derived)
1
Limited Chroma
Back to Modulator
Bars Only in
4 -Tube
Camera
(B) Diagram of modulation process.
Fig. 9 -23. Color encoding in RCA four -tube system.
¡
Blanking
Input
I
Q
I
Flag
Q Flag
y
369
THE SUBCARRIER AND ENCODING SYSTEM
Shifted by
Automatic-Carr ier- Balance
I
J
Voltage
i
+1.5 V-
Automatic Carrier Balance
Bu st
r
r --
'
U
0
Chopped at
3.58 MHz
-DC
0VDC
I
Burst
Flag
Switch
Feedback Amp
Jinn__
3.58 MHz
Shift
AAA
Video
Balance
I
I
& Q
Mixed
To
Video Balance
Mixer
Illeedback
Amp
13
Z
Video
Balance
Q
--L_
Feedback Amp
Q
Burst
Flag
Sv.itcn
DC
0
JWL
--
3.58 MHz
-DC
90° From
I
Modulator Base
-
Automatic Carrier Balance
Fig. 9 -24. Chopper -type modulator.
is switched directly to the monochrome -blanker module (as in "Bars On"),
since the delay through the Y -delay module is eliminated. In this case, the
only delay for which the timing circuit must compensate is from the length
of cable between the camera and auxiliary (rack unit) . This distance is
usually short, so if bars are punched up when framing and registration are
being performed at the camera, the timing of the scan with respect to
system blanking will be the same as at the auxiliary outputs. Otherwise,
370
TELEVISION BROADCASTING CAMERA CHAINS
in order to set horizontal centering properly at the camera, the viewfinder
must be switched to observe color in order to see the output timing in
proper relationship to system blanking.
The I and Q signals go to the modulator module ( Fig. 9-23B), where
each of them is chopped to ground at the color -subcarrier rate ( Fig. 9 -24) .
The I and Q chopping signals come from the subcarrier module and are
phased 90° apart. I burst flag and Q burst flag are adjustable in amplitude
with respect to each other by the burst -phase control ( Fig. 9 -22) and are
applied to the chopping transistors mixed with the I and Q video information, respectively. The modulated I and Q are in quadrature and are
mixed in the mixer amplifier (Fig. 9 -25) . The resultant is a vector that
varies in amplitude and phase with changes in I and Q amplitude and
polarity.
The amplitude of the mixed I and Q modulated signals during the absence of I and Q video information (no color information) depends on
the dc output bias of the I and Q amplifiers ahead of the I and Q modulators. This bias (about 1.5 volts) is set by the dc amplifiers that feed the
I and Q amplifiers. The effect of this dc component can be removed by
feeding the mixer with a subcarrier signal equal but opposite in phase to
the resultant of the modulated I and Q dc component. This required subcarrier signal is obtained by shifting the phase of the subcarrier applied to
the base of the I modulator transistor through a resistor -capacitor network (Fig. 9 -24) .
The above phase -shifted subcarrier will balance out the modulator for
the absence of I and Q information, but the resultant at the output of the
mixer will still contain the video information (represented by a low frequency component of the modulated subcarrier), unless the following
video -balance provisions are included (see Fig. 9 -24) . The I and Q video
signals (ahead of modulation) are mixed, inverted, and fed into the mixer
at a level one -half that of the peak-to -peak modulated subcarrier. This removes the video component (or luminance information) to provide a
fully balanced modulated signal at the output of the mixer (Fig. 9 -26) .
The video -balance controls provide a limited amount of amplitude variation of the modulated I and Q signal in order to match the video- balance
correction signal.
The chroma signal then passes through a low -pass filter to remove the
harmonics produced by the square waves of the chopping modulators (Fig.
9-25). One chroma output of the modulator module is mixed with the
monochrome video in the color -blanker module, and the other chroma output is amplified in the driver module (Fig. 9 -27) to be used in the automatic carrier -balance circuitry described below.
During camera blanking, the chroma output should be balanced out to
zero amplitude. If the dc biases of the I and Q amplifiers are not the exact
level that results in cancellation of the subcarrier signal inserted directly
into the mixer through the series RC phase -shift network, subcarrier will
371
THE SUBCARRIER AND ENCODING SYSTEM
I
/
Vector Sum
1
1
Phase-Shift
Net
-----...
I
I
Zero at Black
3.58 MHz
Ist
10.7 MHz
From
Low -Pass
3rd Har
Modulator
Filter
Trap
1st Har
(Fundamental)
2nd
3rd
Har
Chroma Output Amp
3.58 MHz
I
I
I
3rd Har
1st &3rd
/_s.
10.74 MHz
Har'
1
I
1
5th Har
17.9 MHz
I
1st, 3rd,
5th Har
Courtesy RCA
Fig. 9 -25. Filtering of modulated signals.
TELEVISION BROADCASTING CAMERA CHAINS
372
be present during the camera -blanking period. Since I and Q are modulated in quadrature, the dc inputs to the I and Q amplifiers can be adjusted
to cancel completely any subcarrier at the output, regardless of amplitude
and phase. For this purpose, I and Q black -balance (carrier balance) controls are provided. To maintain this balance, other dc sources to the I and
Q amplifiers are obtained (Fig. 9 -28) These sources are proportional to
the I and Q components of any subcarrier present at the output during
.
l
-
--
Video Component
150% of
Video
Modulated With Carrier Balanced Out
Modulated Signal)
Video -Balance Signal for Cancellation
11/2 Amplitude of Original Video(
Fully Balanced Modulated Signal
Representing the Vector Sum of &
(Chrome With No Luminance)
I
-
J
Q
Courtesy RCA
Fig. 9 -26. Removal of video component.
373
THE SUBCARRIER AND ENCODING SYSTEM
camera blanking. The loop gain of this feedback system is made high to
maintain a high degree of carrier balance.
The chroma amplifier in the driver module has a high gain for low level inputs, but clips when the output exceeds about 1.4 volts peak -topeak. Only the low -level information during camera blanking is of interest,
To
Chroma Amplifier
Driver Module To Be Amplified
Driver Module
for Automatic- Carrier-Balance
Sampling and Detection
From Mixer
and Filters
Delay
Limiting
Amp
To
Color Blanker
With
Monochrome
To Be M xed
Back to Modulator
Module for Detection
During Blanking
i
Limiting
Amp
J
(A) Block diagram.
Blanking
r
."
-
vvv\-^",\-
^v^VUI w
Balanced
Chroma Input to Limiters
Limiters
+0.7
V
-0,7
V
+0
7
V
0.7
V
"
Output
w
`
Prevents Rectification
of Signal
`
y First Limiting Amplifier
(Same Circuit as Used for
3.58-MHz Square Waves)
Second Limi ing Amplifier
Input
(C) Limiter waveforms.
(B) Limiting amplifier.
Fig. 9-27. Driver module.
I and Q detectors in the modulator module are keyed on only
during a portion of the camera -blanking interval. Several cycles of sub carrier occur on the I and Q keying signals, each phased such that the I
detector is switched on in phase with the peak of the I component of the
amplified chroma signal and the Q detector is switched on in phase with
the peak of the Q component (Fig. 9 -29) Since the I and Q components
are 90° apart, the peak of the I component occurs during the time the Q
component is going through zero, so no Q information will be detected by
the I detector, and vice -versa. When the detector is keyed on, if the ampli-
since the
.
TELEVISION BROADCASTING CAMERA CHAINS
374
fled chroma signal is anything except zero or passing through zero, the
capacitor is charged to a level either more positive or more negative than
the dc component of the chroma signal at the detector. This charge is held
throughout the line, amplified, and applied to the respective I and Q amplifier input in proper amplitude to balance out any subcarrier that is
present in the output during camera blanking.
The inverted I and Q signal used for video balance is also used in the
matrix portion of the modulator module to obtain the color-difference signals, Y' R, Y' G, and Y' -B (Fig. 9 -30) When the luminance signal
is removed from these color -difference signals in the detector module, by
combining with the Y signal from the camera, the proper R, G, and B signals are obtained for NAM monitoring and automatic control (described
previously)
Built -in setup facilities are obtained by operation of the test switch on
the monitor module. In all positions except position 1 (Operate) , the color
bars are switched into the matrix module. In position 2, all inputs to the
matrix module are tied together. If the bar generator is then pulled, there
is no color information, and no subcarrier should be present at the output
(except during burst) The black -balance controls on the modulator module can then be adjusted to remove any subcarrier present. If the bar generator is now inserted while the inputs are tied together, all inputs will be
of equal amplitude, and there should be no I or Q signal. The adjustments
to cause I and Q to cancel out under these conditions are the I and Q
white -balance controls in the matrix module. These controls adjust the
levels of the inverted red and inverted green signals, respectively (described previously).
-
-
.
.
.
I
T
Courtesy
Fig. 9 -28. Feedback and detection loop.
RCA
375
THE SUBCARRIER AND ENCODING SYSTEM
Sampling Region
Amplified Chroma From Limiting
Amplifiers in Driver Module
I
Sample
Q
Sample
-1111771117
CI
Occurs During Blanking
C2
I
Error Signal
2
Q
Error Signal
WC)
(DC)
(A) Diagram.
I`/'
I
I
I
V
Error Onlyl
,
I
Sample
I
r--I--I
J
11
;
I
I
v
Black Portion Expanded
I
Q
Sample
Q
Error Only
I
Y.
Error in Both 18
/
(
Q
B) Waveforms.
Courtesy RCA
Fig. 9 -29.
I
and
Q
detectors.
The above procedure causes the subcarrier to be cancelled out whenever
the red, green, and blue inputs are equal in amplitude. If subcarrier appears at the output in test position 3, which disconnects the inputs, it indicates that the three bar amplitudes are not equal. They should be equal
during the white bar. Green -gain and red -gain controls are provided in the
bar-generator module to set these pulses equal to the blue pulse during the
white bar.
376
TELEVISION BROADCASTING CAMERA CHAINS
Test position 4 produces the same chroma presentation on the CRO
that position 3 produces, except that a low -pass filter is used so that any
video- frequency information in the chroma signal can be detected and
balanced out with the video-balance controls in the modulator module.
Test positions 5 and 6 (I only and Q only) have no adjustments associated with them. They are useful for troubleshooting or double- checking
video balance and the quadrature relationship.
Test position 7 sets up a method for making quadrature adjustments
with only the CRO display. The Q channel operates in a normal manner
with bars applied, but the I channel has applied to it a square wave that
goes equally positive and negative around its clamped blanking interval
(Fig. 9 -31A). The CRO then displays a few lines of the resultant of the
Q signal with the positive -I portion of the square wave, and then a few
lines of the resultant of the Q signal with the negative -I portion of the
square wave. Since the positive -I and negative -I levels are equal, these two
resultants will be equal only when I and Q are phased 90° apart (Fig.
9 -31B) Since the CRO displays these two resultants superimposed on
each other, it is easy to compare the amplitudes and make a quadrature
adjustment so that the two CRO presentations become equal.
Test position 8 provides the CRO with the composite color-bar presentation. It is used to set the I and Q signal gains relative to monochrome to
obtain the levels shown in Fig. 9 -32.
.
+1 Amp
To Detector Module
Three Color Tubes Operating on Color
Brightness Only
(luminance) tube provides luminance
information that can be different from
chroma brightness.
Y
NAM & Auto BlacklAuto White
Made From Receiver Matrix
IR- V'I
1G-Y'I
1B-Y'I
+YR
+YG
+YB
V'-R
Y'-G
Y'-B
-0.961 - 0.62Q
-1R-V'
0.271+0.65Q- -1G-V'1
1.111+ 1.70Q
-1B-`0I
Courtesy RCA
Fig. 9 -30. Method of obtaining color- difference signals.
377
THE SUBCARRIER AND ENCODING SYSTEM
The color -bar patterns of Fig. 9 -32 are all those likely to be encountered
in various color camera -chain facilities. These are normally different from
the standard split -field pattern. Note that in addition to the IEEE units, a
voltage scale, such as would be used on an external oscilloscope, is shown.
In Fig. 9 -32A are 100 -percent bars (used only on very special occasions) .
Fig. 9 -32B shows the most usual 75-percent mode, and Fig. 9 -32C illustrates
the white 100 -percent mode. A standard 7.5- percent setup level is used.
Test position 9 is for setting the proper burst phase with the CRO in a
manner similar to that used for the quadrature adjustment. To obtain a reference that is phased 90° from burst, a precision voltage divider is used to
reduce the amplitude of the blue bar an amount that shifts the cyan bar
(third bar) to a phase position 90° from where burst should be (Fig.
9 -33A). This reduction of the blue amplitude shifts the phase of all
the bars (and unbalances the white bar) , but we are concerned only with
the third bar in this test. At the same time, the burst is switched from the
burst -flag width to full active scan width and is caused to go equally
positive and negative, several lines each, by the same blanked square wave
Rate
1
7875 kHz 1112H1
AC Axis After
Signal During
Position-7 Test
L
Passing Through
Large
--2
Lines
2
Lines
--I
--
I Signal After
Clamp Insertion
C 1Equal + & -1
0V
(A) Waveforms.
Q
Modula ed Signal
Resultants Unequal When
I & Q Not in Quadrature
Resultants Equal Only When
& Q in Quadrature
I
(B) Vector diagrams.
Courtesy RCA
Fig. 9 -31. Method of making quadrature adjustments.
378
TELEVISION BROADCASTING CAMERA CHAINS
-
140
W
120100-
C
131
10.911
10.911
c 60-
48
0-
-20-
10.651
10.431
59
(0.411
46
10.101
10.251
7.5
10.051
-9
To
40
I
1-0.281
(
0.141
NOTE
7.5
771.2i15)
1-0.061
-
18
-23
-0.161
1
1.0
0.8
0.6
---0
0.4
35
10.321
14
Setup
93
62
10.341
20
(0.141
20-
100
(0.701
72
'7.71
-
B
116
89
10.631
W 40
MR
G
10.811
100
10.701
80
-40
Y
131
0.2
-23
-0.161
--0.2
Numbers in parentheses are voltages.
(A) 100- percent bars.
W
1401201C0-
77
80
10.541
'60,,,
-
4020-
20
10.141
Y
C
100
100
(0.701
10.701
10.48)
56
38
10.391
77
46
28
10 051
-0.031
(
-20
-40-
1-042811-0.141
NOTE:
15
(0.201
- - - -- 10,111.
7.5
0.8
0.6
(0.321
(0.251
13
1.0
-0.4
10.501
36
1
-_
-
48
10 341
0.091
-
B
72
10.541
(0.271
Setup
R
89
10.621
69
0
20
M
G
-15
-0.111
7.5
10.05
-15
(
-0.111
0.2
- -0
-
-0.2
Numbers in parentheses are voltages.
(B) 75- percent mode.
120
100
-
W
Y
C
100
100
100
(0.701
(0.701
(0.701
MR
G
-
W4020
-L
0-
-20-40-
56
38
(0.391
20
(0.141
77
_
.r
J
20
1-04281 10.141
10.54)
72
46
(0, 321
(0.25110201
13
(0.09)
15
7.5
10,051
(0.111(
-0
1
-0.111
7.5
(0.051
-15
-15
031
1
1.0
0.8
0.6
-0.4
-
10.501
36
10.271
Setup
-
89
69
10.481
-
B
-0.111
NOTE: Numbers in parentheses are voltages.
(C) White 100 -percent mode.
Fig. 9-32. Nominal encoded color -bar signal levels.
-
-0.2
-0
--0.2
379
THE SUBCARRIER AND ENCODING SYSTEM
used for the quadrature test. The two superimposed signals in the CRO
then represent the two resultants of the modified chroma combined with
the positive burst for several lines and then with the negative burst for
several lines. Only when the burst phase is adjusted so that the two resultants are equal during the third bar (Fig. 9 -33B) , will the burst be in
quadrature with the third bar and properly phased.
Blue Reduced
in Amplitude
Burst
Blue
(A) Shift of cyan.
/
Green
New "Cyan
Cyan
Bur sl
Alternate
+ & -
Neg
Burst
New
Burst
Burst
---
"Cyan"
Resultants Equal When
& New "Cyan" in Quadrature
New
"Cyan"
Burst Not in Proper Phase
(B) Phase checks.
Courtesy RCA
Fig. 9 -33. Method of setting burst phase.
9 -12. THE VECTORSCOPE
Fig. 9 -34A illustrates the Tektronix Type 520 vectorscope, and Fig.
9 -34B illustrates the normal vector display with the selector switch in the
vector position. This instrument is designed to measure luminance, hue,
and saturation of the NTSC composite color television signal. Self-cancel-
ling push- button switches permit selection of displays for analysis of television- signal characteristics, and to check vectorscope calibration.
Dual inputs provide time -shared displays for comparison of input output signal phase and gain distortion. A chrominance channel demodulates the chrominance signal to obtain color information from the com-
380
TELEVISION BROADCASTING CAMERA CHAINS
posite video signal for use in vector, line -sweep, R, G, B, I, Q, differential gain, and differential -phase displays. A luminance channel separates and
displays the luminance (Y) component of the composite color signal. The
Y component is combined with the output of the chrominance demodulators for R, G, and B displays at a line rate. A digital line selector permits the display of a single -line vertical. interval test signal from a selected
line of either field 1 or field 2.
(A ) Instrument.
(B) T ypical display.
Courtesy Tektronix, Inc.
Fig. 9 -34. Tektronix Type 520 NTSC vectorscope.
381
THE SUBCARRIER AND ENCODING SYSTEM
The following basic description, supplied through the courtesy of Tektronix, Inc., serves as an excellent review of the color system for the
student:
In color television, the visual sensation of color is described in terms
of three quantities: luminance, hue, and saturation. Fig. 9 -35 shows a
conical representation of these concepts. Black- and -white (monochrome)
TV receivers respond to the brightness (or luminance) signal only. Color TV receivers respond to all three signals, luminance, hue, and saturation.
Hue and saturation together are called the chrominance signal. With the
VECTOR button depressed, the vectorscope displays the hue and saturation
quantities. With the Y button depressed, luminance amplitude is displayed.
Green
Cyan
Yellow
Red
Blue
Purples
Fig. 9 -35. Conical representation of
color concepts.
Achromatic Colors
(Grays)
f
Black
Courtesy Tektronix, Inc.
Luminance is brightness as perceived by the eye. Because the eye is
most sensitive to green and least sensitive to blue light of equal energy,
green is a bright color, and blue is a dark color as conveyed by the luminance signal to monochrome TV receivers. Color -TV receivers utilize the
luminance signal to produce both monochrome and color pictures.
Chrominance consists of two additional quantities: hue and saturation.
Hue is the attribute of color perception that determines whether the color
is red, blue, green, etc. White, black, and gray are not considered hues. Hue
is presented on the vectorscope CRT as a phase angle and not in terms of
wavelength. For example, red (which has a wavelength of 610 nanometers)
is indicated as 104° on the standard color -phase vector diagram (Fig. 9-36)
and the vector graticule (Fig. 9 -37).
Saturation is the degree to which a color (or hue) is diluted by white
light in order to distinguish between vivid and weak shades of the same
hue. For example, vivid red is highly saturated and pastel red has little
saturation. On the vectorscope, saturation is represented by the radial distance from the center (where zero saturation exists) to the end of the
color vector (where 75- percent or 100 -percent saturation exists for a particular color). If the burst- vector amplitude corresponds to the 75- percent
382
TELEVISION BROADCASTING CAMERA CHAINS
marking (Fig. 9 -37), the colors are 75- percent saturated. If the burst vector amplitude corresponds to the 100 -percent marking, the colors are
100- percent saturated.
In an NTSC color-television transmission system, the hue information
and saturation information are carried on a single color subcarrier, at
3.579545 MHz. These signals, in modulated -subcarrier form, are called
chrominance. The hue information is carried by means of amplitude
modulation with the subcarrier suppressed. A subcarrier that supplies
phase information is required for demodulation. No chrominance signals
are present during the horizontal-blanking interval, and a sample of the
subcarrier (called burst) is provided within this interval
To recover the hue information, phase demodulators are employed in
the vectorscope. The phase reference is a color subcarrier that is regenerated
by an oscillator in the instrument. The oscillator is locked in both phase
and frequency to the color -burst signal. When the VECTOR button is
pressed, the vectorscope displays the relative phase and amplitude of the
chrominance signal on polar coordinates. To identify these coordinates,
the vector graticule (Fig. 9 -37) has points that correspond to the proper
phase and amplitude of the three primary colors, red (R) , green (G) ,
.
Red
-
R-Y
90°
104°
Magenta
61°
Courtesy Tektronix, Inc.
Fig. 9 -36. Vector relationship among color components.
383
THE SUBCARRIER AND ENCODING SYSTEM
Inner
Corners of Outer Box Outline
+ 10°
for
and
+ 20%
Box
Indicates
+
2.5° and
+2.5- IEEE -Unit Area
Targer Area
Red Vector
L
MG
N/
R -Y
Yi
Axis
r
L
180°
33° Phase Angle
for +Q Axis
®
Burst -Vector
%
J
toox
75i
Position
B-Y Axis
T
^
75 %and 100%
Amplitude Markings
for Color -Burst Vector
®
Reference
J
Minor Degree
Markings at
2° Intervals
G
®
Major Degree Markings
at 10° Intervals
Calibrated-Test -Circle
Amplitude Reaches This
Outer Ring Marking
1
IL
0° Color -
Suocarrier
11{1111;]_11
L
Courtesy Tektronix, Inc.
Fig. 9 -37. Internal vector graticule of Type R520 vectorecope.
and blue (B) . In addition, the complements of the primary colors are
indicated as follows: cyan (Cr), magenta (MG) , and yellow (YL) .
Any errors in the color- encoding, video-tape recording, or transmission
processes that change these phase and /or amplitude relationships cause
color errors in the received picture. The polar- coordinate type of display
has proved to be the best method for portraying these errors.
The polar display permits measurement of hue in terms of relative
phase of the chrominance signal with respect to the color burst. Relative
amplitude of the chrominance with respect to the burst is expressed in
terms of the displacement from the center (radial dimension of amplitude) toward the point that corresponds to 75- percent (or 100 -percent)
saturation of the particular color being measured.
The outer boxes around the color points correspond to phase and amplitude error limits set by FCC requirements ( ±10°, ±20 percent). The
inner boxes indicate ±2.5° and -}2.5 IEEE units. These limits correspond
to phase and amplitude error limits set by EIA specification RS -189,
amended for 7.5- percent setup. Fig. 9 -38 shows the purpose of the small
marks that intersect the I and Q axes.
The two major distortions to which the chrominance signal is subject
are differential gain and differential phase. Differential gain is a change in
color -subcarrier amplitude resulting from a change in the luminance signal
while the hue and saturation of the original signal are held constant. In the
384
TELEVISION BROADCASTING CAMERA CHAINS
reproduced picture, the saturation will be distorted in the areas between
the light and dark portions of the scene.
Differential phase is a phase change of the chrominance signal that results from changing the luminance signal while the original chrominance
signal is held constant. In the reproduced picture, the hue will vary with
scene brightness. Differential gain and differential phase may occur separately or together. The causes of these distortions are chrominance nonlinearities caused by luminance -amplitude variations. To measure differential phase using the Type R520 vectorscope, no graticule is needed. Instead, the trace overlay and slide -back technique using the CALIBRATED
PHASE control provides the means for performing the measurement.
The IEEE graticule is used primarily for measuring differential gain and
video-signal amplitude. For the measurement of video-signal amplitude,
the graticule is marked in IEEE units. In standard TV practice, 140 IEEE
units equal 1 volt. Hence, with the aid of the IEEE graticule, the corn-
(
I
-1
1111
-Component Axis
V
J
Axis
Q
M
vL
L
J
100
f
T
1
75%
r
B
r
v
'i2Fr
-
--
-
L
Dash line with arrow associates the color vectors with the I -axis markings.
If the Q component of the signal is absent, linear color vectors will move to
associated markings on the
I
their
axis.
Short arrows along the Q axis are intended to show the relationship between the Q -axis
markings and the associated color vectors.
If the I component of the signal is absent, linear-color -vector points will appear at the
appropriate marks along the Q axis.
Courtesy Tektronix, Inc.
Fig. 9 -38. Relationships between standard color -phase vector diagram and
Type -R520 vector graticule.
THE SUBCARRIER AND ENCODING SYSTEM
385
posite video signal will be exactly 1 volt in amplitude when the equipment is adjusted to obtain a display amplitude of exactly 140 IEEE units.
Next, the graticule is used as a guide for checking and adjusting the composite video signal for the following typical proportions:
1.
The white level should correspond to the graticule marking for +100
IEEE units.
The reference black level should correspond to the 7.5- percent setup
marking.
3. The blanking level of the video signal should coincide with the
graticule line for 0 IEEE units.
4. The sync-pulse amplitude should correspond to the graticule line
for -40 IEEE units.
2.
Adjustments to the encoder can be made with the Type R520 vector scope. By use of the vector mode, the I matrix may be checked for accuracy by turning off the Q channel and observing that all six dots align with
the cross marks along the I axis of the vector graticule.
Check that the 100%-75%-MAX GAIN switch is set to 75% and the
GAIN control for the channel is set to "cal." Set the I chroma gain on the
encoder so that the largest-amplitude (red and cyan) dots align with the
I -axis cross marks. All the other dots should align with the appropriate
marks, also.
Some typical causes of I -axis dot misalignment are incorrect matrixing
from the R, G, and B signals to the I signal; value changes of matrix resistor (s) ; incorrect gain of an inverting amplifier associated with the
matrix function; nonlinear amplification of the matrixed signals; or nonlinear amplification of the doubly balanced I modulators.
The remarks given for the I channel apply equally to the Q channel.
After the correct operation of the I or Q channel has been established
independently, with the other channel temporarily disabled, the quadrature phasing between the I and Q channels is facilitated as follows:
Turn on both channels in the encoder.
Adjust the encoder quadrature phasing and the vectorscope A PHASE
(or B PHASE) control so that all color dots lie within their respective
inner boxes on the vector graticule.
3. Burst phasing of the encoder then may be adjusted so that the burst
vector is at exactly 180° on the vector graticule.
4. The amplitude of the burst vector may be set in the encoder.
5. The luminance levels in the encoder now can be set to their correct
amplitudes by using the Y (luminance) mode of operation.
1.
2.
The saturation of the displayed colors can be checked by using the 75percent and 100 -percent burst -amplitude markings on the vector graticule
386
TELEVISION BROADCASTING CAMERA CHAINS
and by noting the position of the 100 %- 75 % -MAx GAIN switch. The
general procedure for making this check is as follows:
1.
While obtaining a normal vector display, check that the
GAIN con-
trol for the channel is set to "cal."
2. Set the 100 % -75% -MAX GAIN switch to a position that causes the
displayed color vectors to appear within the target areas. Note the
location of the burst tip along the 180° axis on the vector graticule
and the position of the 100 % -75% -MAX GAIN switch. If both the
burst -tip location and the switch position indicate 75 percent, the
colors are 75-percent saturated. If the burst -tip location and the
switch position indicate 100 percent, the colors are 100- percent
saturated.
Differential Gain
Differential gain is a change in color -subcarrier amplitude as a function of luminance. In general, any differential gain present in the signal
can be checked by using the differential -gain mode of operation of the
vectorscope and by setting the 100 %- 75 % -MAx GAIN switch to Max
Gain. With a standard 10 -step linearity staircase signal (Fig. 9 -39) applied to the vectorscope (through the equipment to be checked) , any differential gain present will cause a variation in the segment levels.
110 IEEE
Units
100 IEEE
Units -
--
50
Ill
20
1
20 IEEE
Units
o
Fig. 9 -39. Standard modulated stair -step signal.
387
THE SUBCARRIER AND ENCODING SYSTEM
JA
":
7.s_.
10%
-R--
8
60
6
40t.
20
7.5
-40
Courtesy Tektronix, Inc.
Fig. 9 -40.
Differential -gain presentation when modulated
stair -step signal
is
applied.
The major divisions of the IEEE graticule represent percent of signal
gain or loss when the displayed 100 -unit level coincides with the 0-percent
graticule lines. When the right -hand graticule marking for the scale is
used in conjunction with the major and minor graticule markings (Fig.
9 -40), differential -gain measurements can be made within an accuracy of
1 percent. Fig. 9 -40 is an example of a differential -gain presentation using
a modulated stair -step signal. The right side of the graticule is marked in
percent of gain distortion. From white to black luminance levels the indicated gain distortion is 3 percent.
Differential Phase
Differential phase is a phase modulation of the chrominance signal by
the luminance signal. As a result, the hue in the reproduced color picture
varies with scene brightness. Differential phase is read from the calibrated
phase -shift control. Approximately 1 inch of dial movement represents 1°
of phase shift. The vertical deflection of the display is greatly magnified
and inverted on alternate lines; this allows the use of a trace -overlay technique and the slide -back method of measuring small phase changes. The
CALIBRATED PHASE control provides direct readout of differential phase.
When the standard linearity test signal is used, differential phase of 0.2°
can be measured.
Fig. 9 -41 shows an example of a differential -phase presentation using a
modulated stair -step signal. A trace -overlay technique provides excellent
TELEVISION BROADCASTING CAMERA CHAINS
388
CALIBRATED
PHASE
(A) First step overlayed.
(B) Fifth step overlayed.
Courtesy Tektronix, Inc.
Fig. 9 -41. Differential -phase presentation using modulated stair -step signal.
resolution for measuring small phase changes. The change from the reference point in Fig. 9 -41A (first step of stair -step signal overlayed) to the
point of measurement in Fig. 9 -41B (fifth step overlayed) represents 1.4°
of differential -phase distortion.
R, G, B,
and Y Observations
The Type 520 vectorscope provides a luminance channel that permits
the separation and display of the luminance (Y) component (Fig. 9 -42A)
of the composite color signal. The Y component also can be combined with
the outputs of the chrominance demodulators for R, G, and B displays at
a line rate (Figs. 9 -42B, 9 -42C, and 9 -42D). Amplitude measurements of
color-signal components can be made with an accuracy of 3 percent.
A great deal of valuable information can be obtained by this line -sweep
presentation of Y, R, G, and B. To take full advantage of such measurements, see Fig. 9 -43 and Table 9 -1. These review the color- bar -generator
output pattern.
Note from Fig. 9 -42 that the decoded R, G, and B video signals for an
encoder that is properly set up have the same maximum amplitude and
minimum amplitude per step. The latter falls on the 7.5- percent setup level.
All encoders, whether four- or three -channel cameras are used, convert
R, G, and B to luminance for color bars. In the three- channel system, the
same is true for the camera signals.
Fig. 9 -44 shows errors that can be displayed by line -sweep presentations
of decoded R, G, and B video on the vectorscope. Errors of this type
occur in the matrix that converts the R, G, and B signals to luminance.
There may be a change of actual values of matrix resistors, or there may
be an improper setting of an adjustable value.
Fig. 9 -44A shows decoded red video. Note that the first four steps are
incorrect in value and ( from Table 9 -1) that green is the common factor.
389
THE SUBCARRIER AND ENCODING SYSTEM
Gray Yellow Cyan Green Magenta Red Blue
4
2
-7
m
25-"==
0 w}e
..
-6
-8
_.
-
(A) Luminance.
(B) Red.
!A
so
.-
40
a..__
,.
__...._..........
20 _......__..
........
.
7S - -
._
d4
199+
10 %
e0
r-
90
e
6
so
6
_---.
4
4
2
40
-___ -.2
20
_._.-
.
....._
-10%
..
4
--._ .............._....._,_..._.
7.5_
_-SL..r_.._
.
..........
-2
-4
0
6
_o
-29
.
-40
(
(D) Blue.
C ) Green.
Courtesy Tektronix, Inc.
Fig. 9 -42. Line -sweep
presentations of
Y, R, G, and
B
decoded video.
Therefore, an error in the green portion of the luminance matrix results
in approximately the same error in the first four steps of the red or blue
decoded video display.
NOTE: It is helpful for the technician to memorize the actual sequence
of the color -bar pattern: gray, yellow, cyan, green, magenta, red, and blue,
and also the primaries used to form yellow, cyan, and magenta.
Fig. 9 -44B shows decoded green video. Steps 1, 2, 5, and 6 have an error
of approximately the same magnitude. Note from Table 9 -1 that red is
now the common factor, and an error in the red part of the luminance resistive matrix is indicated.
390
TELEVISION BROADCASTING CAMERA CHAINS
Green
Red
Blue
0
0
®
®
©
I
0
I
I- White -I
l--Black-
Yellow
Cyan
Green
Magenta
I
Red
I
Blue
Fig. 9 -43. Color -bar generator outputs.
Table 9 -1.
Steps
Hue
Color -Bar Pattern
Common Factor
Primaries
+G +B
1
White
2
Yellow
R
3
Cyan
G +B
4
Green
G
5
Magenta
6
Red
R
7
Blue
B
8
Black
0
R
R
+G
+B
Green
1
)}
Red
Blue
391
THE SUBCARRIER AND ENCODING SYSTEM
0 0
C)
0
Ref
(A) Red decoded output, error
in green channel.
o_o
-%
-
7.5%
r
10
(First lour steps are higo.
i
Ref
(B) Green decoded output,
error in red channel.
--7.5%
Steps I. 2, 5, and 6 nave
0
error
ol0
of
approximately the same magnitude.
-
Ref
0
(C) Blue decoded output, error
in luminance -chrominance
amplitude ratio.
ö.
----0-----2.5s
Fig. 9-44. Luminance errors in resistive matrix of three- channel encoder.
TELEVISION BROADCASTING CAMERA CHAINS
392
Fig. 9 -44C shows a type of distortion more likely to occur in the video
distribution and switching system following the encoder. Remember that
the decoded output of the blue channel is made up of only 11 percent luminance. Thus, the majority of the signal contributing to decoded blue is
chrominance. Therefore, the decoded blue signal is most sensitive to errors
in the luminance -to- chrominance ratio. The downward slope shown in
Fig. 9 -44C indicates that the chrominance is low in relation to luminance.
A downward slope can indicate either that luminance is excessive or that
chrominance is deficient. Bear in mind that the first step (gray or white)
is all luminance and zero chrominance. Thus, noting the amplitude of the
first blue bar on the left indicates whether the luminance signal or the
chrominance signal is in error. Then the slope of the decoded blue indicates the ratio of luminance to chrominance. An upward slope indicates
that chrominance is excessive in relation to luminance.
Phase vs Time Delay
The time delay between two signals can be checked, because the phase
difference at any particular frequency can be related to time difference.
An example of this is the setup of two color cameras some distance apart.
With the outputs of the cameras connected to the inputs of the vectorscope, and the proper push buttons-cH A, CH B, FULL FIELD Aga, and
VECTOR-depressed, the two signals can be viewed together on a timeshared basis. Any time -delay difference between the two camera links will
appear as a phase difference in the vector display.
This time -delay difference can be determined by noting that 360° on the
graticule equals 280 nanoseconds of time. The difference can be minimized
by adjusting the connecting -cable lengths so that there is no hue or phase
difference from one camera to the other.
NOTE: Detailed setup techniques for the encoding system are presented
in Harold E. Ennes, Television Broadcasting: Equipment, Systems, and operating Fundamentals (Indianapolis: Howard W. Sams & Co., Inc.,
1971) Since the present chapter has covered more advanced details of
encoding, it will be helpful for the student to review Chapter 10 of that
volume at this time.
.
EXERCISES
Q9 -1.
Q9 -2.
Q9 -3.
Q9 -4.
Does a "locked oscillator" require a trigger input to oscillate?
If you notice dots travelling from top to bottom (instead of bottom to
top) of a picture monitor when viewing a color signal, what does this
indicate?
You are adjusting counter chains before setting the subcarrier frequency. Will this frequency adjustment upset the counter adjustments?
In the answer to question 3, why is it stated that the subcarrier countdown is 113.75?
THE SUBCARRIER AND ENCODING SYSTEM
Q9 -5.
Q9 -6.
Q9 -7.
Q9 -8.
Q9 -9.
393
Does the burst flag have any effect on the phase of the color -subcarrier
sync burst?
How would you measure the breezeway?
If you find it impossible to eliminate the flag pulses for the entire 9H
interval, and you have replaced tubes with several new tubes to no
avail, what are the most likely sources of the trouble? (Assume the
circuit of Fig. 9-11.)
If you want to use 9 cycles of color -sync burst, how many peaks should
you count on the interlaced color -burst pattern?
If Q2 of Fig. 9 -13 should become open, how would this affect the burst
flag?
Q9 -10. In the encoder, with the Y channel (luminance, or monochrome,
channel) turned off and with all inputs tied together on a color -bar
signal, what pattern would you expect at the output with I and Q on?
Q9 -11. In all instances where we have stated that the Y, I, or Q channel is on
or off, exactly what signal is controlled?
Q9 -12. What does a white -balance control actually do in the circuit?
Q9 -13. Assume you are looking at the encoder output under these conditions:
scope in the wideband position, standard (100 percent) bar input
signal, IEEE scale calibrated for 1 volt (peak -to -peak) from -40 to
plus 100 IEEE units, encoder gains properly adjusted for 0 -100 units
for white pulse, I and Q gains and gain ratio properly set. What is
the maximum peak -to -peak value you should read for the following
conditions:
(A) Y and Q channels off, I channel on.
(B) Y and I channels off, Q channel on.
Q9 -14. When sync is inserted in the encoder, why must it be inserted after
aperture compensation and before Y delay?
Q9 -15. When the vectorscope is in the vector display mode, is luminance value
indicated?
CHAPTER
10
Color Picture Monitoring
Systems
NOTE: It is imperative for the student to review NTSC color fgndamentals at this time. See Harold E. Ennes, Television Broadcasting: Equipment,
Systems, and Operating Fundamentals ( Indianapolis: Howard W. Sams
& Co., Inc., 1971 ) , Chapter 2 (in particular, the text associated with
Figs. 2 -36 through 2 -44 concerning color receivers and monitors).
The major divisions of a color monitor are shown in Fig. 10 -1. Note
that the primary difference between a monitor and a conventional receiver
is that the monitor does not have rf and i -f stages. In the monitor, the composite color signal (Ea,) is fed (normally from a distribution amplifier)
directly to the decoder circuitry. The major difference between types of
decoders used in color monitors is that some employ wideband color ( I
and Q) demodulation, whereas others use narrow -band color -difference
(or X -Z) demodulators.
As shown by Fig. 10 -1, if the color monitor receives E,,, without inserted
sync, an external sync drive is required. If the color signal has inserted
sync, the monitor normally is operated on the internal sync position, using
sync separated from the video itself.
10 -1. ANALYSIS OF BASIC
I
-AND -Q DECODER
The composite color signal (Es,) is connected to J1 or J2 (Fig. 10 -2),
and the opposite jack is terminated unless loop- through is made to another
decoder or terminated line. Signal E,,, is amplified by V1. The output of
VI feeds the sync separator (V3) and, through a 1 1./..s delay line, the Y
amplifier (V2) . Delay is necessary because the bandwidths of Y, I, and Q
all differ, and consequently the channels have different delay characteristics.
Since Q has the narrowest bandwidth, the delay is greatest in this channel,
and the Y and I signals are delayed accordingly. Time differences greater
than 1/2 picture element (approximately 0.05 µs) in the region close to
394
COLOR PICTURE MONITORING SYSTEMS
395
the subcarrier frequency, and greater than one picture element elsewhere
in the spectrum cause luminance and chrominance errors. Therefore, the
delays in Y and I are used to cause time coincidence of all three channels
for the final matrixing process.
Amplifier V2 is termed a Y amplifier because a burst trap is incorporated
to suppress a considerable region around the subcarrier frequency. As previously explained, the subcarrier is an odd multiple of one -half the line
frequency as well as an odd multiple of one -half the frame frequency, to
minimize beats with luminance video components. However, this holds
true only for still pictures. When motion is present in the scene, a small
amount of beating takes place. By suppressing the region of strongest
chrominance information in the Y channel, such interference is made
negligible.
Composite
Color -Signal
InputlEml
Sync Input
(Optional)
-a
-y
Red
Decoder
Red
Green
Video- Output
Amplifiers
Blue
Green
Color
Kinescope
Blue
Sync
Conve gence
Circuitry and
Control Panel
Deflection
High Voltage
Regulated
Power
Supply
Deflection
High Voltage
Fig. 10 -1. Major divisions of a color monitor.
The encoded signal input also feeds bandpass amplifier V6 through the
chroma control. Transformer T2 passes only those frequencies between
approximately 2.5 MHz and 4.2 MHz, where the chrominance sidebands
are present. These signals are passed to the control grids of synchronous
demodulators V7 and V8. The color -sync burst is eliminated from stage V6
(in the particular circuitry under discussion) by a negative keying-out
pulse from the monitor flyback transformer.
The sync -separator stage (V3) strips conventional sync from E,,, and
delivers it to the deflection input channel for the color monitor. A small
capacitor in this same stage serves as a color -burst takeoff for delivery of
separated burst to amplifier V4. This stage is keyed on at the burst interval
following horizontal sync by a positive pulse from the monitor flyback
transformer. At other times, the burst amplifier is biased beyond cutoff. A
hue capacitor at the grid of V4 adjusts the phase of the burst sine wave to
the in -phase condition (I signal) for control of the local oscillator.
396
TELEVISION BROADCASTING CAMERA CHAINS
á
t
t
E
r
LL
d
C
t
® .__r
a
E
E
O
A
OO
..
ó
_
ó
m
;3
c
C
E
Fig. 10 -2. Basic block diagram of l -Q decoder.
COLOR PICTURE MONITORING SYSTEMS
397
Phase discriminator V5 is the heart of the automatic phase-control
(APC) loop. This discriminator receives the amplified color burst from
V4, and the local- oscillator signal through phase -amplifier stage V13. Any
phase difference between the two signals results in a dc voltage change
that controls the mutual conductance of reactance tube V11. This tube, in
turn, controls the phase of the crystal oscillator (V12A). When no burst is
present in the applied signal (monochrome), a control voltage in stage
V5 drops below a reference potential and causes the color killer (V 12B )
to bias the chroma circuit below cutoff.
Synchronous demodulators V7 and V8 are gated by the reinserted sub carrier voltage. Grid 3 of V8 receives the in -phase (I) subcarrier voltage.
Grid 3 of V7 receives the quadrature (Q) subcarrier voltage through the
quadrature- adjust transformer. Because of the 90° phase relationship, V8
is gated on when V7 is off, and V7 is gated on when V8 is off. The chrominance video is applied to the control grids of both demodulators; the Q
demodulator (V7) is sensitive only to Q video components, and the I
demodulator (V8) is sensitive only to I video components.
The above statement is true only for double sidebands in quadrature, as
exist for the Q band to 0.5 MHz and for the I band to approximately 0.6
MHz. However, the I band has a single sideband extending to 1.5 MHz so
that smaller areas of colors along the orange -cyan axis may be defined. To
eliminate cross modulation, filters are used following I and Q demodulation. Luminance components also are eliminated by the filters. The chromi-
nance channel carries hue and saturation information, and the Y channel
carries the luminance information.
The Q -filter output feeds phase splitter V9A to deliver -Q and +Q
signals to the matrix. The I filter and delay feeds amplifier VIO to equalize
the wider -band gain with the narrower -band gain of the Q channel. Amplifier V10 then feeds the I phase splitter to deliver
and +I signals
to the matrix.
In the matrix, the minus and plus I and Q signals are added to plus Y
signals. The results are shown in Table 10 -1.
-I
Automatic Phase Control (APC)
The APC loop in the system of Fig. 10-2 consists of phase discriminator
V5, reactance tube V11, crystal oscillator V12A, and phase amplifier V13.
The action of the APC discriminator is illustrated by Figs. 10 -3 and 10 -4.
The purpose of this stage is to maintain a balanced condition that exists
when the two input sine waves are of the same frequency and phase. When
the local oscillator tends to drift, a dc correction voltage adjusts the oscillator phase and returns the circuit to a balanced condition.
The phase detector (VS) receives a signal directly from the burst amplifier (V4) , and a second signal from the local oscillator (V12A) through
amplifier V13. The burst signal is fed in push -pull by T1 to the cathode of
V5A and the plate of V5B. Without local -oscillator voltage, and with the
TELEVISION BROADCASTING CAMERA CHAINS
398
Table 10 -1.
Decoded Signal, Fully Saturated Color Bars
Relative Amplitudes
Matrix
Color
Y
I
Q
Total
1.0
0
0
+0.30
-0.19
-0.13
-0.33
1.0
1.0
Cyan
Green
0.89
0.70
0.59
Magenta
0.41
Red
0.30
+0.26
+0.57
Blue
0.11
-0.30
+0.33
+0.13
+0.19
White
Yellow
1.0
0
0
+0.20
+0.14
+0.33
+0.33
White
Yellow
Red
+0.95
+0.63Q +Y
I
Green
-0.281
-0.64Q+
-1.11
I
+1.71Q +Y
0
0
1.0
LO
0
1.0
1.0
1.0
1.0
0
0.89
0.70
0.59
-0.09
Cyan
Green
Magenta
0.41
Red
0.30
-0.08
-0.16
Blue
0.11
+0.09
-0.14
-0.20
0
0
1.0
-0.53
-0.36
-0.89
1.0
White
Yellow
Blue
-0.57
-0.26
+0.16
+0.08
1.0
0.89
0.70
0.59
-0.36
Cyan
Green
Magenta
0.41
Red
0.30
-0.30
-0.66
Blue
0.11
+0.36
+0.66
+0.30
+0.89
+0.36
+0.53
0
0
0
0
1.0
0
1.0
balance control properly adjusted, the currents (IA and IB) through the
V5A and V5B branches are equal and 180° out of phase. Therefore, the
voltage developed across C3 is zero, and no correction voltage is applied to
the grid of reactance tube V11.
IA
h
Cl
Rifilar
Wound
E2
Ti
//
From Burst Amp
Bal
C2
R1
IB
Local- Osc
Voltage
To
Reactance-Tube Grid
E4
C3
V13 Plate Load
(Fig.10 -2I
Fig. 10 -3. Functional diagram of APC discriminator.
399
COLOR PICTURE MONITORING SYSTEMS
(A) E4 is zero.
rEl
E2
E2t
(C) E4 goes positive.
(B) E4 goes negative.
Fig. 10 -4. Vector diagrams for APC discriminator.
The phase relationship of the voltages in the balanced condition is shown
in Fig. 10 -4A. As shown by this diagram, the vector sums El + E3 and
E2 + E3 are equal. Since IA and IB are equal, no voltage is developed
across C3, and E4 is zero.
If the phase of the local- oscillator signal drifts, E3 will no longer be
exactly 90° from the burst signal, as shown in Fig. 10 -4B (E3 lags normal
phase). In this case, IA increases relative to IB, and capacitor C3 charges
negatively. The resulting E4 of negative value is applied to the reactancetube grid to correct the oscillator phase and rebalance the discriminator
currents. If E3 leads the normal phase, as indicated in Fig. 10 -4C, current
IB increases relative to IA, and E4 becomes positive.
Amplifier and Color Killer
Fig. 10 -5 is a partial schematic diagram of a typical monitor and illustrates terminations and circuits associated with the delay line. It also shows
a point of burst takeoff that differs from that of Fig. 10 -2.
Bandpass
To
Burst Amp
i
Ist
1µt
Bu st
Takeoff
6CL6
Video Amp
Y Delay
.
1
6
7
:i
R121500
To Second
a- Video Amp
(Y)
1500
3
-
+300V +300V
/300
Delay
4.5 -MHz
Trap
Terminations
To Bandpass Amp
R21a
(
Chroma Section)
Fig. 10 -5. Delay-line terminations and burst takeoff.
400
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 10 -6 shows the first stage of the chroma section. Control R2 of the
chroma input is ganged as a second section of R1 of the luminance input
(Fig. 10 -5). Potentiometers R1 and R2 make up the overall picture contrast control. Since the proper ratios of Y- channel gain to chromachannel gain must be maintained throughout the system, these controls
operate together in this particular arrangement. Proper balance of respective channel gains then may be adjusted by individual gain controls
later in the chroma channels.
The amplified composite signal appears in the plate circuit of the band pass amplifier, where a filter restricts the bandpass to approximately 2.3
to 4.2 MHz. The color- carrier sidebands and the higher Y- signal frequencies are passed, but the lower frequencies, which include the horizontal- and vertical -sync pulses, are eliminated.
The grid return of this amplifier is through R5 and C5. This network
constitutes a bias time constant and is shunted by the color -killer triode
through a separate winding on the horizontal- output transformer. The grid
of the 6BL7 is returned to plus 300 volts and to a point in the color -sync
phase discriminator. During monochrome transmission, no color burst is
transmitted. The phase- discriminator circuit is inoperative, and no negative voltage is supplied to the grid of the color killer. The color killer conFrom Cathode of
Bandpass Filter
1st Video Amp
(Fig. 10 -5)
lApprox 2.3 -4.2 MHz)
6
1/2 6118
Bandpass Amp
2
R2
Contrast
°°. j-0-6-0-or
15k
.011
3
39k
500
L
J
0.01
56k
100
0.047
C4
R4
T.01
33 pF
-47pF
Over-All
100kCutoff
Chroma Gain,
Time
R3,
Constant
,300
Flyback Pulse
i"i
10k
5µS- --i
Time- Constant
Pulse
8
Blas Time
r-
V
To
and Q
Demodulator
Grids
Ví0
i-
iis-
Constant
6BL7
2
Color Killer
Phase
C51
TD.
L.
R5
Discriminator
27k
r
Winding on Ho iz
Output Transformer
Fig. 10 -6. Bandpass
100 meg
+300 V
amplifier and associated circuits.
I
COLOR PICTURE MONITORING SYSTEMS
401
ducts because of the positive pulses applied to the plate of the tube. This
condition results in a charge on C5. The time constant of C5 and R5 is such
that a negative voltage sufficient to cut off the bandpass amplifier is maintained between conduction periods. By cutting off the bandpass amplifier,
the possibility of spurious chroma response during monochrome transmission is eliminated.
For color signals, in which the color -sync burst is transmitted, the grid
of the color killer is held negative by a potential developed in the phase
discriminator. Since the tube then cannot conduct even with a positive
plate, C5 is not charged, and the bandpass amplifier functions normally.
It is advantageous to eliminate the color -sync burst from the I and Q
demodulation circuits and subsequent chroma amplifiers. In the circuit of
Fig. 10 -6, this is accomplished by feeding a negative pulse to the screen
of the 6U8 at horizontal- flyback time. Since the flyback pulse is only approximately the duration of horizontal sync (5 .ts) , it is necessary to delay or widen the keying pulse so that the screen is held negative past the
occurrence of the color -sync burst. In this circuit, R4 and C4 constitute a
time constant sufficiently long that the charge of C4 through R4 holds the
screen negative until slightly after the end of the color burst, when active
line scanning starts.
Burst Separator and Amplifier
It is necessary to separate the color -sync burst from the composite video
signal in order to drive the synchronizing section of the chrominance circuits. One method for doing this is illustrated in Fig. 10-7A. The control
grid of the burst amplifier receives the color burst through the tuned takeoff
transformer in series with the plate of the first video amplifier (in this
particular circuit). Amplification of only the burst is assured by gating
the cathode of the 6U8 at the proper time during the blanking interval.
During the positive interval of the flyback pulse, the cathode remains too
far positive to allow conduction of the tube. During flyback time, the pulse
is negative and reaches the conduction potential just before burst time.
The cathode is held negative by the gate pulse (hence the tube is conducting) until just after the end of the burst interval. The gated color
burst is amplified and fed to the phase discriminator through the tuned
plate transformer.
Fig. 10 -7B shows some waveforms that illustrate the effect of proper
burst -transformer time constant, L/R, on the pulse -passing characteristic.
So long as L/R is much greater than the pulse duration, which is 2.5
microseconds, the color -burst envelope is passed without distortion. Should
L/R become less than this duration, with a radically increased R or decreased L, the envelope is differentiated, and the color burst is lost.
Another method for extracting the color burst is shown in Fig. 10 -8.
In this example, a negative gating pulse is fed to the grid of a keyer, is
amplified, and is inverted in polarity. The resultant signal is fed to the
402
TELEVISION BROADCASTING CAMERA CHAINS
grid of a burst -amplifier stage. The cathodes of the burst amplifier and the
keyer are connected together. The heavy conduction of the keyer tube
during scan time produces at the cathode a positive voltage that is sufficient to cut off the burst amplifier. The positive pulse that is fed to the grid
of the burst amplifier is sufficient to overcome the bias voltage applied to
the cathode, and the tube conducts. Thus, the color burst is extracted from
the composite video signal and is passed to the phase -detector circuit.
3.58-MHz
Burst -Takeoff
1st Video Amp
112
Bur
Transformer
6U8
t Amp
3.58 MHz
11
11
To
10k
L
To Y
Phase
Disc
0.01
Horiz
Flyback
Channel
12k
Gate
Coil on Horiz- Output
220k
Time
Transforme
Constant
+300 V
2 µF
5µs-1
Gate Pulse
-v,
Bus
(A) Circuit diagram.
To Y
To
Burst Amp
LIR
»
2.5 µs
L
/R« 2.5µs
12.5µs
Color Burst
3.58 MHz
Pulse Passed
Without Distortion
Pulse
Differentiated
(B) Burst waveforms.
Fig. 10 -7. A gated burst amplifier.
Fig. 10 -8 includes a circuit that can be used to kill the color -sync burst
in the bandpass amplifier. The triode section of the 6U8 is employed as a
cathode follower to feed the grid of the pentode section through a 1N34
diode. For the duration of the negative gating pulse, the positive terminal
of this diode is held too far negative to allow conduction, and grid 1 of
the pentode receives no signal. During intervals between gating pulses,
the video, which is of positive polarity at this point, is passed and amplified. Since the gating pulse usually is taken from the horizontal- deflection
403
COLOR PICTURE MONITORING SYSTEMS
+285 V
+285 V
To Video Amp
13h1
18k
112
6AN8
-
1/2 6AN8
Burst
Keyer
Amp
7
2
1150 pF
100k
22k
1800
0-tr
10k
(8I
To
3
Phase
Del
if
0.002
270pF
10047
Neg Gate Pulse
6
I
10.01
(2200
+285
V
B+
6118
To
Bandpass Amp
Bandpass Filter
6
Positive
Signal From
Video Amp
7
44001
Chroma
Gain
Cath
1
8e
Fig. 10 -8. Gating of burst amplifier and bandpass amplifier.
system, it should be realized that the stability of the horizontal AFC is of
great importance for color receivers. The phase of the gating pulse must
not change radically over the normal range of the horizontal -hold control.
Color Sync and Phase
The output of the burst amplifier feeds a phase -detector circuit that also
receives a signal from the local 3.58-MHz oscillator. The phase detector
has an output voltage that indicates by changes in polarity and magnitude
any difference in phase of the two signals. This correction voltage is fed
to a conventional reactance-tube circuit that locks the local oscillator on
frequency and at a specific phase relationship with the transmitted color
burst.
Before looking at typical circuits in this section of the color monitor, let
us review system requirements for the I and Q circuits. Fig. 10 -9 aids in
this review. Remember that a definite phase relationship is necessary so
that I (wideband color) lies along the orange -cyan axis of the color
triangle.
The voltage induced in the secondary of the transformer shown in Fig.
10 -9A lags the primary reference-burst current by 90 °. This is conven-
404
TELEVISION BROADCASTING CAMERA CHAINS
tional transformer action. Since this secondary is center tapped, the voltages at opposite ends are 180° apart. Thus, El and E2 are 180° apart and
in quadrature with the reference burst, as shown by the phase diagram in
Fig. 10 -9A.
To Phase
El
Disc
33 pF
El
Ref
Burst
Ref
Burst
Phasor Control
E2
E2
Pri
Sec
240 pF
(A) Phase -shift network.
90° Lag
+
Q
\
Ref
Burst
/
\
90° Lag
A 33°
I
lead at phasor control produces
I
axis for
A
90° lag from
I
will produce
Q
axis for
Q
I
demodulator.
demodulator.
(B) Shift to I axis.
(C)
I -Q relationship.
Fig. 10 -9. Phasing of cw signals for l -Q demodulation.
We may now see that if voltages El and E2 are exactly in quadrature
with the color -burst phase, adjustment of the phasor control (R -C combination that produces a leading phase angle) to 33° will cause the output
voltage to lie along the I axis (Fig. 10 -9B) If El and E2 are not exactly
in quadrature with the reference burst (as is usually the case because of
the leakage reactance), the phasor control covers an adequate range to
provide proper phasing. This phase -shifted voltage is fed to the phase
discriminator, which compares it with the reference burst. A correction
voltage then holds the local oscillator at this phase angle (I axis) as described in the following.
Observe in Fig. 10 -9C that a signal lagging the I axis by 90° will supply a carrier along the Q axis. The importance of this relationship lies not
so much in practical adjustment of monitor circuits as in the fact that it
permits a more accurate visualization of the system function. A phasing
.
,
COLOR PICTURE MONITORING SYSTEMS
405
control, ordinarily on the front panel, properly phases the local oscillator
so that the locally injected carriers fall on the I and Q chrominance- signal
axes. Improper adjustment causes inaccurate color reproduction such as
reds going blue, blues going green, and, most important, wrong flesh tones.
Fig. 10 -10 shows one type of APC loop for color synchronizing. Diodes
V1 and V2 constitute the phase- discriminator circuit for comparison of
the local -oscillator signal through T1 with the transmitted burst signal
through T2. Note that the diodes are connected to conduct on the same
half-cycle of the signal. No conduction occurs without voltage from the
burst amplifier.
During conduction, the plate of V2 goes negative with respect to
ground. This point connects to the color -killer grid (Fig. 10 -6) to prevent
that tube from conducting during color telecasts.
Transformer T1 and the phase control serve the function described previously. The discriminator then acts to supply a correction voltage for the
reactance tube so that any deviation from the I axis is corrected as in
conventional AFC circuits.
The in -phase I carrier is taken from the secondary of the transformer
in the cathode circuit of the crystal oscillator. The Q amplifier is fed from
the cathode of the oscillator.
Actually, APC circuits differ widely. Some monitors use lumped resistance- inductance circuits for the quadrature phasing. The principles,
however, remain the same.
A receiver or monitor that demodulates on the R -Y and B -Y axes
differs from the one just described in that no 33° phase relationship with
signal lies along the sine axis.
the reference burst is necessary. The B
demodulator through a 90°
It is then only necessary to feed the R
phasor to demodulate the red color-difference signal. This particular feature does not aid in simplification, since 33° networks are quite simple
(Fig. 10 -10); and whether this 33° phasor is used or not, a phasing control must be used for accurate placement of the local- carrier phase.
-Y
-Y
Synchronous Demodulators
Fig. 10 -11 illustrates one method of feeding the synchronous demodulators and the matrix. The chrominance signal is fed to the control grids
of the demodulators. The suppressor grid of the I demodulator is driven
by the in -phase cw, and the suppressor grid of the Q demodulator is driven
by the quadrature cw. The chrominance signal, of course, contains both I
and Q color information.
In the I demodulator, the output contains the vector sum of the I- signal
sidebands from the chrominance channel and the in -phase cw. The Q- signal
sidebands in the chrominance signal, since they are in quadrature with
the I sidebands, produce zero output in the I demodulator. This is the
action of a synchronous demodulator; the output is zero for components
90° in phase from the cw drive.
406
TELEVISION BROADCASTING CAMERA CHAINS
Fig. 10 -10. Color-synchronizing circuit.
407
COLOR PICTURE MONITORING SYSTEMS
Fig. 10 -11. Demodulators and phase inverters for
I
and Q signals.
408
TELEVISION BROADCASTING CAMERA CHAINS
The output of the Q demodulator contains the vector sum of the quadrature cw and the Q- chrominance sidebands. The single sideband of the I
signal above 500 kHz produces a quadrature component, and therefore
introduces cross talk into the Q- demodulator output. This is the reason for
the 500 -kHz filter in the plate load of the Q demodulator in Fig. 10 -11.
Note that the I demodulator has a gain control and an extra amplifying
stage. Since the Q sidebands are equal (double sidebands) , the Q gain is
twice that of the I channel above 500 kHz. The extra I amplifier compensates for this difference, and the gain control allows exact adjustment for
proper gain ratio.
CW
O
Signal on
Grid 3
Plate -Current
Pulse From 111
Fig. 10 -12. Action of synchronous
demodulator.
O G/
Quadrature
Signal on Grid
1
Zero
Maximum
In -Phase Signal
on Grid
1
Fig. 10 -12 illustrates the basic operation of a synchronous demodulator.
Note from Figs. 10 -10 and 10 -11 that the suppressor grids (grid 3), which
receive the cw signals, are biased at a negative 5 volts. The peak voltage
of the cw signals may be approximately 30 volts. Parameters of the circuits are such that the tube conducts heavily on peak regions of the cw
signal even though there is no chrominance modulation on grid 1.
Tube conduction with no chrominance signal is shown in waveform 2
of Fig. 10 -12. The instantaneous signal voltage on grid 1 determines the
amount of that current that reaches the plate. Thus, the instantaneous voltage on grid 1 is multiplied by the instantaneous voltage on grid 3. This
action is sometimes referred to as product demodulation.
409
COLOR PICTURE MONITORING SYSTEMS
On grid 1, signals that are in phase go positive when grid 3 goes positive. This results in heavy plate current on the positive peaks. The pulses
follow the modulation on grid 1.
A quadrature signal on grid 1 (waveform 3 of Fig. 10 -12) is zero at
the time of the plate -current pulse; therefore no signal is produced in the
output. An in -phase signal (waveform 4 of Fig. 10 -12) adds vectorially to
the cw signal.
If the cw signal is not of the same frequency as the chrominance signal,
the usual heterodyne beat frequency results. This undesired beat component is modulated by the amplitude variations on grid 1, and as a result
there are spurious products in the output circuit. The fact that these signals
pass through the low -pass filters into the chrominance amplifiers emphasizes the importance of proper APC loop-phasing adjustments.
Note that the plate of the Q phase splitter (Fig. 10 -11) is dc coupled
to the matrix, whereas the cathode is ac coupled. This is necessary to isolate
the plate and cathode dc voltages in the matrix. Similarly, the +I signal is
ac coupled, and the
signal is dc coupled to isolate these separate
voltages.
-I
Matrixed Signals
The matrix extracts the color -difference signals (R Y, G Y, and
B
Y) from the filtered output of the I and Q demodulators. These signals, along with the luminance, or Y, signal, excite the color picture tube
with instantaneous values such that the overall function matches the corresponding scanned point at the studio camera.
The I -Q matrix at the sending end reduces the R -Y component to:
-
-
0.877 (R
Similarly, the B
-Y component
- Y), or R1.14
is
0.493 (B
-
(Eq. 10 -1.)
reduced to:
-Y),or B 03
(Eq. 10-2.)
In this manner, both the I and Q channels contain some of both color difference components so that only a two -phase color signal is required
for three chrominance primaries. It is the purpose of the matrix in a receiver or monitor containing an I -Q demodulator to recover 1.14 (R Y)
and 2.03 (B Y) . This means that the color- difference components are
recovered in their original forms before I and Q matrixing at the transmission end; therefore, the R -Y gain is 1.14 and the B -Y gain is 2.03.
In the wideband monitor, this is achieved in the matrix operation.
This action is emphasized in Fig. 10 -13, which shows the R -Y and
B -Y components multiplied by 1.14 and 2.03, respectively. We may now
note the values of I and Q necessary to extract the color-difference components existing before modulation. These are found to be:
-
-
410
TELEVISION BROADCASTING CAMERA CHAINS
R
-Y = 0.951 + 0.63Q
B-Y=-1.11I+
(Eq. 10-3.)
(Eq. 10-4.)
1.71Q
We know that Y is:
Y= 0.30R +0.59G +O.11B
Rearranging equation 5:
G
(Eq. 10 -5.)
.
-Y= -0.51
(R
- Y)
- Y)
-0.19 (B
(Eq. 10 -6.)
This is the action performed in monitors in which the color -difference
signals are demodulated directly in the matrix to extract the G
component. In terms of I and Q for the wideband color receiver, substituting
Equations 10 -3 and 10 -4 into Equation 10 -6 gives:
-Y
G
-Y = -0.281 - 0.64Q
(Eq. 10 -7.)
Equation 10 -7 is also shown in Fig. 10 -13 for matrix operation necessary
to extract the G -Y color -difference signal in terms of I and Q.
+Q
+I
1.71
0. 95
Q
/
1
0. 63 Q
B-Y
12. 031
-Q
O.
//
-0. 64
28
I
i'
Q
z.
..-
z
G-Y
z'
-1.111
Fig. 10 -13. Conversion of
I
\
and Q signals to color- difference signals.
The Y signal and both polarities of the I and Q signals are fed to the
output matrix to recover the original R, G, and B signals. This matrix
must receive the correct ratios of signals from the phase -splitter outputs.
This depends on the adjustment of the chroma control, the quadrature
adjustment of the demodulators, and the setting of the I -gain control.
COLOR PICTURE MONITORING SYSTEMS
411
As previously discussed, the I channel requires an extra amplifier stage
to equalize its wideband gain relative to the narrow -band gain of the Q
channel. The I -gain control provides control of this amplification.
The result of the combined matrix function is shown by Table 10 -1.
For example, in order to recover the original red signal, +0.95I and
+0.63Q are added to the luminance signal of +0.30 to give unity signal
at the red output.
Conditions for proper matrixing are first determined by adjustments
as follows:
I and Q signals, the hue control is adjusted
for zero Q response in the I channel. Remember that the I and Q
sidebands vanish for a white signal, which represents the zero axis.
Signals above white are positive, and signals below white are negative. When hue is properly adjusted, the local oscillator is locked to
the I- carrier phase, and no Q response (either positive or negative) is
obtained in the I channel.
2. The quadrature- adjust control must be adjusted for zero I response
in the Q channel. This now assures that the demodulators are gated
properly with reference to their respective axes, and will not deliver
contaminated signals to the matrix.
3. The chroma and I -gain controls must be adjusted so that the blue
output has zero I response and maximum Q response. This results
in proper luminance -to- chrominance gain ratios so that the fixed
matrix network results in correct output signals.
1.
With input representing
We may now examine the actual dynamics of the matrix function. Fig.
(Y) and
chrominance input levels are shown at the left of the drawing. The Ychannel input is taken as unity, with voltage developed across R1, R2,
and R3 in series with their respective output adder resistors, R7, R8, and
R10. The chroma and I -gain controls are adjusted so that +Q is 1.71 times
Y, and -I is 1.11 times Y.
Now refer to Fig. 10 -11, and note that the signal +Q appears at the
plate of the Q phase splitter. The plate load is 33k in parallel with R5
(27.2k) in parallel with R4 (10k) . This is an effective plate load of
about 6k. The -Q signal appears at the cathode. The cathode load is 3k in
parallel with R12 (10k), or 2.3k. Thus, the plate provides 6/2.3 or 2.61
times the gain at the cathode. With the chroma gain adjusted so that +Q
is 1.71 times Y, -Q is 1.71/2.61, or 0.64, times Y, which appears at one
input, R12, of the green matrix. Observation of the required matrix proportions for I and Q in Fig. 10 -13 and Table 10 -1 shows that the blue
matrix requires +1.71Q, and the green matrix requires -0.64Q. These
proportions are indicated in Fig. 10 -14. Note that the +1.71Q is applied
in parallel to blue- matrix resistor R4 and red -matrix resistor 115. Resistor
10 -14 is a simplified diagram of this section. Relative luminance
412
TELEVISION BROADCASTING CAMERA CHAINS
lY
R1
Y
Red
R7
470
Q
GP
Green
,
R8
Fig. 10 -14.
Matrix for color monitor.
470
Q
Blue
10k
R102470
R4
10k
R5 provides 2.72 times the resistance of R4. Therefore R5 provides a
Q- signal ratio of 1.71/2.72, or 0.63, as required for the red matrix.
is available at the plate of V4B (Fig. 10 -11), and
For the I channel,
+I is available at the cathode of V4B. The I gain is adjusted so that the
signal at the V4B plate is 1.11Y. As in the Q channel, the total plate and
cathode loads are proportioned so that when the I gain is adjusted such
that
is 1.11Y, then +I is 0.95Y for the +I input at the red matrix.
Each output resistor is the adder across which appears the algebraic sum
of the inputs to the three preceding resistors. Only those signals containing
red appear across R7, since the green and blue components were cancelled
by the polarity and amplitude relationship of the Y, I, and Q signals fed
to the red channel. The amplitude of the red voltage determines the degree
of saturation of the red component. The same principle holds for the
green and blue outputs.
It will be noted by serious readers who have followed the actual computations in preceding examples that slight manipulation of values was made
to result in required values. The reason is that, in the interest of clarity,
absolute NTSC numerical values have not been used, and numbers beyond
two decimal places were disregarded. For example, we have used the
popular I and Q proportions:
-I
-I
-
-
= 0.60R 0.28G 0.32B
Q = 0.21R 0.52G + 0.31B
I
413
COLOR PICTURE MONITORING SYSTEMS
whereas absolute values are:
-
-
= 0.599R 0.278G 0.321B
Q = 0.213R 0.525G + 0.312B
I
10 -2. COLOR PICTURE-TUBE CIRCUITRY
The color picture tube to be discussed is the three -gun, shadow-mask type
most common at the time of this writing. The three guns are physically
fixed in position relative to mask holes and phosphor dots on a screen.
One gun and grid assembly is arranged to strike only phosphor dots that
emit red light when struck by the beam; one gun strikes only green dots;
and the remaining beam strikes only blue dots. When the beams are properly converged, one picture element is a triangular arrangement of three
adjacent dots. When the observer is at normal viewing distance, the dots
are not visible as such, and a colored image of satisfactory definition is
reproduced. The phosphor screen contains about 7000 dots per square
inch. The screen is aluminized, and no ion trap is necessary.
By the nature of this basic functioning of the tube, the physical construction of the gun and electrode assembly must be precisely controlled.
Adjustment magnets, coils, and voltages then allow sufficient correction
for any slight physical misalignment and characteristics of electrical
deflection.
Each gun assembly contains a cathode, a control grid (grid 1) , and a
screen grid (grid 2) . Grid 3 ( focus electrode) and grid 4 (convergence
electrode) are common to all three assemblies. The ultor is the high voltage electrode, and this potential is also connected to a wall coating on
the large bell of the envelope. External to the picture tube are beam positioning magnets, the color -purity coil, the deflection yoke, and the
field -neutralizing coil (when used) .
Fig. 10 -15 illustrates the function of one gun assembly of the tri -gun
arrangement. Numbers along the top designate the various forces acting
on the beam. The voltages shown are typical values for a 17 -inch tube.
Obviously, the ultimate goal of the electron beam is the high voltage of
the ultor, or phosphor screen. Force 1 is one of varying repulsion depending on the potential of grid 1 relative to that of the cathode. Force 2 is
an accelerating force with strength dependent on the grid -2 (screen) potential, which is adjustable.
There is one beam -positioning magnet for each gun. The magnetic field
adjusts the individual beam relative to the other beams transversely to the
direction of beam travel ( force 3) . An adjustable current through the
color -purity coil creates an electromagnetic field with the same action
( force 4) , except that all three beams are influenced by this current. Its
purpose is to assure that all beams travel straight down the neck of the
tube toward the mask and phosphor screen. Construction of the tube is
414
TELEVISION BROADCASTING CAMERA CHAINS
2
E
>
c,
8
8
3
z
I
r
o
I
444,
13
6-
z
8
Fig. 10-15. Functional diagram of one gun in three-gun CRT.
415
COLOR PICTURE MONITORING SYSTEMS
Portion of Mask
Electron Beams
Fig. 10 -16. Principle of mask in color picture tube.
such that when this occurs, the beam for red strikes only red phosphors,
the beam for green strikes only green phosphors, and the beam for blue
strikes only blue phosphors. This is maximum color purity.
Force 5 compresses (focuses) the beam; this force depends on the ratio
of the grid -3 and grid -4 potentials. It will be noted that variation of the
voltage on grid 3 ( focus grid) causes a variation of beam convergence,
and a change in the voltage on grid 4 causes a variation of focus.
The mask contains a large number of holes ( Fig. 10 -16) , which allow
the beams (when properly converged) to strike only the proper phosphor
elements. Proper convergence exists when all three beams converge at the
same hole, then diverge to their respective phosphors. The mask and phosphor screen of the tube are flat (in 70° deflection tubes). Fig. 10 -17 illustrates the fact that under beam deflection, the point of convergence traces an
arc. To maintain both convergence and focus at the portions of the raster
away from the central area, horizontal and vertical dynamic waveforms are
applied to grids 3 and 4.
The deflection coil performs the same function as it does for monochrome picture tubes. The yoke is huskier because of the greater deflection
current required for color tubes.
d2
Fig. 10 -17. Variation of convergence
-dl
with deflection.
Convergence Arc
416
TELEVISION BROADCASTING CAMERA CHAINS
Current in the field -neutralizing coil around the face of the tube minimizes the effects of stray magnetic fields. Control of the beams must be
precise; even the magnetic field of the earth must be nullified.
Fig. 10 -18 shows the circuits associated with a three -gun color picture
tube. Each grid receives excitation from its specific video channel, is dc
controlled, and is returned through individual background controls. The
cathodes are driven positive during vertical- blanking intervals by pulses
from the vertical-output transformer. This assures complete retrace blanking. The dc voltage of each screen can be adjusted individually to aid in
balancing phosphor efficiencies.
The convergence-amplifier grid receives waveforms from the cathodes
of the horizontal- and vertical- deflection amplifiers. The shapes of the waveforms at these points are very nearly parabolic. The controls that are shown
provide any correction necessary to shape the applied voltages properly.
The amplified output modulates the dc applied to the convergence electrode. Since the distances from the guns to the top, bottom, left, and right
regions of the screen are greater than the distance to the center of the
raster, the beams from the three guns would not register properly without
correction. Dynamic convergence corrects the applied voltage in step with
the scanning position of the beams.
The dc voltage applied to the focusing electrode is adjustable for optimum control of the scanning -spot size from each gun. The focusing electrode also receives a portion of the parabolic waves to maintain focus
during scanning.
The picture -tube high voltage is regulated by a shunt -regulator tube
across a flyback voltage -doubler power supply. During a black picture and
blanking intervals, when no beam current is drawn, the regulator becomes
the power-supply load. For a maximum white picture, the kinescope becomes the load, and the regulator tube absorbs very little power. This
method maintains a constant load on the high -voltage power supply and
allows good regulation for the picture -tube anode.
In addition to the contrast (video gain) control, controls that affect
picture quality are the brightness and background controls. Usually, dc
restoration is provided by a triple -diode tube. Note from Fig. 10 -18 that
the anodes of this tube are connected to the arms of variable potentiometers
in the background control circuitry. The anode of the red -channel dc restorer is returned to the arm of a brightness control, to which the blue
and green background controls are returned through the picture -tube cathode circuitry. Adjustment of the brightness control makes the picture -tube
grids more or less negative relative to the cathode; hence this control determines the beam current and raster brightness.
Assume the blue- background and green-background controls are at the
extreme counterclockwise position. This places the three grids at the same
bias. Because of differences in the efficiencies of the three phosphor dots
of each element, the background of a picture that should be white would
COLOR PICTURE MONITORING SYSTEMS
417
Fig. 10 -18. Dynamic- convergence circuits and screen and background controls.
418
TELEVISION BROADCASTING CAMERA CHAINS
be tinged with a color component. The green- screen, blue- screen, and red screen controls normally are adjusted for white balance on high lights. Since
restorer action refers to blanking level, the background controls normally
are adjusted for white balance with the brightness control adjusted for
low raster brightness. This method assures proper balance over the range
from low lights to high lights of the video signal.
10 -3.
ADJUSTMENT OF COLOR MONITORS
Following is a basic outline of typical monitor adjustments that are
applicable to color monitors in general. We will then consider refinements
and additions in modern circuitry. The proper sequence for check-out of
the monitor, or adjustments after installation of the kinescope, is as follows:
Horizontal-drive and high -voltage adjustment
linearity adjustments (adjustment for correct aspect ratio
and linearity by conventional bar -generator methods)
3. Color- purity adjustment
4. Screen adjustment for white, and color -balance adjustment
5. Convergence adjustment
1.
2. Size and
Horizontal -Drive and High -Voltage Adjustment
To adjust the horizontal drive and the high voltage, perform the following steps:
Plug a 0 -1.5 mA or 0 -2 mA meter into the regulator- current jack
usually provided on the high -voltage chassis.
2. Turn the brightness and contrast controls fully counterclockwise
(off) The regulator current should be close to 1 milliampere, or as
specified in the instruction book for the monitor.
3. Adjust the horizontal -drive control (rear of chassis) for maximum
regulator current. From 0.8 to 1 mA (or slightly more) is normal
( with brightness and contrast controls off) .
4. If at least 0.8 mA of regulator current is not available with the adjustment in step 3, measure ( with a VTVM and high -voltage probe)
the potential at the ring on the flyback rectifier. Lower the voltage
with the high -voltage- adjust control (rear of chassis) until the regulator current reaches at least 0.8 mA. Note that as the ultor voltage is
lowered, the regulator current increases, and vice versa. It is more
important to obtain good raster brightness without blooming than
to have an absolute 20 kV (or other specified voltage) on the ultor.
Because of differences in color kinescopes, a variation of several
kilovolts may be observed between tubes for optimum performance.
5. Return the brightness and contrast controls to the normal operating
positions.
1.
.
COLOR PICTURE MONITORING SYSTEMS
419
conventional bar -generator methods, adjust the aspect ratio of the
raster and the linearity of the sweeps.
6. By
Screen -Purity Adjustment
To avoid shading of the color screen, it is necessary for the individual
gun emissions to be properly aligned down the center of the tube structure.
If color impurity is noted in the raster, proceed as follows:
Loosen the yoke -assembly bracket screws and slide the yoke as far to
the rear as possible. Remove the alignment magnets from the neck
assembly. Turn the blue-screen and green -screen controls fully counterclockwise. Turn the red -screen control fully clockwise. Since the
yoke has been slid back from its normal position, deflection is not
complete, and the red beam should be coloring the screen except at
the extreme sides. Disregard coloring around the edges at this time.
Adjust the color -purity control (or magnet) until approximately the
same area of impurity exists around all sides of the raster. A slight
rotation of the color -purity coil (or magnet) may be necessary in
conjunction with coil- current adjustment. When a color- impurity
magnet is used, tabs normally are provided for positioning.
2. Next, slide the yoke forward until the entire screen is red with maximum purity. The yoke may not need to go entirely to the front of
the slot. If the purity is good except at one extreme side or corner,
adjust the field -neutralizing control (when provided) for maximum
purity. This control determines the current through the coil at the
outer front of the kinescope.
3. Check the blue and green screen purities by turning the other screen
controls off and the one to be checked on. A slight adjustment of color
purity and/or field -neutralization may be needed to effect the best
compromise in screen purities. Tighten the yoke -assembly bracket
screws and replace the alignment magnets.
1.
NOTE: It may be necessary to demagnetize the picture tube with a degaussing coil to obtain best color purity.
Color Balancing
Turn the contrast control to minimum ( fully counterclockwise) . Set the
brightness control for maximum raster brightness. Adjust the red -screen,
blue -screen, and green- screen controls to obtain a low- brightness white
raster.
Feed a black- and -white test -pattern signal to the monitor, and be sure
the three channel gains at the camera control are equal, as indicated on the
CRO's. Bring the monitor contrast control up slightly over midway, and
adjust the brightness for a reasonable presentation. Adjust the monitor
green-video gain and blue -video gain until the reference white high light
of the test pattern prevails on the monitor screen.
420
TELEVISION BROADCASTING CAMERA CHAINS
Leave the contrast control at its present setting, and turn the brightness
down to low level. Adjust the blue-background and green- background controls (dc- restorer controls) for proper low- brightness white.
Convergence
Check convergence by placing the normal/convergence switch in the
convergence position. This ties all three video channels together. With a
test -pattern signal (crosshatch or dots), all portions of the image should
be black or white. Any color fringing is an indication of misconvergence.
Remember that camera misregistration would not contribute to color fringing in this case, since a common video signal is being amplified by the
monitor. If slight misconvergence is noted, the dc- convergence control
should bring the beams into convergence. When serious misconvergence
or bad fringing at the sides and corners requires adjustment of the
dynamic-convergence controls, tube or circuit malfunctioning may be indicated. After maintenance, it is desirable to adjust the convergence controls
with a dot generator as described below.
Place the normal /convergence switch in the normal position. Feed a
dot -generator signal to the monitor. (NoTE: Use a pattern in which the
dot size is no more than two raster lines, not the large dot size of ten raster
lines.) Adjust the height, width, and linearity controls for proper aspect
ratio and linearity. The dots may be used for this purpose, or the more conventional bar generator may be used.
Fig. 10 -19 illustrates the basics of convergence. The kinescope is installed with the blue gun on top, the green gun to the left, and the red gun
to the right as viewed from the rear (Fig. 10 -19A) . Fig. 10 -19B shows the
necessary conditions for proper convergence. With the correct dc- convergence voltage, the beams converge at the proper hole in the mask, then
diverge to adjacent dots forming the triangular picture element, as shown.
The dc- convergence control has the greatest effect in the central picture
area.
Fig. 10 -19C represents a front view of one picture element properly
converged. The phosphors are so closely spaced that from a normal viewing
distance the color dots blend into one white dot. If the dc-convergence
voltage is too high, the blue dot will be high, the red dot will be separated
several elements to the left, and the green dot will be several elements to
the right ( from a front view of the screen-Fig. 10 -19D) . Note that
misconvergence does not cause loss of any primary color; the beams always
strike their intended phosphors, but they may be as much as I/4 inch or
more from the proper triangular area defining adjacent dots. With low
dc- convergence voltage, the blue dot is lower than the other two dots.
The first rule is to obtain proper convergence on a horizontal line and
a vertical line through the center of the raster. Place the following controls
in their approximate midrange position: horizontal dynamic- convergence
amplitude, vertical dynamic- convergence amplitude, vertical dynamic -con-
421
COLOR PICTURE MONITORING SYSTEMS
(A) Relative gun positions.
Electron learns
Phosphor Dots
Portion of Mask
(B) Paths of electron beams.
(C) Picture element converged.
(D) Picture element not converged.
Fig. 10 -19. Principles of convergence.
422
TELEVISION BROADCASTING CAMERA CHAINS
vergence phase, and horizontal dynamic- convergence phase. Adjust the dcconvergence control clockwise to a point at which the blue dot is high.
Then by adjustment of the individual red, green, and blue beam-positioning magnets, bring the dots into the approximate relative positions shown
in Fig. 10 -20A. The dots should describe an equilateral triangle. In Fig.
10 -20B, the arrows indicate direction of dot movement under influence
of the dc- convergence voltage. Raising the voltage causes blue to move
upward and red and green to move farther apart. Lowering the voltage
causes the dots to converge as shown by the arrows. With a little practice,
the technician is able to position the dots accurately so that lowering the
dc- convergence voltage perfectly converges the dots through the central
area of the picture -tube screen.
Arrows indicate influence of
Arrows indicate influence of adjustment of magnets.
(A) Beam positioning magnets.
DC
convergence- electrode voltage.
(B) Dc- convergence control.
Fig. 10 -20. Basic convergence adjustments.
It may now be noticed that convergence occurs in the center of the tube,
but not at the top, bottom, and sides of the raster. This requires adjustment of the dynamic-convergence amplitude and phase controls.
A peaking adjustment that affects the total amplitude of the dynamic
voltage normally is provided in the convergence transformer. This control
should be rotated to find the arc through which the most radical change in
spots can be observed, with the horizontal dynamic- convergence amplitude
control in the maximum clockwise position. (That is, find the arc of rotation in which the spots shift apart, converge, and go through the opposite
phase most rapidly.) Leave the control at the midpoint of the arc. Return
the horizontal dynamic- convergence control to midrange.
Observe the extreme right and left dots on the center horizontal line.
If the blue dot is high, adjust (counterclockwise) the horizontal dynamicconvergence amplitude control. If the blue dot is low, increase this dynamic voltage. Adjust the control until equal displacement is noted for all
dots. Then adjust the horizontal dynamic-convergence phase for best dot
COLOR PICTURE MONITORING SYSTEMS
423
coincidence. Readjustment of the dc- convergence control will then converge the dots horizontally. Considerable practice is necessary to be able
to accomplish convergence with only a few trials. Ordinarily, a back -andforth adjustment must be made all through this procedure.
Next, observe the extreme top and bottom dots, and adjust the vertical
dynamic-convergence amplitude control for equal displacement error. Another adjustment of the dc- convergence control should converge the dots
down the center vertical line. Remember that adjustment of dynamic voltage waveforms affects the total convergence -electrode voltages. A change in
one control requires a change in other controls. Note also that convergence
depends on the ratio of grid -4 voltage to neck -coating voltage ( high voltage) , and that focus is established by the ratio of grid -4 voltage to grid -3
( focus grid) voltage. Therefore, it is necessary to readjust the focus control often.
If the raster has become converged at the center and one side of the
screen before the other side converges, adjust the horizontal dynamic -convergence phase control for equal displacement, and then readjust the dcconvergence control. If the raster was converged in the center and either
the top or bottom (with the other portion misconverged ), adjust the
vertical dynamic- convergence phase for equal displacement, and then readjust the dc convergence.
There is a slight limitation in some present -day color kinescopes that
affects the degree of convergence in one corner of the raster. If convergence is good in all portions except a corner or small edge, the kinscope
probably is at fault.
IMPORTANT NOTE: Modern studio -type color monitors employ many
additional convergence controls, as will be outlined in the following
section.
10 -4. THE RCA
TM -21 COLOR MONITOR
One of the most important applications for a color monitor is in control
rooms where operators face the problems of setting up and matching
color cameras. A properly designed and operating color monitor offers the
following benefits:
It provides a better check of registration during actual programming
than the black- and -white master monitor does.
2. If the monitor has good deflection linearity ( within 1 percent in
both directions), a good check of camera deflection linearity is
possible.
3. Provision for underscanning, to show the corners of the picture,
permits better checking of camera framing, camera -lens aberrations,
and camera deflection transients. Underscanning also makes cue
marks in the picture corners readily visible.
1.
424
TELEVISION BROADCASTING CAMERA CHAINS
4. A highly stabilized method of black -level setting permits better
evaluation of camera shading characteristics and clearly indicates the
effects of camera pedestal adjustments.
5. Precision decoder circuits and linear output amplifiers produce a
picture of improved color fidelity, so that camera color fidelity can
evaluate more accurately.
6. Good picture sharpness facilitates checking of camera focus.
Fig. 10 -21 is a basic block diagram showing the five major chassis of the
RCA TM -21 color monitor. This monitor serves as a studio color standard
in many stations, and its description is warranted to point up the details
important to a standard color monitor. It demodulates on the I and Q
(wideband color) axes.
Composite
Color-Signa
Sync Input
(Optiona I
l
-
Red
Decoder
--o-
Green
Blue
Red
Video-Output
Amplifiers
Green
Blue
I
Sync
Deflection,
Regulated
High Voltage
Convergence
& Protection
Power
Supply
Protection
1
Deflection
High Voltage
Courtesy RCA
Fig. 10 -21. Block diagram of TM -21 monitor.
The heart of the decoder design (Fig. 10 -22) is a stabilized video driver
stage that drives the monochrome channel and the burst-controlled oscillator from its plate circuit, and the two chrominance demodulators from
its cathode. The dc component is restored at this stage by means of a feedback-stabilized clamp. One of the gating stages involved in this feedback
clamp has been made to serve as a burst separator as well, thus eliminating a
separate tube for this function. In the video driver, the plate signal current is inherently equal to the cathode signal current, so there is no possibility of gain variations in the plate circuit relative to the cathode circuit.
Prior to the video driver stage, the input signal is raised to a relatively
high level (about 12 volts peak -to -peak) by an amplifier equipped with a
nonselective, or wideband, gain control. Through use of this high level at
the driver stage, virtually all of the voltage gain required in the entire decoder is supplied by an amplifier that handles all signal components simul-
425
COLOR PICTURE MONITORING SYSTEMS
II
o
Ó
.110
Y
E
G
o
Ó
eE
m
t
=
E
i
Courtesy RCA
Fig. 10 -22. Block diagram of TM -21 decoder.
426
TELEVISION BROADCASTING CAMERA CHAINS
taneously. This technique eliminates the problem of matching the gains
of several individual amplifiers. In the stages following the video driver
( which must necessarily be split into separate channels) , it is possible to
sacrifice voltage gain for the sake of stability and still deliver signals at
about a 1 -volt level at the output of the decoder. The amount of degeneration (or feedback) that it has been possible to incorporate by following
this approach has made practical the elimination of several conventional
gain controls ( normally provided in decoders to compensate for circuit
variations)
One of the channels following the video driver stage is the burst -controlled oscillator, which consists of a crystal- controlled 3.58 -MHz oscillator
shunted by a reactance tube. The control voltage for the reactance tube is
derived from a phase detector that compares the oscillator output with the
separated burst provided by the video driver. Special attention has been
given to drift problems in this oscillator, so that the phase of its output
remains stable relative to the phase of the chrominance signal delivered
from the cathode side of the video driver.
In conventional decoder designs, the burst -controlled oscillator normally
delivers two subcarrier outputs (90° apart in phase) to the chrominance
demodulators. A popular method of deriving the two outputs is to use a
pair of tuned circuits, one tuned above resonance and the other below
resonance, to achieve the required phase shift. In the TM -21 decoder, however, a potential phase stability problem has been avoided by providing
only a single output from the oscillator. This output is tied directly to
both demodulators so that there can be no relative phase drift between
them. The required 90° phase difference is provided in the video channel
by passing the input signal to the I demodulator through a precision delay
line equivalent to 90° at 3.58 MHz. The delay line is manufactured with
a tolerance of -±-10; hence it is possible to eliminate the conventional
quadrature phase control. The presence of the delay line in the I video
channel poses no problem, because it is very simple to take it into account
when adjusting the total delay of the I channel relative to the narrow -band
Q channel.
The demodulators themselves are a stabilized diode type, as shown by
the simplified schematic in Fig. 10 -23. In essence, the circuit is a fast acting clamp. The diodes are closed periodically at a 3.58 -MHz rate, and
their effect on the signal is to connect the output side of the 120 -pF capacitor to ground through the center tap of the 3.58 -MHz transformer. The
charge stored in the 0.01 -µF capacitor through the rectifying action of the
diodes serves to make the diodes conductive only during the extreme peaks
of the subcarrier cycle. Because the clamp is closed only momentarily, the
output side of the 120 -pF capacitor is normally free to follow the variations
in the input signal. The average output level, however, is a function of the
input level at the instants when the diode conduction occurs, as illustrated
by the waveforms in Fig. 10 -24. This average level is affected by both the
.
427
COLOR PICTURE MONITORING SYSTEMS
amplitude and the phase of the incoming chrominance signal, and represents the desired demodulated signal.
The major advantages of this demodulator circuit are: (1) it has no
video -gain drift problem, since it behaves in principle like a fast -acting
switch, and (2) it is insensitive to the level of the cw subcarrier signal,
provided the cw signal is always of higher amplitude than the modulated
rf signal.
Demodulated Video Output
Low -Pass Filtering)
(Requires
Modulated
RF
Input
'
120pF
-'-'4-
3.58-MHz
CW
Input
100
Courtesy RCA
Fig. 10 -23. Clamp -type demodulator.
The convergence chassis contains purely passive circuits for modifying
certain waveforms derived from the deflection chassis before applying
them to the convergence yoke surrounding the kinescope gun structures.
The word "convergence," as applied to color -display devices, refers to the
process of adjusting the positions of the red, green, and blue beams so that
the respective images are registered in all parts of the screen. Because the
effective distance between the guns and the screen assembly varies with
the deflection angle, it is necessary to control the convergence with dynamic waveforms containing both horizontal-frequency and vertical frequency components. The basic waveforms consist of a parabola and a
sawtooth at each frequency, but these must be mixed in different proportions for each gun.
As indicated in Fig. 10 -25, both the output tubes and the output transformers of the TM -21 serve as signal sources for the convergence circuits.
Two features of the convergence-circuit design that contribute to a straightforward setup procedure are: (1) The controls are arranged so that the
red and green rasters may be adjusted as a pair, relative to each other, after
which the blue raster may be brought into registration relative to the red green pair. (2) Every control has been made to direct some type of movement in either the horizontal or vertical direction instead of along the
120° axes.
The large number of convergence controls needed for a tricolor tube (16
in the case of the TM -21) need not seem too formidable if each one performs some readily understood function. Those used are so designated and
arranged on the control panel that it is easy to visualize them as trimming
adjustments for the deflection circuits. There are five basic types of controls: The position controls are trim adjustments for the centering func-
428
TELEVISION BROADCASTING CAMERA CHAINS
tion, while size and linearity carry the same connotation as in conventional
deflection systems. The tilt and bow controls produce these effects on the
lines of the grating pattern commonly used to facilitate convergence adjustments; the bow control affects the curvature of the lines, whereas the tilt
controls are used to make the lines parallel.
The controls are grouped in two ways. The vertical, static, and horizontal
adjustments are located in separate columns. The upper controls in each
column adjust the red and green rasters relative to each other, and the lower
controls adjust the blue raster relative to the red -green pair. A screen selector switch just to the left of the convergence control panel makes it
possible to view any of the rasters separately, or to view only the red -green
pair.
Diodes close during these instants.
AAAAAAAAAAA
1;; 11;;/1YV
I
Subcarrier Signal
I
I
I
I
I
I
I
I
I
I
I
I
I
K
I
Typical Amplitude Variation'
Typical Phase Variation
...-----
t -
I
I
I
I
I
I
I
i
1
I
AtAA-Á- A Ai A
V
,
-Yr...,I
if
I
11
v
'\i \.I
I\
1
CW
V
v1v__,
--,
T
I
i
Zero-Signal Axis
_
\\
Typical Modulated Wave
at Demodulator Input
I
I
I
Demodulator Output
Before Filtering
I
I
I
N
I
to that waveform is clamped to zero-signal axis
during diode conduction periods.
I
I
I
Zero-Signal Axis
/I
I
I
Demodulator Output After
Low -Pass Filtering
Courtesy RCA
Fig. 10 -24. Waveforms in clamp -type demodulator.
429
COLOR PICTURE MONITORING SYSTEMS
i
E
1
Lö
tt
i
Courtesy RCA
Fig. 10 -25. Deflection, high -voltage, and protection circuits in TM -21 monitor.
430
TELEVISION BROADCASTING CAMERA CHAINS
The BRIGHTNESS control produces the same effect as conventional brightness controls, even though it operates in an unusual manner. Instead of
varying the bias on the kinescope, it varies the level of a special pulse
added to the signal in place of the normal sync and the burst signals (Fig.
10 -26) This technique is made feasible by the use of keyed clamps in the
output amplifiers; these clamps operate during the time interval of the
added pulses. The advantage of this brightness -control technique is that
it eliminates the need for individual red, green, and blue background controls. The single BRIGHTNESS knob automatically exercises the proper degree of control over the three color channels because the added pulse is
passed through the standard decoder matrix.
The CONTRAST control varies the gain of the input stage in the decoder.
It varies the luminance and chrominance components of the picture equally.
The SYNC SELECTOR enables the operator to switch between external sync
and the internal sync pulses separated from the composite video signal. In
the remote position, this switch brings the sync interlock into operation so
that the use of internal or external sync is controlled from a remote point
(such as a switcher) through a dc control lead.
.
The APERTURE COMPENSATOR adjusts the degree of high -peaking in
the luminance channel for optimum picture sharpness without objectionable overshoots.
By moving the TEST switch through its several positions and making
specific adjustments at each step, the monitor is brought into proper operating condition. The following paragraphs go through each of these steps
with reference to Figs. 10 -26 and 10 -27.
In the first position, the signal is automatically disconnected, but the
brightness pulse remains. In this position, the RED SCREEN control is adjusted for cutoff. The SCREEN SELECTOR switch can be set at R to facilitate this adjustment by cutting off the green and blue beams to avoid confusion.
In the second (screen balance) position, both the signal and the brightness pulse are disconnected, and the green and blue screen controls may
be adjusted relative to the previously set red screen to produce a gray
screen of approximately 20 percent brightness. The SCREEN SELECTOR
Pulse Adjustable Through About
This Range by Brightness Control
(A) Input signal.
(B) Brightness pulse added.
Courtesy RCA
Fig. 10 -26. Brightness -control waveforms.
431
COLOR PICTURE MONITORING SYSTEMS
During
Picture Scanning Time
Set for Cutoff
1
/
Red Screen
Adjustment
/
0
/
t
Blue Screen Adj
50%
/
=
ADprox 20%
f
0
L
Ec
f/
Green Screen
Grid-to-Cathode
Voltage
previously set
at
red, t20%luminasce.
about 20% luminance.
Set to balance
j'
100%
50%
Pulse is clamped to this
grid -to- cathode voltage
20%
during horizontal
blanking.
Level During Picture
Scanning Time
0
¡
Brightness Pulse of
Fixed Amplitude
No- Signal
`All three
Condition.
guns have
identical bias.
Note All light output scales are normalized such that
100% R + 100% G + 100% B - 100% white,
(A) Adjustment of red screen
(B) Adjustment of blue and
for cutoff.
green screens.
100%
R, G, & B
Coincident
50%
Ec
Green Gain
Set for gray balance
at cutoff.
Brightness pulses in
all channels are
clamped to
level.
Blue Gain
Red gain is fixed.
this same
,r
Ec
(Normalized)
(C) Adjustment of blue and
(D) Normalized curves
green gains.
(correct adjustments)
.
Courtesy RCA
Fig. 10 -27.
Procedure for setting color balance.
switch must, of course, be in the RGB position for this adjustment. As
shown in Figs. 10 -27A and 10 -27B, this step brings the kinescope transfer
characteristics into coincidence at the point corresponding to about 20 percent of the maximum signal level.
In the next position (monochrome) , both the brightness pulse and the
signal are applied, but the chrominance circuits are disabled. In this position, the green and blue gain controls may be set to provide proper color
balance in all parts of the gray scale. This adjustment is facilitated by the
use of a signal containing a gray -scale pattern.
As shown in Fig. 10 -27C, the absolute signal amplitudes required for
the three guns are different because of different phosphor efficiencies. When
the proper adjustments are made, however, and the signal scales are nor-
432
TELEVISION BROADCASTING CAMERA CHAINS
malized, the effective transfer characteristics are essentially coincident (Fig.
10 -27D) .
The next position, unity chroma, is the normal operating position, in
which the signal is applied to both the luminance and chrominance channels. The CHROMA control is inoperative in this position, and the saturation of the colors in the picture yields a good indication of the quality of
the incoming signal. The PHASE control may be set conveniently while the
TEST switch is in the unity chroma position by examining the blue component of a standard color-bar signal (use the B position of the SCREEN
SELECTOR switch). When the phase adjustment is correct, the standard
color -bar signal produces four blue bars of equal brightness. If the phase
adjustment is incorrect, the blue bars are of unequal brightness. This test
is very sensitive, particularly if the brightness is temporarily reduced to
place the blue bars near cutoff on the kinescope characteristic.
In the final position of the TEST switch, variable chroma, the conditions
are the same as for the unity- chroma position, except that the CHROMA
control is made operative. This position is intended for operation in applications where the monitor is used to make the most pleasing pictures, even
though the signals available are slightly substandard. The CHROMA control is simply set for the most pleasing overall effect.
The TEST switch can be used to make a rapid test of the convergence
adjustments in the monitor. If there is any uncertainty in the viewing of a
color picture as to whether observed misregistration is a fault of the signal
or of the monitor, it is only necessary to place the TEST switch in the monochrome position. If the color fringes disappear, they are clearly a fault of
the signal, but if they remain, it is necessary to touch up the convergence
adjustments of the monitor itself.
10 -5. THE X
AND Z DEMODULATOR
It will be recalled that the I and Q signal components contain some of
each of the R
and B
components, from which G
can be
extracted conveniently. The reader should already be familiar with demodulation with respect to the I and Q axes and the R
and B
axes.'
The latest and simplest form of "narrow -band color" demodulation is
-Y
-Y
-Y
-Y
-Y
termed X and Z demodulation.
See Fig. 10 -28. Through transformer T1, the local 3.58-MHz oscillator
provides signal voltages to the suppressor grids of demodulators V1 and
V2. The signal applied to the suppressor of V1 is in phase with the burst
signal and is arbitrarily termed the X signal. The voltage at the suppressor
of V2 is shifted approximately 90° by L1, Cl, and R1. This signal is
'For example, see Harold
E. Ennes, Television Broadcasting: Equipment,
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams
& Co., Inc., 1971) .
COLOR PICTURE MONITORING SYSTEMS
Fig. 10 -28. Basic circuit for X and Z demodulation.
434
TELEVISION BROADCASTING CAMERA CHAINS
termed the Z signal. The cw signals thus supplied provide the gating signals for the demodulators. The chrominance signal from the chroma band pass amplifier is applied to the control grids of V1 and V2 in parallel.
The demodulated outputs at the plates of V1 and V2, after filtering by
the 620 -µH coils and 10 -pF capacitors, are applied to chroma -difference amplifiers V3 and V5. The signal currents in the R -Y and B -Y amplifiers
combine in the common cathode resistor, R3, to develop the signal input
voltage at the cathode of G -Y amplifier V4. Note here that R2 and C2
at the grid of V4 serve only to couple a small amount of signal from the
plate of V3; this provides slight degenerative feedback to assure the correct phase and amplitude for the G -Y output signal. The picture -tube
retrace -blanking signal normally is coupled into the common cathode of the
chroma -difference amplifiers as shown.
As a result of the phase relationship of the X and Z demodulators, the
R -Y and B -Y currents of V3 and V5 develop across R3 a voltage that
is a G -Y signal. Remember that:
G- Y=- 0.51(R -Y) - 0.19(B -Y)
that a positive G -Y value is composed of negative
Thus, we note
values
of both R
and B Y. The cathode currents of V3 and V5 are in phase
with the input signal, and the cathode injection at V4 results in a plate
signal 180° out of phase with that at the V3 and V5 plates. This meets the
polarity requirements.
This type of demodulation occurs at a high level, and the output amplitude at the color -difference amplifiers is sufficient to drive the grids of the
picture tube. This fact, coupled with the relatively simple "matrix" circuitry just described, results in a relatively inexpensive but effective color
monitor as compared to monitors using the more elaborate I -Q demodulator. Also, the separate filtering and additional gain stages required in I -Q
demodulation are unnecessary with the X and Z demodulator.
-Y
-
10 -6. USE OF THE COLOR
MONITOR IN MATCHING TECHNIQUES
The final check for color balance in multiple-camera setups is accomplished most readily by using a switch bus for the color monitor and vector scope, as illustrated in Fig. 10 -29. With this technique, the effect of slight
differences in color monitors themselves is eliminated, since a common
color monitor is used for all signal sources.
It is assumed at this point that all cameras have been individually adjusted as correctly as possible. Very briefly, the results of these adjustments
should be as follows:
1.
Chroma and white level are the same from all cameras, and system
phase as well as burst phase are the same for all cameras at the
switcher output (on color bars)
.
435
COLOR PICTURE MONITORING SYSTEMS
All Same Length
To
Cam
1
DA
Cam
2DA
Studio Switcher Inputs
Cam 3 DA
Color
Monitor
Monitor Switch Bus
From Switcher Output
Fig. 10 -29. Technique for matching color cameras.
The same video output level is present from all cameras when the
pickup tubes are operating at the proper potentials and looking at
the same reference white chip on the gray scale. When image- orthicon
tubes are involved, this will be at the knee or slightly over the knee.
3. The gray -scale steps (looking at the same chip chart under the same
lighting conditions) are at the same levels relative to capped level
for all cameras. Slight adjustments in target voltage and/or gamma
circuitry are required to correct any differences.
4. Luminance -to- chrominance ratio and luminance -to- chrominance tracking are correct and the same for all cameras.
2.
We now are ready to color balance all cameras by using the common
color monitor as the final check. All cameras should be color balanced on
the same gray scale under the same light conditions. Use the following
procedure:
Switch between all cameras on the color monitor. If different "colors"
of the gray scale exist, go to the following steps.
2. Turn the camera chroma off. With a properly adjusted color monitor,
the background should be gray and show no color. Turn the chroma
on. Note any color in the background on the color monitor; balance
this output with a slight adjustment of the appropriate black -level
control on the camera. For example, if the background is slightly
green, adjust the green black -level control for cancellation of color. It
1.
436
TELEVISION BROADCASTING CAMERA CHAINS
is much easier to see this on a good color monitor than on the waveform oscilloscope. Repeat this procedure for each camera.
3. If the high -light chips are slightly colored, adjust the appropriate
"paint pot" (video vernier gain control) to cancel the color. Repeat
this step for all cameras.
4. The final check of the overall system (through the studio switcher)
is best made by using a split- screen presentation for two cameras at a
time. First check the system by feeding a signal from the same camera
(on color bars) to both banks of the special- effects bus with vertical wipe operation. There should be zero hue and saturation shift as the
vertical wipe is made on the same signal source. If a hue shift occurs, the switcher paths are not properly phase delayed. If a satura-
tion (color intensity) shift occurs, the two paths of the system have
different response at 3.58 MHz.
Then perform the same operation with two cameras looking at the
same chip chart. The color monitor should show no color shift as the
vertical -wipe transition is made.
EXERCISES
In an R -Y, B -Y receiver or monitor, why is only the luminance channel signal delayed?
Q10-2. Give the color-system primaries and their respective complementary
colors.
Q10 -3. Describe the chrominance signal for the corresponding complementary color.
Q10 -4. What is the purpose of the color burst?
Q10-5. What is the purpose of the bandpass amplifier in a monitor?
Q10 -6. What is the purpose of the color -killer circuit?
Q10 -7. Give the basic operation of a synchronous demodulator.
Q10 -8. Describe the difference between the cw reference signals used for
I -Q demodulation and those used for color- difference (R-Y, B -Y)
demodulation.
Q10-9. In a color-encoding system, what would be the effect of carrier unbalance on white and gray areas displayed on a properly adjusted
color monitor?
Q10 -10. In case of carrier unbalance in the encoder, will dominant hues be
shifted on the color monitor?
Q10 -1.
11
CHAPTER
Preventive Maintenance
Maintenance procedures for specific parts of a camera chain (such as
video amplifiers and preamplifiers, power supplies, processing circuitry,
yokes, etc.) have been covered in applicable sections of the preceding chapters. It remains to cover the general system checks of a camera chain, and
to suggest pertinent preventive -maintenance scheduling.
11
-1. THE VIDICON FILM CHAIN
When performance of a film camera chain deteriorates, there may be
some uncertainty concerning where to start a search for the problem. The
experienced maintenance technician is able to analyze the symptoms and
know approximately which area to explore. The following subsections
should orient the reader in the general techniques to be used in checking
performance specifications.
Steps in Obtaining Maximum Film- Camera Resolution
NOTE: Before proceding, review "Measuring Detail Contrast" in Section 6 -4, and "Checking Gamma Circuitry for Film Cameras" in Section
6 -6 of this book.
The problem here is to obtain maximum detail contrast (by means of
aperture correction in a properly adjusted camera chain) while maintaining the lowest possible noise level. You should perform the following steps
before undertaking the longer procedure of a complete video -sweep alignment; such alignment may not be necessary.
(A) Use open gate on the projector at normal lamp voltage and lens
settings. Adjust the target voltage to obtain a beam current of 0.3 .tA.
(B) See Fig. 11 -1. Turn any manual video -gain control to maximum
gain. Connect the oscilloscope at TP 1. You should know what peak -topeak level the camera head should deliver at 0.3 µA of beam current (this
usually is indicated on the schematic or in the instruction book) Usually
.
437
438
TELEVISION BROADCASTING CAMERA CHAINS
this level is around 0.4 to 0.5 volt peak to peak. If this signal level is low
under the above conditions, replace the camera -head tubes one at a time
until proper gain is restored.
If you do not know what this level should be, and /or if the camera
chain does not employ either a beam -current meter or a calibration pulse,
proceed as follows:
Assume the target load resistance is 47k. For a beam current of 0.3 .tA,
the voltage swing is: 0.3(10 -8) (47,000) = 0.014 volt (white to black).
Now see Fig. 11 -2. Temporarily disconnect the vidicon target lead (with
target voltage off), and substitute the signal as shown. If you adjust the
signal-generator output (which can be a 15.75 -kHz square wave, pulse, or
Control Chassis
Open Gate
on Proj
0.4
0.7
V
o
ECamera Head
Manua Gain
Control
Amp
-
Remote -Gain
Processing
Tube
Amp
V
Output
TP2
TP1
Remote Control
Fig. 11 -1. Camera -chain level check.
stairstep) to 1.4 volt peak to peak and use a 40 -dB pad, the input level to
the preamplifier will be the required 0.014 volt to simulate 0.3 µA of
target current. (It may be necessary to bypass temporarily the camera
head high -peaker cathode circuitry with a 0.47 -µF capacitor.) Now measure the peak-to -peak level at the input to the control chassis. Restore the
camera head to normal operation, and (with open projector gate) adjust
the target voltage to get the same peak-to -peak level as measured with the
signal generator. This target voltage is that which is required to obtain
0.3 µA of target current. Even if the equipment employs a metering circuit,
if you doubt its accuracy, check it by the above procedure.
(C) With a test -pattern signal, readjust any manual gain control to
obtain 0.1 to 0.15 volt (pk -pk) video (or the safe maximum value of
video found by the previous linearity test -Section 6 -6) .
(D) With a test- pattern slide or film loop, adjust all high -peaker controls
for minimum smear or streaking. Aperture compensation should be removed for setting these controls to observe only medium- and low frequency response.
1.4 V
pk-pk
0.014V pk-pk
40-dB
Attenuator
0.1
To
Camera -Preamp
Target Load Resistor
7552
Ext Sync if Desired
(Horiz Drive)
Fig. 11 -2. Connection of signal generator to target load resistor.
PREVENTIVE MAINTENANCE
439
(E) Restore the aperture compensation, and use the maximum boost
required to obtain 600 -line response with good detail contrast as observed
on a master monitor. If noise is apparent when the aperture compensation
is adjusted to obtain just 600 lines of horizontal resolution, do the following:
Check for excessive beam current. Reduce the beam until high lights
are just resolved.
2. Be certain of camera physical and electrical focusing.
3. Determine that the camera output is normal (Step B above) .
4. Replace video amplifier tubes in the control chassis one at a time
for reduction of noise without loss of resolution.
5. If there still is no improvement, the camera chain needs a complete
video alignment.
1.
NOTE: As a general rule, a "bad" vidicon will show excessive target lag
before it deteriorates in resolution. Target lag is image retention under
vertical or horizontal movement of the images. Such a vidicon can be
retired to a slide (only) camera, since no movement (except slide change)
is involved. After considerable extended use in the slide camera, the vidicon will go "soft" and lose resolving power.
Excessive overpeaking (by either video alignment or use of high -peak
controls to get white -following -black for apparent sharpness) will result
in video bounce (excessive on drastic scenic changes) and white or black
compression. In case of excessive bounce on scenic changes, check both
the high -peak and low- frequency compensation controls. Adjust the lowfrequency compensation control (where used) for a flat vertical interval
as observed on an external CRO at the output of the control chassis. Remember that most master monitors incorporate their own tilt controls ( a
separate adjustment) for the master monitor CRO circuitry.
Parameters of Target -Voltage Set
A surprisingly large amount of "maintenance" time is spent needlessly
when optimum adjustment procedures are not followed in the beginning.
Be sure you understand such procedures (normally considered under "operations" rather than "maintenance ") so that you know when actual maintenance is required.
Thus far, our discussion of the vidicon film camera serves only as a
starting point in obtaining good film telecasts. It has been assumed that
the projector -lamp voltage and lens f/ stop were set properly. Also, we
still have to consider focus current, beam alignment, and possible shading
problems.
Here are some "fine points" that help in "squeezing everything possible"
from the vidicon film setup:
440
TELEVISION BROADCASTING CAMERA CHAINS
Block all light from the vidicon faceplate, set the target voltage at
20 volts as a start, and turn the beam control off (maximum bias) .
2. With the target volts- current meter in the current position, rotate the
zero- adjust control so that the meter indicates upward on the scale.
(You do not want a zero reading because you are going to take a
difference measurement.)
3. While watching the CRO or the monitor kinescope, bring up the
beam current until the information at black level is "wiped -in" (dark current information is discharged) . This will be quite evident if you
watch closely.
4. Measure the difference in meter readings between 2 and 3 above.
For the best overall picture quality, this difference (actually a measurement of dark current) should be no more than 0.01 pA. Use
1.
the maximum target voltage possible while maintaining dark current
no more than 0.01 /IA. This is for film application only, not studio
use for live pickups.
5. Now zero the target meter with the beam control off, and remove the
light block from the vidicon faceplate. Project a resolution slide into
the vidicon. The camera lens should always be wide open. Adjust the
projector lens wide open (minimum depth of field) to obtain sharpest mechanical focus with the beam adjusted to just discharge the
highest high light. Then stop the projector lens down two stops from
wide open. This will provide adequate depth of field for warped or
buckled film.
6. Now adjust the projector -lamp voltage to obtain 0.3 p,A of beam
current with the gate open and with the target voltage arrived at in
Step 4 above.
You have now arrived at the optimum settings of target voltage, lens
aperture, and lamp voltage. NOTE: If the film chain incorporates a color
camera, you must maintain the projector -lamp voltage at its normal operating value to preserve color temperature. We are currently discussing
monochrome operation only. If a monochrome camera is multiplexed with
a color camera, you should use an additional neutral -density filter on the
monochrome -camera axis to obtain proper operation and light balance.
If the vidicon camera chain does not employ a target-current meter, or
if the metering circuit is not dependable in reading very small values of
current as is required in dark-current readings, keep the following in
mind: As the target voltage is increased, there is a point at which edge
flare (excessive dark current) is reached. When the target voltage is reduced below a certain value (depending on the individual vidicon) , a
point is reached at which portholing (darker around edges than at center)
occurs. It is possible to adjust the target voltage above the point at which
portholing just occurs, and an entirely "flat" raster (with respect to shading) results. This value then can be varied slightly if an unrealistic pro-
PREVENTIVE MAINTENANCE
441
jector -lamp voltage is required to obtain 0.3 µA of target current at the
target voltage being used.
NOTE: If you attempt to use too low a target voltage, a condition known
as target bounce can occur on drastic scenic changes. A small amount of
dark current (up to 0.01 MA for a film camera) stabilizes voltage excursions caused by target -current variations through the load resistor in
accordance with the charge on the photocathode. As a broad general rule,
optimum results in a monochrome film camera are obtained with a target
voltage somewhere between 18 and 25 volts.
Focus Current vs Focus Voltage
For a vidicon, the magnetic field strength at the center of the focus coil
normally should be 40 gausses (produced by 40 mA of focus current) .
For an image orthicon, the value is 75 gausses (produced by 75 mA of
focus current) . Therefore, the manufacturer usually recommends a fixed
value of focus current. However, with some tubes (either vidicons or
I.O.'s) it is possible to vary the ratio of focus current to focus -electrode
voltage slightly and obtain sharper delineation of the high- frequency
wedges of the test pattern. The focus -electrode voltage is, of course, what
is varied by adjustment of the beam -focus control (or orth -focus control
in RCA image-orthicon terminology) . This assumes that the beam current
has been aligned properly for optimum conditions, and that the proper
mode of focus exists for an image orthicon.
Remember that when the focus current is varied, the scanning size varies.
More current (greater focus field) stiffens the beam, and the photocathode
or target area is underscanned (kinescope image increases in size) . As the
focus field is reduced, the scanning beam travels farther for a given deflection voltage, and overscanning results (kinescope image decreases in
size). Therefore, after a change of focus current is made, remember to readjust the scanning for normal area before judging resolution of the test
pattern. You must, of course, readjust the beam -focus ( voltage) control
for optimum electrical focus each time you vary the current.
Linearity and Shading
The light output of a photoconductive device such as the vidicon is directly proportional to velocity of scan. So long as the scanning waveform is
linear, no inherent shading takes place in the vidicon. If the waveform is
nonlinear, the velocity, or rate of scan, is different for different areas of
the target, and shading occurs.
The best way to check a film camera is to remove the lens and swing
the camera to one side so that a flat-light source (such as a 40 -watt bulb
placed about 4 to 6 feet away) can be used. With a new vidicon (or an old
one that has not been abused), it is possible to adjust the linearity controls
for absolutely flat shading. In most cases, the image linearity can be made
well within 2 percent by this method. If there is shading when the camera
442
TELEVISION BROADCASTING CAMERA CHAINS
is returned to its normal position in the film chain, you know the shading
is the result of optics, not the camera. The above procedure assumes the
target voltage has been set for optimum operation as described earlier. It
is also assumed that you are already familiar with the conventional method
of checking camera sweep linearity: first getting the monitor linear by use
of the grating generator, then adjusting the camera sweeps (and linearity
controls) to obtain a linear presentation of the test- pattern image on the
same monitor.
A far better method of exactly measuring geometric distortion (non linearity of camera sweeps) is the use of the ball chart (RETMA [now
EIAI linearity chart). This aid can be obtained in slide form for a film
camera, or in chart form for a studio camera. The pattern is designed so
that the black outlines of the circles define 2-percent linearity, and the
inner portions of the circles define 1- percent linearity.
NOTE: The use of the linearity chart is illustrated and described in
Chapter 10 of Harold E. Ennes, Television Broadcasting: Equipment
Systems, and Operating Fundamentals (Indianapolis: Howard W. Sams
& Co., Inc., 1971).
11 -2.
1.
2.
3.
GENERAL PREVENTIVE MAINTENANCE
Preventive maintenance in general consists of:
Lubrication per manufacturer's specifications
Cleaning of all equipment and optics at periodic intervals
Performance checks at periodic intervals
The extent of the periodic interval is determined by manufacturer's recommendations, or as dictated by experience with any particular installation.
Cleaning
In general, cleaning procedures can be scheduled at daily, monthly, or
longer intervals, as follows.
Daily -Wipe and clean all exposed surfaces of the camera. Use only lens
tissue on turret lenses or zoom-lens front surfaces. When necessary, use a
cotton ball dipped in isopropyl alcohol, and use a light circular motion over
the optical surface. Wipe the surface dry with a wad of lint -free cheesecloth or diaper cloth. If warranted, clean the faceplates of pickup tubes in
the same manner. WARNING: Turn the power off when doing this.
Monthly-Inspect all exposed color optics and front -surface mirrors in
the optical path. If necessary, clean dichroic surfaces, but use only pure
grain alcohol or a 1 -to-6 ratio of benzene and grain alcohol. Use a cotton
ball and diaper cloth as described above. On front -surface mirrors, use isopropyl alcohol diluted with a small amount of distilled water (to prevent
rapid evaporation), and wipe dry with diaper cloth before the solution
on the mirror surface evaporates.
PREVENTIVE MAINTENANCE
443
NOTE: It is very important, as a final step in lens and optics cleaning, to
use a camel's -hair "static" brush to remove dust that inevitably falls on
the surfaces because of static deposits.
Every Three Months -This interval depends entirely on the cleanliness
of the operating area. In the case of portable and mobile use, these operations should be performed monthly (or more often) .
Clean all air filters in the blower path of the camera, and in the forced air paths of the cooling systems of power supplies, etc. Use only the solvent
for cleaning that is recommended in the instruction book for the particular
installation.
With a low- pressure air blower, blow out the camera head, control panels,
and rack equipment to eliminate dirt, dust, and general contamination.
Wipe clean with a soft paint brush and cloth.
IMPORTANT NOTE: At this time, it is imperative to make a thorough
visual inspection of component parts. Observe resistors for discoloration
or signs of heating. Inspect capacitors for signs of bulging or leakage.
Inspect terminal boards for cracks and loose connections. Be sure all fuse
mounts and fuse caps are tight. The time spent here is invaluable in preventing future breakdowns, which is the basic reason for performing preventive maintenance.
Lubrication
Lubrication of modern camera chains normally is scheduled about every
six months. For operation under extreme conditions of dust, heat, and
humidity, lubrication procedures must be carried out more often. Caution:
Some of the older camera chains employ fan and blower motors that require
lubrication every 200 hours. Always check the instruction book for a particular system, and work the lubrication schedule into a check -off sheet at
the recommended intervals.
Items that require lubrication include: turret shaft and detent rollers;
turret gears and bearings; camera -head counterbalance linkage; focus, iris,
and zoom mechanisms ( which usually involve multiple gears, bearings,
and shifts) ; and blower and fan motors. It is quite important to check the
manufacturer's specifications for the type of oil or grease to use on the
items involved.
It is important not to overlubricate. Always wipe off excess oil or
grease, and use any special applicators that are supplied with a camera
chain by the manufacturer.
Plugs and Receptacles
There are two main types of plugs and receptacles used to interconnect
the various components. The first type of plug is used with a coaxial line
and consists of a metal shell with a single, center pin that is insulated from
the shell. When the plug is inserted into the receptacle, this pin is gripped
444
TELEVISION BROADCASTING CAMERA CHAINS
firmly by a spring connector. There is a knurled metal ring around the plug;
this ring is screwed onto the corresponding threads on the receptacle. The
insulation in these plugs is heavy in order to withstand considerable voltage.
The second type of plug is used for connecting multiconductor cables.
The plug usually consists of a number of pins that are insulated from the
shell. The pins are inserted into a corresponding number of female connectors in the receptacle, although in some cases the plug has the female
connectors in it and the male connectors are in the receptacle. This type
of plug usually has two small pins or buttons that are mounted on a
spring inside the shell and protrude through the shell. When the shell is
properly oriented and placed in the receptacle, one of these pins springs
up through a hole in the receptacle, firmly locking the plug and receptacle
together. When it becomes necessary to remove the plug, the other pin
is simply depressed, and the plug can be removed.
Connections between all plugs and their cables are made inside the plug
shell. The cable conductor may be soldered to the pin, or there may be a
screw to hold the wire to the pin. Remove the shell if it is necessary to get
at these connections for repair or inspection. If there is a clamp holding the
cable to the shell, loosen the clamp screws. Usually there are several screws
holding the shell; these are removed and the shell is pulled off. In some
cases, it is found that the shell and plug body are both threaded; then the
shell may simply be unscrewed.
Inspect the following:
1. The part of the cable that was inside the shell for dirt and cracked
or burned insulation
2. The conductor or conductors and their connection to the pins for
broken wires; bad insulation; and dirty, corroded, broken, or loose
connections
3. The male or female connectors in the plug for looseness in the insulation, damage, and for dirt or corrosion
4. The plug body for damage to the insulation and for dirt or corrosion
5. The shell for damage such as dents or cracks and for dirt or corrosion
6. The receptacle for damaged or corroded connectors, cracked insulation, and improper electrical connection between the connectors and
the leads
Tighten the following items:
1. Any looseness of the connectors in the insulation. If tightening
is not
possible, replace the plug.
2. Any loose electrical connections. Resolder if necessary.
Clean these items:
1. The cable, using a cloth and cleaning fluid
2. The connectors and connections, using a cloth and cleaning fluid. Use
crocus cloth to remove corrosion
445
PREVENTIVE MAINTENANCE
The plug body and shell, using a cloth and cleaning fluid. Use crocus
cloth to remove corrosion.
4. The receptacle, with a cloth and cleaning fluid if necessary. Use crocus
cloth to remove corrosion.
3.
Adjust the connectors for proper contact if they are of the spring type.
Lubricate the plug and receptacle with a thin coat of petroleum jelly if
they are difficult to connect or remove. The type of plug with the threaded
ring may especially require this.
11
-3. TROUBLESHOOTING
Conditions requiring emergency maintenance can be classified broadly
into three categories:
1.
Erratic video level
2. Erratic black level
3.
Erratic horizontal and /or vertical deflection
A dead camera chain or a combination of all three of the above conditions
generally points to the common unit for the entire chain -the power supply. Maintenance of power supplies is covered in Chapter 3.
Erratic Video Level (Black Level Constant)
Review Fig. 6 -25. The first step is to check the camera chain with test
pulses, when these are provided. If the pulse inserted at the preamp input
is erratic relative to the reference pulse ( inserted at the camera end of the
cable) , then the trouble is obviously in the preamp, video amplifier, or
processing amplifier in the camera head. If both the inserted and reference
pulses are erratic, the problem is in the control console or control -room rack
equipment. Test points normally are provided at each module output for
quick scope observation. If the pulses are stable, but video with the pickup
tube looking at a scene is erratic, the problem lies in the pickup -tube
circuitry.
In a multiple -channel color camera, trouble of this nature is usually existent in only one of the channels at a time. If the trouble is exhibited in all
the
channels, then it occurs at a point after combination of all channels
encoder. In this case, the B, R, G, and M monitoring points will show no
level changes on an individual channel.
Review Fig. 6 -28 for typical circuitry involved in remote control of video
gain. Since the condition discussed here is erratic video level with constant
black level, the problem will normally exist prior to black -level clamping.
-in
NOTE: Bear in mind here that if the trouble is following the camera head output, both video and black level would be erratic (in the typical
arrangement illustrated in Fig. 6 -25).
Review Chapter 7 for automatic and manual control circuitry.
446
TELEVISION BROADCASTING CAMERA CHAINS
Erratic Black Level (Peak -to -Peak Video Level Constant)
Now we will consider the case in which the overall amplitude is erratic,
but the peak -to -peak video remains constant with the changing pedestal.
Again, the first step is checking the test pulses. This problem sometimes
results when a clamp -pulse width or timing adjustment is on the edge of
the proper setting. The maintenance engineer should familiarize himself
thoroughly with all such adjustments for the equipment in the installation
with which he is concerned.
Check first for the existence of clamp pulses. Then, using video of varying APL's, check for the effectiveness of clamping at the clamped point.
Sometimes the black -level control pulses are fed into a video stage prior
to the clamp, and are timed (by adjustment or design) to occur within
the clamp interval. Check these points on a dual -trace scope.
Obviously, the wide variety of circuitry employed in different camera
chains prevents establishment of a set pattern of testing. The maintenance
engineer must study and analyze his particular instruction -book descriptions. Any point that is not clear is sufficient reason to contact a factory
representative for clarification.
Deflection Problems
It will be recalled that if either the horizontal or vertical deflection fails
entirely, the pickup tube (s) will be biased off, and no video will be obtained. When this condition exists, the test pulse is passed, but no video is
obtained from the pickup tube. In this case, the first place to check is the
protection stage (review Fig. 7 -22); observe the presence or absence of
horizontal- or vertical- deflection pulses, and check back from there. In the
event of erratic deflection, location of the trouble in the horizontal or vertical circuits is self- evident from the monitor display.
See Fig. 11 -3 for a review of a typical four -channel color-camera deflection arrangement. When different types of tubes are employed ( for example, an I.O. for luminance and vidicons for chrominance) , different deflection amplitudes must be provided. Fig. 11 -3 shows a typical arrangement for both horizontal- and vertical-deflection circuits. Master size and
linearity adjustments affect all four channels. These generally are set for
the monochrome tube. Deflection for the color -channel tubes then is obtained through a waveshape and attenuation network.
We can see from this configuration that the green size control becomes
the master for all chroma tubes, and individual controls are provided for
red and blue. Linearity controls for red and blue are variable resistors in
series with the deflection coils. This arrangement is for the purpose of fa-
cilitating registration.
If all channels show erratic deflection, the problem obviously is located
prior to the multiple yoke take -off: in the deflection output stage or ahead
of this stage. If the erratic deflection is in one channel only, troubleshooting
447
PREVENTIVE MAINTENANCE
To
Mono Deflection Coil
Deflection
Input
Attenuation Network
G
Size
To
R
Green Deflection Coil
Size
To Red
B
Size
To
Deflection Coil
Blue Deflection Coil
Fig. 11 -3. Basic deflection arrangement for four -channel camera.
is isolated to the corresponding network. Review Section 7 -4 of Chapter 7
for typical deflection circuitry.
EXERCISES
Q11 -1. How can you tell whether shading is introduced by the pickup tube,
or by the amplifiers and /or shading controls?
Q11 -2. What is the most common cause (other than "optics ") of shading in
the vidicon film camera?
Q11 -3. How many adjustments, in addition to the shading controls, affect
shading in the image- orthicon camera?
Q11 -4. Why should nearly all adjustments that affect shading in the image
orthicon camera be made with the lens capped?
Q11 -5. What type of tube checker should you use for preventive maintenance,
and what should you check for?
Q11 -6. If a slightly excessive beam current is used on a vidicon and a split
image results, what are the two most likely sources of trouble?
Q11 -7. You have noticed that when setting up I.O. cameras outdoors with
long camera -cable runs, you seem to be continually "running out of"
image -focus (photocathode focus) range, particularly after the cables
have been heated by direct sunlight. Can you do anything about this?
Q11 -8. What are the problems most likely to be associated with the image orthicon tube itself, rather than amplifiers?
APPENDIX
Answers to Exercises
CHAPTER
1
A1-1. (A) 30 W /ft2. (B) 90 W /ft2.
Al -2. (A) 15 W /ft2. (B) 50 W /ft2.
Al -3. (A) 720 foot- candles. (B) 640 foot -candles. In solving this problem,
use the information of Table 1 -2. Note that the multiplying factor for
illumination is 4 (15 feet is one -half of 30 feet, and light decreases as
the square of the distance). Thus, for (A) the chart of Fig. 1 -2B shows
180 foot- candles, and 180 X 4 is 720 foot -candles. Also, since the distance is halved, the dimensions are halved, and the off -axis dimension
to use with Fig. 1 -2B becomes 10 X 0.5, or 5 feet. Since the chart
in Fig. 1 -2B shows 160 foot-candles, then the answer is 160 X 4, or
640 foot- candles for (B) .
Al-4. (A) 45 foot -candles. (B) 20 foot- candles.
A1-5.
I= W/E = 5000/115 = 44 amperes
(approx)
.
CHAPTER 2
A2 -1.
Remember the color triangle; green and blue form cyan. From Table
2 -1, for cyan:
I=-0.60
Q=-0.21
The vector sum is:
Vl(- 0.6)2+
(- 0.21)2 = x/0.36 +0.04=
0.4
=0.63
This is the amplitude of fully saturated cyan (unity green and blue
inputs, zero red) .
Now see Fig. A -1. If you take
as the "adjacent side," then:
-I
cot B
B
side
- adjacent
opposite side
I
-0.6
Q
-0.21
= 19.4° from -I
axis
448
2.86
449
ANSWERS TO EXERCISES
Since the
-I axis is at 303°
303
A2 -2.
A2 -3.
42 -4.
and cyan is lagging this by 19.4 °:
- 19.4 = 283.6, or simply 284°
If these I and Q signals were applied to the grids of a color kinescope
that was perfectly linear (not only to cutoff, but beyond cutoff into
the negative light region) , they would swing equal amounts above
and below cutoff, and the net luminance would be zero. But the kinescope is not linear, and it cuts off at black (blanking or picture black,
depending on monitor adjustments). Therefore the ac axis is no longer
zero, and the bars become visible, since these pulses are occurring
during active line scan.
Any pulse into the red input of the encoder will cause only the red
gun of the color picture tube to be excited. (The same is true for the
other inputs and their respective guns.) If this red input occurs at a
time when the green and blue inputs to the encoder are zero, only the
red gun will be activated; therefore the red is fully saturated regardless
of amplitude. If there is no mixture with white (meaning a measurable amount of the other two primaries) , then, by definition, the color
is fully saturated. This normally occurs only with a signal from a
color -bar generator. Color -bar pulse amplitudes can be reduced to
75 percent of unity value at the encoder input, to prevent 'overshoots"
when transmitter color specifications are being checked. Primary signals and their complements are still at maximum saturation.
The value indicated on detail C -C of Fig. 2 -10B is the minimum allowable front porch. Now study Note 4 of Fig. 2 -10A. Put down
the stated values of x, y, and z from Fig. 2 -10B:
= 0.02H = Minimum front porch
= 0.145H = Minimum without front porch
= 0.180H = Maximum with front porch
x -I- y = 0.02H + 0.145H = 0.165H = Minimum with front porch
x
y
z
+1
123°
+Q
33°
00
0.21
0.6
213°
-o
284°
Cyan
Fig. A -1. Vector for cyan.
303°
-1
450
TELEVISION BROADCASTING CAMERA CHAINS
-
A2-6.
The maximum- minimum range is 0.180H 0.165H = 0.015H. If
0.02H is minimum, then 0.02H + 0.015H = 0.035H maximum front
porch, not accounting for rise and fall times of blanking. But allowance for these times must be made. The maximum specified rise and
fall time is 0.004H. So 2 X 0.004H = 0.008H can be taken up by
this time. Then 0.035H
0.008H = 0.27H is maximum front porch
width. To convert to microseconds:
Minimum = 0.02H = ( 0.02) (63.5) = 1.27µs
Maximum = 0.027H = ( 0.027) ( 63.5) = 1.71 µs
Standard = 0.025H = (0.025) (63.5) = 1.59 µs
The Tektronix Type 524 oscilloscope provides a 0.025H marker for
the purpose of setting the front -porch width.
The color -sync burst normally is maintained at the same peak -to -peak
amplitude as the sync pulse. If the sync is adjusted to 0.3 volt (above
blanking) , then the burst amplitude is 0.3 volt peak -to -peak. The FCC
specification is 0.9 of sync amplitude to 1.1 of sync amplitude.
No. It is eliminated during the 9H interval of equalizing and vertical -
A2-7.
sync pulses.
8 to 10 complete cycles.
-
A2 -5.
0dB
-6dB-
Fig. A -2. Ideal detector -diode curve.
0 2
0.75 1.25
4.18 4.5
MHz
A2 -8.
Note from the statement concerning this subject in the FCC rules that
the color -transmitter response is tightly controlled at the color sub carrier frequency of 3.58 MHz. Note also that the applicable specifi-
cations are tied to the "ideal detector -diode curve" shown in Fig. A -2.
The first 0.75 MHz of the modulation frequency range is double sideband; hence the diode response is 100 -percent in this region. Due
to vestigial -sideband transmission, the diode response at 1.25 MHz
will be down 6 dB if the transmitter has 100- percent response to this
frequency. In essence, then, the actual attenuation -vs- frequency response of the color transmitter must be within plus or minus 2 dB
for frequencies up to 4.18 MHz (using response at 200 kHz as a
reference) to meet FCC requirements.
A2 -9. No. The camera head normally employs a gamma of less than unity
(black stretch) .
A2 -10. Yes. If not, luminance distortion (which also affects colors) will occur.
A2 -11. Phase sensitivity.
451
ANSWERS TO EXERCISES
A2 -12. Multipath reflections with attendant phase shift at various frequencies
in the radiated signal. Also, of course, the receiver circuits can affect
color reproduction.
A2 -13. 1. Yellow would go greenish.
2. Cyan would go bluish.
3. Green would go bluish (cyan) .
4. Magenta would go reddish.
5. Red would go yellowish.
6. Blue would go toward purple (magenta)
7. White would remain white.
A2 -14. Errors in the input -signal amplitudes at the encoder (all inputs must
be equal for white) . Improper I or Q white balance adjustments.
Carrier unbalance. I or Q coefficient errors in the receiver.
A2 -15. Low overall chroma gain.
A2 -16. Good human "flesh tones."
A2 -17. A red with reduced luminance value. This could be caused by lack of
proper gamma correction (black stretch) in the camera chain, or by
video unbalance in the encoder.
A2 -18. Differential gain, in which the chroma gain decreases with an increase
in brightness.
.
CHAPTER
3
A3 -1. The panning head, or cradle.
A3 -2. (1) Center -of- gravity adjustment. Adjusted for camera balance. (2)
Pan -drag adjustment. Adjusted for proper friction in panning. (3) Tilt drag adjustment. Adjusted for proper friction in tilting of camera head.
(4) Pan brake. Locks mount in azimuth. (5) Tilt brake. Locks mount
vertically.
A3 -3. Two or more prompters on which the copy moves together, line for
line, regardless of speed in forward or reverse direction.
A3 -4. The purpose is to cause the rectifier output to change, in coordination
with the output voltage, in such a direction that minimum voltage drop
across the series regulator (s) occurs. This reduces power dissipation in
all series regulator elements.
A3 -5. Remote -sensing leads connect from the supply to the load end of the
camera cable, and therefore permit sensing the voltage at the load
rather than at the supply terminals. Thus, the load voltage can be held
constant regardless of the length of the camera cable.
A3 -6. So that any RFI components occur during blanking and do not appear
as beat patterns during active scanning.
CHAPTER 4
A4 -1.
A4 -2.
A4 -3.
To magnify electromagnetically the image charge on the target area.
By means of the magnetic field produced by the external focusing
coil, and by varying the photocathode voltage.
By the magnetic field of the external focusing coil, and by the electrostatic field of grid 4.
452
A4 -4
TELEVISION BROADCASTING CAMERA CHAINS
Yes, the grid -3 voltage adjustment facilitates maximum possible collection by dynode 2 of the secondary electrons from dynode 1. It is adjusted for maximum signal output consistent with best freedom from
shading.
Grid 5 serves to adjust the shape of the decelerating field between
grid 4 and the target. It is adjusted to obtain best uniformity of electron
landing over the target area, that is, for best corner resolution and
minimum shading.
A4 -6. No, not for the same signal -to -noise ratio. More aperture compensation can be used with the vidicon before noise becomes troublesome
than is the case for the image orthicon.
A4 -7. Sweep linearity adjustment.
A4 -8. Dark current increases with an increase in target voltage.
44 -9. Dark current is practically the same at any target voltage.
A4 -10. None. (Plumbicon is the registered trademark of N.V. Philips of
Holland for their lead -oxide vidicons.)
A4 -5.
CHAPTER 5
A5 -1
Because the input coupling from the pickup tube to the preamp is at
high impedance, with shunt capacitance (both output capacitance of
the pickup tube and input capacitance of the preamp) . Also, peaking
circuitry to flatten the amplifier response is frequency -selective, and RC
or LC coupling causes a phase shift between low and higher frequencies. Hence, correction circuitry is needed to make all signal delays for
low and high frequencies the same.
AS -2. Only when it feeds the coaxial line in the camera cable. When another
amplifier after the preamp is located in the camera head, the output
impedance usually is 93 ohms.
A5 -3. No. Peaking circuitry only corrects the frequency response of the amplifier itself. Aperture correction takes place in a following video -
processing amplifier.
A5 -4. The dynode -gain control for the image orthicon, or the target -voltage
control for the vidicon.
AS -5. An attenuation of 40 dB is a voltage ratio of 1 /100. This amount of
attenuation is most convenient in feeding pickup -tube video preamps,
since it allows normal output levels from the test equipment.
A5 -6. Insufficient beam current to discharge the highest high lights.
A5-7. Excessive beam current causes loss of resolution and, under some conditions, a "splitting" of the image. In some cases, an effect similar to
"target flutter" in an I.O. occurs in the corner of the picture. This also
can be caused by improper beam alignment.
CHAPTER 6
A6 -1.
Phase shift is the additional phase angle between the input and output signal voltages over and above the normal 180° phase reversal of
the stage. This may be expressed by the expression:
Phase shift
=O
- 180°
453
ANSWERS TO EXERCISES
In this equation O is the total displacement, expressed in degrees, between the input and output voltages. The low- frequency phase shift
can be calculated directly from circuit constants as follows:
(1)=-{-tan-1
Xe
RG
where,
is the phase shift,
Xc is the reactance of the coupling capacitor,
RG is the grid resistor of the following tube.
4)
This formula simply states that the phase shift is an angle whose tangent is the ratio of the capacitive reactance to the grid resistance. The
plus sign indicates a leading phase shift; this is true because capacitive reactance causes the current to lead the applied voltage, and a
leading voltage is developed across the grid resistor. It can be seen
that the larger the capacitance (less reactance) , the smaller will be the
resultant phase shift.
A6 -2. The coupling capacitance.
A6 -3. Low -frequency degeneration in the cathode bias circuit; insufficient
time constant (RC) in the coupling circuit; degenerative feedback
through the power supply; changed components in negative- feedback
circuits; faulty clamping in clamp-type amplifiers.
A6 -4. Increased stray capacitance to ground (detrimental to high- frequency
response) and a tendency to "motorboat" at a low frequency.
A6 -5. If the level of the 60 -Hz square -wave signal is excessive, the resultant
clipping or compression will remove the tilt, and a poor response may
appear to be satisfactory.
A6 -6. Since you know that the fundamental frequency for a T pulse (one
picture element) for a 4 -MHz system is 4 MHz, the duration of two
picture elements corresponds to one -half the T frequency, or 2 MHz.
The 2T pulse contains one-half the harmonic spectrum of the T pulse.
46 -7. The repetition rate is the line -scanning frequency, 15.75 kHz.
A6 -8. Yes, plus an extra equalizer for the viewfinder feed.
A6 -9. Not necessarily. It is used only in the luminance channel in systems
fixed for operation with one luminance channel and three chroma
channels. In some film systems, however, voltage may be removed from
the luminance tube, and the system may be operated with three
channels (luminance derived from the chroma channels by matrixing) .
In the latter case, aperture correction is available for all channels.
A6 -10. Emitter voltage = +1.8 V (approx)
Emitter current = 1.8/500 = 0.0036 A = 3.6 mA (approx)
Collector -load voltage drop = (3.6) (3.3) = 11.8 V (approx)
11.8 = +8.2 V (approx)
Collector voltage = +20
Voltage gain = 3300/500 = 6 (approx)
A6 -11. Output peak -to -peak signal = (5) (0.2) = 1 volt (or slightly less,
depending on the input impedance of the following stage).
A6 -12. Q1 base voltage = +0.7 V (approx)
Q2 emitter current = 10.7/10,000 = 0.001 A = 1 mA (approx)
Voltage across 121 = (0.001) (3000) = 3 V (approx)
-
454
TELEVISION BROADCASTING CAMERA CHAINS
Q2 emitter voltage = +0.7 + 3 = +3.7 V (approx)
Therefore, Q2 base voltage (Q1 collector) = +3.7 + 0.7
(approx)
= +4.4
V
A6 -13. The maximum current swing would be ±1 mA in 75 ohms. Therefore,
the maximum peak-to -peak output voltage is (0.002) (75) = 0.15 V.
A6 -14. Nonuniformity of target or photocathode sensitivity across the useful
scanned area; nonuniformity of light incident on the chip chart; defective or dirty chip chart; defects or improper adjustments in optics,
shading, multiplier focus, or beam alignment.
CHAPTER 7
A7 -1.
A7 -2.
A7 -3.
A7 -4.
A7 -5.
A7 -6.
A7 -7.
A7 -8.
So that the cable delay between the control
unit and the camera will
fall within the duration of horizontal blanking. (Camera blanking is
formed from the drive pulses.)
No. If the RC time constant is long, the output-pulse duration is the
same as the input -pulse duration.
The output -pulse width is the product of the capacitance of the
coupling capacitor and resistance of the base resistor. The output pulse
is not narrower than the input pulse if the RC time constant is long
compared to the input -pulse duration.
Undelayed.
Undelayed.
No, not within the usual operating range of the target voltage.
Yes.
Yes. In most late -model cameras, the same circuitry is used in either
mode; the difference is in the source of the error voltage.
A7 -9. Slight differences between cameras in spectral distribution of the
dichroic systems. Differences in accuracy in achieving a white balance.
Differences in I.O. landing and shading characteristics. Differences in
characterstics of amplitude versus gray scale. Also, since two or more
cameras cannot have the same viewing angle for any subject, slight
color differences in the scene actually do exist.
A7 -10. Black level, gamma correction, multiplier focus, shading, and dynode
gain.
A7 -11. Black level and amplifier gain.
CHAPTER
A8 -1.
A8 -2.
218 -3.
8
No more than 20 to 1 for ideal control. The range can be determined
by measurement with a spot- brightness meter. Also, the whitest material should give the same amplitude on the CRO as that given by
reference white on the chip chart under the same light. The blackest
material should match step 9 on the chart; this is 3- percent reflectance.
Thus, the ratio of 60 percent (white) to 3 percent (black) is 60/3,
or 20 to 1.
15 IEEE units.
Optical and electrical focusing. Amount of aperture correction used
(determined by signal /noise ratio). Scene lighting and contrast ratio.
Luminance -to- chrominance ratio. Camera registration. (You should
455
ANSWERS TO EXERCISES
A8 -4.
A8 -5.
48 -6.
A8 -7.
A8 -8.
A8 -9.
A8 -10.
A8 -11.
A8 -12.
A8 -13.
A8 -14.
A8 -15.
A8 -16.
A8 -17.
be able to see 400 to 450 lines of horizontal resolution at the center
of a test pattern.)
On three -channel systems not employing the cancellation technique,
adjust the final registration for 400 lines minimum horizontal resolution at the center of the test pattern. In the cancellation technique, the
polarity of all video signals except green is reversed. Thus, when a
negative picture from any one channel is combined with the positive
picture from the green channel, adjust the channel being compared
to green so that complete cancellation takes place at least through the
large center circle of the registration chart.
How it looks in monochrome. Does it have "snappy" contrast?
Yes, it is. On the small area covered by the tight shot, the color temperature of the light can be different, especially with old incandescent
lamps. Also remember the effect on skin tones of backgrounds; a
tight shot (little background) can eliminate this effect.
There is a way you can rebalance the camera on such shots (if
necessary) during operations, if you have a reference white and black
in the scene. Have the camera monitor on NAM monitoring, and
carefully adjust for NAM balance on black and white (black and
white balance controls). This takes some experience and is not recommended unless the balance is very noticeably bad.
Degree of saturation (purity) of background. Area of background
relative to face: for a facial closeup, the background has minimum
effect; if the background is comparatively large (skin area small relative to background) , the effect is maximum.
±-100 K, or ±-10 volts.
Yes. In fact, more dimmers sometimes are necessary for color than
for monochrome. This is because areas in which skin tones exist (where
performers will be facing the camera) should not be dimmed. Lighting in other areas can be manipulated by dimmers for special effects
and mood scenes.
(A) 1/2 to 1 times base light.
(B) A maximum of 11/2 times base light.
In the blue region.
It helps by increasing the response toward the red end of the spectrum.
In parallel.
In series.
Skew is a condition in which the raster in a channel (or channels) of
a color camera does not have the same shape as the raster in the reference channel. It results from slight differences in deflection yokes. It is
corrected by introducing a small amount of vertical sawtooth (of the
proper amplitude and polarity) into the horizontal sawtooth.
60- percent reflectance.
They adjust the ratio of resistance to inductance to achieve similarity
(linearity) of the horizontal- deflection circuits.
CHAPTER 9
A9 -1.
No. It requires the higher- frequency trigger input only for "locking"
to the subharmonic.
456
A9 -2.
A9 -3.
A9 -4.
A9 -5.
A9 -6.
A9 -7.
A9 -8.
A9 -9.
TELEVISION BROADCASTING CAMERA CHAINS
The subcarrier frequency is too high; the sync generator is on crystal
control ( instead of subcarrier- frequency control ) ; or the sync generator
is on free -run operation.
No. The subcarrier- frequency oscillator is a temperature -controlled
crystal circuit. It is improbable ( after normal operating temperature
is reached) that an error greater than ±-50 Hz will exist. (Tolerance
is ± 10 Hz.) But even assuming a -!- 100 -Hz error, since the total
division is 113.75, the final error would be 100/113.75 = 0.87 Hz. So
once the counters are properly centered, the frequency adjustment can
be made at any time, as the count is quite broad.
The frequency counted down from the 3.579545 -MHz oscillator is
31.46852 kHz. The countdown usually is done by three divider stages
(5 X 7 X 13 = 455) and a X4 multiplier, to make the total division 455/4, or 113.75, times. Double check this by multiplying:
31.46852 kHz X 113.75 = 3.579545 MHz (to the nearest hertz).
No. The phase of the color -subcarrier burst is adjusted in the encoder.
The flag pulse influences only the timing (position) of the burst.
The best way is to use a time base of 0.1 µs /cm and use 0.1 -µs markers. Trigger the scope with horizontal -drive pulses. Or, you can use a
time base of 1 ,us/cm if scope- triggering instability occurs on the extremely short time base. Also, you can use 10 ,us/cm with X10 sweep
magnification. There should be five of the markers between the trailing
edge of horizontal sync and the first (full) cycle of burst.
If bursts occur during the entire 9H interval, obviously the 9H -eliminate pulse has been lost completely. However, if some of the bursts
are eliminated, but not for nine complete lines, the burst -eliminate
pulse has too much slope, and the eliminate keyer (V3) does not remain off for the entire interval. The most likely cause is insufficient
clipping at V2. This could result from a low output level from V1,
leakage in coupling capacitor C3 or bypass capacitor C4 (either of
which would reduce the bias), or leakage in coupling capacitor C5. The
first check should be to observe the pulse at the ViB plate to see if
the multivibrator is capable of obtaining a 9H pulse width. If it is, a
badly sloping pulse from the clipper is likely, as noted above.
18 peaks.
Not at all. But the 9H- eliminate pulse would be lost, and bursts would
continue throughout the vertical interval.
A9 -10. If all inputs are tied together, starting with the first (white) interval,
a continuous "white" pulse is sent for the line duration, since at any
time at least one of the multivibrators will be sending a pulse. So with
Y off and I and Q on, if the encoder is properly balanced, only a thin
line representing the subcarrier zero axis should be observed at the
output. (The subcarrier should be zero for any "monochrome" condi-
tion.)
A9 -11. The switches for the Y, I, and Q channels control only the
video channels. Thus, with all channels off, there still is a
output if the subcarrier is not properly cancelled out for a
(black) condition. Remember this means both the I and
rier- balance controls are involved.
A9 -12. The white -balance control adjusts the gain of the matrix
respective
subcarrier
no- picture
Q subcar-
signal in-
457
ANSWERS TO EXERCISES
verter so that all three signals are amplified identically. Note that this
implies identical amplitudes at the input, as well as maintenance of
proper values of the precision resistors used in the matrix network.
49 -13. (A) For the I -only chroma signal, the two maximum -amplitude signals are cyan and red. The I video component of red and cyan is 0.6
of unit luminance. So when the subcarrier is modulated, the sidebands
reach twice this value, or 1.2 of unit luminance. Since 100 IEEE units
= 0.714 volt, 1.2 X 0.714 = 0.85 volt peak to peak. This is 120
IEEE units.
(B) For Q only, the maximum -amplitude peaks are green and magenta. The Q video component of green and magenta is 0.525 of unit
luminance; with modulation, the sidebands reach 1.05 of unit luminance. Since 100 IEEE units = 0.714 volt, 1.05 X 0.714 = 0.75
volt peak to peak. This is 105 IEEE units.
Whenever you are in doubt concerning problems with I and Q gain
ratio, you can set up the encoder gains by this method to determine
whether either one is in error. When you then turn all channels on
and adjust the composite gain for 0 -100 IEEE units on the white pulse,
the chroma signals should be correct in absolute amplitudes as well as
gain ratios.
A9 -14. If sync were inserted before aperture compensation, it would be overcompensated and would contain spikes on the leading and trailing
edges. If sync were inserted after the Y delay, it would need to be delayed externally to match the encoder delay.
A9-15. No. But the effect of luminance can be observed if the Y channel in
the encoder is turned off with both I and Q on. Any change in amplitude of the individual vectors indicates differential gain. A change in
phase of the vectors indicates differential phase.
CHAPTER 10
A10 -1.
Since the demodulation of color -difference signals is narrow -band,
only the chroma frequencies below 500 kHz are used. Hence, identical delays occur in the color channels, and only the Y signal need be
delayed to achieve time coincidence.
A10-2.
A10 -3.
Primary
Complementary
Red
Cyan
Green
Blue
Magenta
Yellow
The chrominance signal for the primary color has the same amplitude
as the signal for the complementary color, but differs in phase by
180 °.
410 -4. To synchronize the receiver or monitor local 3.58 -MHz oscillator
A10 -5.
A10 -6.
with the subcarrier- frequency oscillator at the sending point.
To amplify and pass only those frequencies between approximately
2.3 and 4.2 MHz, where the color -subcarrier sidebands lie.
To gate off any input to the chrominance section during a monochrome transmission.
458
TELEVISION BROADCASTING CAMERA CHAINS
A10 -7.
It detects the amplitude variations of one phase of a multiphase
modulated carrier.
A10 -8. The basic difference is a 33° phase shift relative to the color burst.
The cw reference for I leads the one for R -Y by 33 °, and the reference for Q leads the one for B -Y by 33 °.
A10 -9. We know that when the system is normal, white or gray areas appear
during a signal interval of zero subcarrier (I and Q are cancelled
out) . If either one of the modulators (I or Q) becomes unbalanced
during the active line interval, a white or gray area will become
colored as a result of the addition of the subcarrier vector during
this interval. Also, during intervals of the scan, the subcarrier may be
cancelled by the unbalanced carrier vector, and the color may become
desaturated or white. The overall result is a white -to -color and color to -white error that changes with picture content and is quite objectionable.
A10 -10. Yes. We know that the unwanted carrier has a constant amplitude
that is added vectorially to every color vector. A positive unbalance in
the I modulator shifts all hues toward orange; a negative unbalance
shifts them toward cyan. A positive unbalance in the Q modulator
shifts all hues toward yellow -green; a negative unbalance shifts them
toward purple.
CHAPTER
11
A11 -1. Any shading component originating in the pickup tube (particularly
in the I.O.) will vary as a function of the light amplitude, which becomes video amplitude at the output. The shading -generator signal
remains fixed at all light levels or video amplitudes.
A11 -2. Nonlinear sweeps. A perfect sawtooth current through the deflection
yoke has a constant rate of change and, hence, produces a constant velocity of scan across the target. Any departure from a sawtooth current
waveform means the rate of change is not constant, and shading will
result.
A11 -3. Assuming the tube is operating in the proper mode of focus, the adjustments for grid 3 (multiplier focus) and grid 5 (decelerator grid)
are the most critical. (Adjust the grid -3 voltage with the lens capped.)
Proper use of the alignment controls also may improve shading characteristics.
A11 -4. Good shading characteristics depend primarily on uniform collection
of the secondary electrons as the return beam scans a small area (about
IV4 inch) of the first dynode. Variations of the secondary- emission
ratio over the first dynode are amplified in the remaining dynode sections. The most severe case of shading is represented with the lens
capped, since the beam is completely returned to the first dynode.
Since the amplitude of the shading component is a function of the
return beam, it is greatest in dark areas, where the return beam is
maximum.
A11 -5. Use a reliable dynamic -transconductance type with provisions for
checking high- resistance interelectrode shorts, and provisions for
testing voltage -regulator tubes. It is not necessary to keep a record of
ANSWERS TO EXERCISES
459
measured tranconductance. Use the GOOD -BAD scales, which automatically include the manufacturer's tolerance ( even on new tubes) . Run
the shorts check, and test all voltage- regulator tubes for firing voltage
and regulation within the specified current range. This type of tube
testing should be done about every 90 days. After tube replacement,
remember to run a complete performance analysis on the units involved, making any necessary adjustments. Recheck after a four -day
run -in time.
A11 -6. This is a result of low emission caused by either low filament voltage
or a weak vidicon tube.
A11 -7. The first thing you can do is to lower the focus current for the duration of the remote telecast in order to get a reasonable picture. For example, if you normally use 75 mA of focus current, reduce this to
between 70 and 72 mA, and readjust sizes, alignment, etc. The picture sharpness may suffer slightly but not as much as it would as a result of exceeding the image -focus range. The problem occurs as a result of a drop in insulation resistance in some camera cables when
heated. The critical conductors are the two leads that go to one side
and the arm of the photocathode -voltage potentiometer, variously
termed IMAGE -FOCUS or PC FOCUS on the camera control unit. For
instance, these leads are connected to pins 15 and 21 in the RCA TKI1 camera chain. Each of these conductors is one of a group of seven
conductors arranged in a circle of six with the seventh in the center.
If you have this problem, remove the protective bell cover at each end
of the camera cable, and check to make sure that a center conductor
is used for one pin in one group and for the other pin in the other
group. If it is not, simply interchange the conductor with whatever pin
is connected to the center conductor. This will increase the conductor to- ground resistance.
A11 -8. These can be outlined as follows:
Spots
(A) Spots that defocus when the photocathode focus ( image focus)
is varied can be caused by dirt on the faceplate. Open the lens iris. If
the spots grow in size and contrast decreases, clean the tube faceplate
and /or other optical components in the system. If no change in size occurs, the photocathode itself is blemished. Adjust for the best point to
minimize the effect.
(B) Spots that remain unchanged when the pc focus is varied, but
defocus when the beam focus (orth focus) is varied result from defects
on the target or field mesh. You might be able to return the tube to the
factory for correction; check with the manufacturer.
(C) A large white spot near the center of the raster, if it is observed
with the lens capped and does not change with adjustment of the
focus control, is an ion spot. You must return the tube to the factory
for reprocessing. This sometimes occurs in tubes that have not been
operated periodically. Be sure to rotate I.O.'s, including all spare tubes,
at least once a month.
Portholing (dark corners in picture)
Usually, portholing is a result of improper adjustment. Open the
lens to operate over the knee of the transfer curve. Then adjust the
460
TELEVISION BROADCASTING CAMERA CHAINS
voltage on grid 5 (decelerator grid) for best beam landing (best corner brightness) . You might need to align the beam on a different loop
(mode) of focus and readjust the grid-5 voltage. You also may need
to readjust the grid -3 voltage (multiplier focus) slightly to provide
maximum uniform signal output. This adjustment normally should be
made with the lens capped.
If the above procedure fails to affect portholing, change the lens and
note whether the picture improves. If it does, the lens may be a vignetting lens (highly improbable). Also, the yoke might be magnetized.
Demagnetization methods are described in Chapter 4.
Noise in Picture
The first step when the picture is noisy is to cut off the beam and
check for amplifier noise with the control -unit gain control at reference
operating level. If no noise is apparent, turning the beam up will bring
in the noise; this indicates a noisy I.O. If there is some noise present
with the beam off, sometimes you can handle this situation temporarily
by increasing the orth -gain (dynode voltage) control (when provided
on the camera) to override the amplifier noise. This is only a temporary
solution for use when time does not permit servicing the amplifier.
If the noise definitely is coming from the I.O., check the targetvoltage adjustment, and see if adjustment of the grid -3 voltage (multiplier focus) for maximum signal output will minimize the noise. Be
sure you are using a sufficiently high lighting level to allow operation
over the knee at normal lens 1/ stops.
Coarse Mesh Pattern in Picture (In Field -Mesh Tubes)
A coarse mesh pattern in the picture is caused by alignment of the
beam on the wrong loop (mode) of focus. If you are unable to obtain
the proper voltage mode of operation, try using a different focus -coil
current, within 4 or 5 mA of the normal current. Remember that a
field -mesh I.O. will "align" properly on only one mode of operation
( operable grid -4 volts from near center to minimum) .
Soft Picture (Poor Resolution)
The only way you can be sure whether a soft picture is caused by the
I.O. or by the associated amplifiers (unless lengthy video sweep procedures are used) is to interchange the tube with one that gives a good
picture in another camera. Always be sure that the lens and faceplate
are thoroughly clean. Also, the blower motor and filters must be in
good condition so that the tube is not too hot. To be sure there are no
magnetic -field problems, turn off the blower motor and other adjacent
electrical machinery temporarily and note the effect; if the resolution
changes with the location of the camera, this is almost certainly a clue.
Never forget to double -check the settings of such controls as filamentvoltage switches (usually in the camera) and cable -length switches
(usually in the camera control or processing unit) . Whenever the cable
length is changed, these switches should be reset properly. Otherwise,
the resolution can suffer, and other problems can be encouraged.
Index
Carrier
balance, automatic, 370 -374
unbalance, 67 -68
Cascode amplifier, 198
Cathode interface, 208 -209
Center of gravity adjustment, 92
Character, 80
Chips, 73
Chopping, 370
Chroma gains, 69
Chrominance signal, 381
Clamp
back -porch, timing circuitry for, 241 -244
-onblack mode, 310, 312 -313
white mode, 310, 313 -314
pulse -transformer type, 240 -241
slow- acting, 329 -332
tube -type, 238 -239
Clamping, 258 -259
circuitry, 238 -244
keyed, 237 -238
stage, 317
Cleaning, 442 -443
Clocked flip -flops, 76, 77
Color(s )
angles for, 60
balance between cameras, 291-292
balancing, 419 -420
bar pattern, 69
cameras, types of, 36-37
complementary, 39
displacement, 62
"distortions" in, 57 -72
killer, 399 -401
lighting for, 21
NTSC, basic problem in, 59
phase, 403 -405
purity coil, 413 -415
signal(s), 53
equation for, 38
polarities of, 46
standards, 36 -54
sync, 403 -405
timing system, 350 -354
systems, integration with monochrome, 274
temperature, 21
triangle, 43, 60
Colorplexer, 350
Colpitts oscillator, 341
Command system, digital, 83-84
Complement, 74
Composite signal, 41
Conditions, 73
Connectors, camera, 129-133
Contours out of green, 249
Control( s )
camera, 311
circuitry, 281 -284
operating, 325 -327
camera -position, 314 -315
Convergence, 420 -423
circuits, 427
controls, 427 -428
A
A/D converters, 88
Air conditioning, 23 -24
Alignment procedure, video, 212-213
AND gate, 74
APC circuits, 405
Aperture
correction
amplitude -limited, 247 -249
circuitry, 244-251
vertical, 249-251
corrector, 320
effect 143, 145
APL, effect on waveform, 239 -240
Aspect ratio, 234
ATC multivibrator, 278
Automatic phase control, 397-399
circuits, 405
AWL, 283
B
Ball chart, 442
Bandpass amplifier, 399-401
Bandwidth, 222
formula for, 223
high- frequency, 223-227
I and Q, 38
limiting, 364 -365
low- frequency, 231 -233
meaning of, 222-233
Bar(s)
100- percent, 55
patterns, color, 377
75- percent, 55 -57
signal, color, 55 -57
Batten, 25
Beam
alignment, 323
current, 324
Binary numbering system, 79
Bit, 80
Black level, reference, 54
Blanking
camera, 273
insertion, 258 -260
level, 54
Blocking oscillator, 292 -294
Boxcar circuit, 241, 275 -278
Breakdown
region, 101
voltage, 101
Brightness
control, pulse type, 430
derived, 279-280
Build -out resistor, 202, 203
Burst
amplifier, 401 -403
amplitude, 365
flag, 350
key generators
adjustment of, 354 -358
transistor, 353 -354
tube, 352 -353
phase, 365
error, 59 -61
setting, 377-379
separator, 401 -403
Counter(s)
multivibrator, 345
solid- state, 345 -348
Crane, 90
C
Cable corrector, 320
Calibration signal, 328
Camera
cable, 123 -131
chain, monochrome, RCA TK60, 308 -315
color
digitally controlled, 81 -88
four-Plumbicon, 315 -321
Marconi Mark VII, 315 -321
control
RCA TK -44A, 321 -327
unit, 317
operation, modes of, 312 -314
dolly, 96
Crawl, power -line, 339 -340
Cross talk, 61
Cutoff frequency, 224
D
Damper tube, 296
Damping resistance, 216-217
Dark current, 173, 174, 177
Dc restoration, 236 -237
Decoder, I-and -Q, 394 -413
Decouplers, 111, 305
461
462
INDEX
Deflection
circuitry
camera, 292 -298
vertical, 297
coil driving circuitry, 295 -297
problems, 448 -447
reversal controls, 315
Delay, 48 -49, 394-395
cable, 271
compensation, 364 -365
line
tapped, 352 -353
video, 426
phase vs time, 392
pulse, 276
Gamma, 177, 251
circuits, 290
for film cameras, checking, 254 -256
for live cameras, checking, 257 -258
correction, 39, 57 -59, 251 -258
corrector, 321
mismatch, 287
Gates, 74
Gaussian rolloff, 226
Geometric distortion, 158 -159
Gray scale
reproduction, 235-236
steps on, 236
Grounding, 114 -118
Demodulator(s )
H
diode, 426 -427
I and Q, outputs of, 63
synchronous, 405 -409
X and Z, 432 -434
Detail contrast, measuring, 245 -247
Detector(s)
ideal, 48
NAM gated, 280-281
Dichroic mirrors, 183, 184
Differential
gain, 69 -70, 383 -384, 386 -387
phase, 59, 70 -72, 383 -384, 387 -388
Digital concepts, 72 -81
Dimmer
autotransformer, 31
SCR, 32 -34
Diode, testing of, 121 -123
Discharge tube, 293
Display, polar, 383
Dolly, 90
Dot
crawl, 338, 339, 350
structure in NTSC color, 337 -339
Driver tube, 295
Driving signals, camera, 271
Dummy loads, power -supply, 112 -113
Duration response, 231
Dynode, 138
E
Edge flare, 440
Encoder, 358 -362
adjustments, 385
Encoding process, 358 -379
Envelope delay, tolerances for, 48
Equalization, cable, 251
Equalizer, high-light, 247
Erratic
black level, 446
video level, 445
78
Exposure, proper, 325
EXCLUSIVE OR,
Feedback
cycle, 293
pair, 201 -202, 264 -265, 268
FET, 198, 201
Field- neutralizing coil, 416
Flip -flop, 76
Flyback time, 296
Focus
current, 441
mode in I.O., 324
voltage, 441
Foot -candle, 11
Forbidden combinations, 76
Frequency
measuring service, primary, 349-350
response, transmitter, 47 -48
standard, 349-350
G
Gain(s)
control, remote, 266, 287
standardized, 284
H, 49
Halo, 138
Heat filters, 288
High
peaker(s), 145, 212,213
adjustment of, 218 -221
voltage
adjustment, 418 -419
supply, 106 -107
Horizontal -drive adjustment, 418 -419
Hue, 381, 382
control, 60
signal, 37
vector, 42
IEEE
graticule, 384 -385
units, 55
Illuminant C, 60
Illumination, 175 -177
scene, 327
Image orthicon(s), 134 -152
characteristics, table of, 140
comparison of, 142
cooling system, 161 -162
field -mesh, 159-161, 323, 324
image section, 136
multiplier section, 138-139
non-field -mesh, 323, 324
problems in, 141 -143
resolution, 143 -145
scanning section, 136-138
setup
controls, 145 -152
techniques, 147 -152
simulator, 213
Impedance, sweep-generator, 210 -212
Integrated circuit, 73
Integrators, 304
Intercom, 310
Interconnecting facilities, 123 -133
Interlace, 235, 338
Interphone, camera, 333 -335
Inverting amplifier, 74
Iris control, 308
servo, 190-192
J
JK flip -flop, 78
K
Keyer circuit, 269-270, 306
k factor, 225
Lamp, quartz, 12
Lens shade, 327
Level
calibration, 284 -292
camera -head output, 437 -438
control pulses, 279 -284
Life, lamp, 22
463
INDEX
Light
amount of, 11 -23
output, 15
Lighting, 312
control, 26 -34
multiplying factors for, 13
power requirements for, 24 -25
Limiter, 317
Linearity, 441 -442
controls, sweep, 296 -297
correction, feedback, 297-298
Lines, active, 339
Lobe, 229
Low -frequency filter, 195 -196
Lubrication, 443
Luminaire, high -efficiency, 17
Luminance, 381
M
Maintenance, 260 -270
power -supply, 112 -123
preamplifier, video, 205-221
preventive, general, 442 -445
video amplifiers, 261 -267
yoke, 157 -159
Mask, 415
Matching, color monitor used in, 434 -436
Matrix, 362 -364, 434
coupling to, 409
errors in, 388-390
function, 411 -413
Matrixed signals, 409 -413
Memory capacitor, 332
Modulation
luminance, 54
process, 365 -379
Modulator, doubly balanced, 67, 68
Monitor(s ), color
adjustment of, 418 -423
RCA TM -21, 423 -432
Multipath -reflection errors, 59
Multiplier, frequency, 343
Multivibrator, 294-295
bistable, 348
cathode coupled, 352
N
gate, 74
Negative
logic, 73
picture amplifier, 320
Nonadditive mixing, 279-281
Noninverting amplifier, 73
Nonlinearity, amplitude, 70
NOR gate, 74
Not allowed combinations, 76
Notch filter, 49
NTSC, 36
Nuvistor, 308
NAND
o
Observations, R, G, B, and Y, 388 -392
One, 73
Operating procedure, camera, 312 -315
Optics, camera, 182 -190
Orbiter switch, 314 -315
OR gate, 74
Oscillator, locked, 340 -343
Oscilloscope, 350
calibration, 206 -209
Outdoor scenes, 326
Overload protection, 105 -106
Overpeaking, 439
Overscan switch, 315
Overshoots, transmitter, 46
P
Pads, video, 220 -221
Pan and tilt cradle, 90 -93
Panning head, 90
operation, 92 -93
Parabolic waveform, 305 -306
Peaking circuits
adjustable, 217
vacuum -tube, 193 -196
Pedestal, 90
dollies, 93-95
description of, 93 -95
operation of, 95
Phantom oR, 78
Phase
control, adjustment of, 218 -221
-frequency characteristic, 203 -204
sensitivity, 59
splitters, 304
Photoconductive tubes, 325
Pickup -tube protection, 299 -302
Picture
element, 227 -228
tube, color, 413 -415
circuitry, 413 -418
controls associated with, 418 -418
Pigtails, 25
Pilot white pulse, 281 -282
Pincushioning, 158
Plugs 443 -445
Plum¡,icon, 180-182
Polarities, I and Q, 62
Pole compensation, 289
Portholing, 440
Positive logic, 73
Power
distribution, 25 -26
camera chain, 107 -111
requirements for lighting, 24 -25
supplies
camera -chain, 99 -107
tests for, 119 -120
Preamplifiers
vacuum -tube, 196 -198
video, solid- state, 198 -205
Preregulation, 103
Primary colors, 36
Prism optics, 185 -186
Probes, oscilloscope, 205 -206
Processing amplifier, 260
Product demodulation, 408
Programming current, 102
Prompting equipment, 97-99
Pulse (s )
distribution amplifiers, 274
duration, boxcar, 276
reference, 284
Purity adjustment, screen, 419
Q
Q signal, 37
Quadrature
adjustments, 376
component, 64
distortion, 61 -66
R
Radiation outside channel, 49
Receivers, "narrow- band," 63-64
Receptacles, 443 -445
Reference
diode, 101
voltage generators, 110-111
Registration controls, color- camera, 325
Regulator
high -voltage, 416
vacuum -tube, 99-100
Reimage lens, 183
Remote gain control, 317, 320
Resolution, 233 -235, 324 -325
chart, television-camera, 233-236
film- chain, 437 -439
horizontal, 222, 225
limiting, 235
vertical, 223
Response-switch circuitry, 333
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement