sytek :: Cooper Broadband Network Technology 1988

sytek :: Cooper Broadband Network Technology 1988
r
Broadband Network
Technology
An Overview for the
Data and Telecommunications Industries
By Edward Cooper
Sytek, Inc.
Mountain View, California
Sytek
Sytek Press
1225 Charleston Road
Mountain View, California 94043
Prentice-Hall, Inc.
Englewood Cliffs, New Jersey 07632
International Standard Book
Number: ISBN 0-13-083379-7
Editor: Christopher L. Poda
Book Design:
.
Richard Klein Design, Inc.
EditoriallProduction Supervision:
Diana Drew
Cover Illustration: Robert Pleasure
Interior Illustrations,: Gemini Graphics
Text Typesetting: George Graphics
Manufacturing Buyer: Ed a'Dougherty
Copyright © 1986, 1984 by Sytek,
Incorporated
This 1986 edition published by
Prentice-Hall, Inc.
A Division of Simon & Schuster
Englewood Cliffs, New Jersey 07632
All rights reserved. No part of this
publication may be reproduced, stored in al
retrieval system, or transmitted, in any
form or by any means, electronic,
mechanical, photocopying, recording or
otherwise, without the prior written
permission of the publisher.
Printed in the United States of
America
10 9 8 7 6 5 4 3 2 1
ISBN 0-13-083379-7
025
Prentice-Hall International (UK) Limited,
London
Prentice-Hall of Australia Pty. Limited,
Sydney."
Prentice-Hall Canada Inc., Toronto
Prentice-Hall Hi~panoamericana, S.A.~
Mexico
"
'
Prentice-Hall of India Private Limited;
New Delhi
Prentice-Hall of Japan, Inc., Tokyo
Prentice-Hall of Southeast Asia Pte.jUd.,
Singapore,.
,l
Editora Prentice-Hall do Brasil, Ucla.,
Rio de Janeiro
.
Whitehall Books Limited,
Wellington, New Zealand
Preface
During the 1970s, advances in microelectronics led to the development of small,
powerful electronic workstations arti;! personal computers. As hardware prices dropped
and versatile applications software became available, many individuals and
organizations began buying and using these systems. At the same time, corporate
electronic data processing departments made remote access to their centralized
computers possible by distributing terminals and other input/output devices
throughout the company's offices.
The problem of connecting many terminals, printers, and other remote peripheral
devices to a large, centralized computer was solved by using point-to-point multiconductor cables, modems with telephone lines, and even dedicated coaxial cables.
These techniques led to wiring mazes and maintenance problems in even mediumsized plants and office buildings. A bettq:r solution to the interconnection problem was
developed by combining the technologies of broadband cable communications and
programmable microprocessor and interface circuits. This same approach was also
used to connect separate personal computers and workstations together into local area
networks (LANs). Broadband networking allows many different devices to be
connected throughout a large area with a single cable.
Broadband radio frequency (RF) coaxial cable systems have been in service since
the 1950s. The designs, components, and tools for such systems have been developed
and refined by several different companies. Mass production has lowered the cost of
broadband cable components greatly. Similarly, the cost of intelligent interface devices
has dropped because of continuing advances in the semiconductor industry.
Broadband local area networks can now be used to link sophisticated data
processing equipment, and can provide a reliable communication backbone for
transmitting audio and video signals throughout a facility. Data transmission became
practical with the development of the RF modem, which translates digital data into RF
signals and vice versa. A packet communication unit that combines an RF modem with
a standard RS-232 interface port permits data transfer among digital devices from
many different vendors.
This book will help managers, engineers, and operators involved with data
communications to understand the principles and techniques of broadband networks.
It will provide readers with the knowledge and vocabulary necessary to communicate
effectively with broadband consultants and equipment vendors. The reader will gain
an understanding of the capabilities and limitations of the broadband approach, and of
the job of the broadband system designer. This should ease the task of working with
specialists to design the best network for your application.
This book is not intended to be a comprehensive reference on broadband
communications, or to be the definitive tutorial on broadband system design. RF
system design is learned through years of experience in designing networks, and not
by reading books.
Readers desiring to gain an overview of broadband communications should begin
with chapter one. Technical personnel might skim over the first chapter, read chapters
two and three, and concentrate on the detailed information in chapters four, five, and
six.
Chapter one provides general information on the development of broadband
systems and their applications, especially in local area networks.
Chapter two discusses several topics that are important in understanding
broadband networks.
Chapter three contains details regarding different cable system architectures that
are currently used, including a comparison of single and dual cable systems.
Chapter four describes characteristics of the active and passive components that
comprise broadband networks.
Chapter five covers system-level design problems that must be solved for each
network. Sample calculations show how various design factors interact with each
other.
Chapter six includes details on system alignment procedures, common problems
that can occur, and possible solutions to these problems.
Chapter seven lists sources of further information.
The appendices following the text contain specific references to terms and
symbols, CATV system frequencies, tools and test equipment, equipment
manufacturers, and books in the field. Details of RF calculations and system grounding
are also included.
For additional information consult the reference materials listed. To discuss your
specific requirements please call the Cable Design and Consulting Group of Sytek,
Inc., Mountain View, California (415/966-7347). In addition, we hope to continue to
provide informative overviews, such as this one, in other areas of broadband
communications.
We would like to thank the many people who contributed their time and
knowledge in making this book. In particular, Linda, Richard, and Cheralynn Cooper,
who had to put up with Ed's typing on a terminal for several weeks. Christopher's
parents, Louis and Josephine Poda, have provided support and encouragement which
is greatly appreciated.
Mike Pliner of Sytek provided the necessary support for the extensive effort that
was required. Several other individuals provided suggestions on the format and
technical content: Helmut Hess and Cecil Turner of the RF Systems Division of
General Instrument Corporation; Ken Howell of TRW; and Marcia Allen, Mike
Kalashian, Peter Filice, and Don Koller of Sytek, Incorporated. Ralph DeMent of DEC,
and all our professional contacts encouraged us to write this overview. Allen Day,
Susan Lindsay, and Michele Bisson were very helpful in the design and production
process.
Welcome to the world of broadband communications!
Edward B. Cooper, author
Christopher L. Poda, editor
ii
Contents
Contents
1.
Background and Applications
Introduction .'
Broadband Communications.
The CATV Connection.
Coaxial Cable .
Frequency Multiplexing.
Geographic Independence
The Business Community ..
Office of the Future ..
Work-at-Home . . . . . . . . . .
Small Businesses
The Industrial Community.
The Educational Community
Network Examples ..
2
2
2
3
4
4
. . . .
6
6
7
7
7
7
8
9
2.
Key Broadband Concepts .
Introduction
Network Topology
Physical Topology ..
The Headend
The Distribution Network
Logical Topology.
Network Implementation: Single and Dual Cable Systems.
Components
Design Issues . .
Achieving Proper Signal Levels
Signal Amplitude: the Decibel (dB)
Unity Gain Trunk Design
System Losses ..
Transparent System Design .
Summary.
12
12
12
12
13
14
14
15
18
19
19
20
22
22
24
25
3.
Single and Dual Cable Systems .
Introduction
Single Cable Systems .
Subsplit System.
Midsplit System.
28
28
28
28
29
iii
Contents
Highsplit System.
. ......... .
Converting From A One-Way To A Two-Way System ..
Dual Cable Systems . . . . . . . . . . . . . .
. ........ .
CATV Dual Trunk Systems
. . . . . . . . .. . ...... .
Connecting Networks Together.
Comparing Single and Dual Cable Systems .
System Bandwidth
Multiple Cable Systems.
Amplifier Capacity
Components.
Installation .
Maintenance ..
Interface to Other Networks
Redundancy.
Outlets
Future Developments.
Conclusions ....
4.
iv
Broadband Components .
Introduction
The Coaxial Cable ...
Types of Cables ..
Trunk Cables ...
Feeder Cables
Drop Cables
........................ .
Installation Considerations
Cable Attenuation.
Frequency Variation.
. ............. .
Temperature Variation ......... .
Amplifiers ................................. .
General Amplifier Characteristics.
Amplifier Gain Control.
Types of Amplifiers
Trunk Amplifiers ..
Bridging Amplifiers.
. ........ .
Line Extender Amplifiers ........................ .
Internal Distribution Amplifiers .
Amplifier Module Additions.
Attenuators .
Bidirectional Amplification
......... .
Equalizers
.............
Feeder Disconnect
Power Supplies ..... .
Grounding .. .
Passive Components
........................ .
Connectors and Hardware ............ .
Directional Couplers.
Multi-taps ....
Programmable Taps
31
31
31
33
33
36
36
36
36
37
37
38
38
38
39
39
40
42
42
42
43
43
43
43
44
45
45
46
46
47
47
48
48
49
50
50
50
50
50
50
51
51
53
53
54
54
56
58
Filters
Outlets
Terminators
5.
v
Factors Affecting System Design ..
Introduction
Initial Considerations.
System Structure.
System Frequency Considerations .
Bandwidth Requirements ......... .
Closed Circuit Television (CCTV)
Data Communications.
Special Services ............... .
Television Distribution
Bandwidth Allocation.
Physical Layout.
Signal Levels .
The Video Reference Level
Narrow Bandwidth Carrier Levels
Narrow Bandwidth Advantages
Noise Level
Noise Figure ..
System Carrier-to-Noise Ratio .............. .
Noise at a Splitter /Combiner
Amplifier Selection
Design and Performance Calculations.
Amplifier Gain ......... .
Amplifier Cascade. .
. ............... .
Amplifier Output Level .................. .
System Noise ...
Intermodulation Distortion
The System Level Graph
Using the System Level Graph.
Reliability and Redundancy
Broadband Component Quality
Periodic Maintenance.
Equipment Replacement and Repair ...... .
Redundant Trunks and Components ....... .
Status Monitoring Systems
Technical Control Systems.
Headend Design .
System Diagrams.
Standard Headend .
Large CATV Multichannel Headends ..
Composite Triple Beat Distortion ..
Harmonically Related Carriers.
Interval Related Carrier.
59
60
60
62
62
62
64
64
65
65
65
67
67
68
68
70
70
72
74
75
75
76
77
79
79
80
80
81
81
83
84
85
86
87
87
87
87
88
88
89
89
89
94
94
94
95
Contents
Drawing Standards.
System Specifications.
Summary.
95
96
98
6.
System Alignment.
Introduction ..
. ...... .
The Need for Alignment ..
Test Equipment Required
Documentation .....
Coaxial Cable Certification ..
Alignment Methods ...
Flat Amplifier Output.
Flat Amplifier Input.
Flat Midspan ... .
Summary ... .
Alignment of Single Cable Two-Way Systems
Non-amplified System .. .
Forward Path Alignment .... .
Return Path Alignment .
Alignment of Dual Cable Systems .
Inbound and Outbound Cables.
Potential Problems and Solutions.
Radiation and Signal Ingress .
Checking an Outlet ..
Monitoring System Performance
Summary.
100
100
100
102
103
103
104
104
105
106
106
106
107
107
108
110
110
111
112
113
113
113
7.
For Further Reading
116
Appendix A: Definition of Terms
118
Appendix B: Abbreviations and Acronyms
126
Appendix C: Broadband Symbols
127
Appendix D: Frequency Allocations
128
Appendix E: RF Calculations
The dBm and dBm V ............ .
Noise
Carrier-to-Noise Ratio
Broadband Design Equations
130
Appendix F: Test Equipment ..
133
Appendix G: Tools for Installation and Maintenance
Trunk Cable Tools
Drop Cable Tools
Miscellaneous
136
vi
130
130
131
132
136
137
137
Appendix H: Broadband Equipment Suppliers
Full-Line Manufacturers
Support Manufacturers and Distributors
Satellite Equipment Manufacturers
138
138
139
140
Appendix I: System Grounding
Introduction
Broadband System Grounding
Grounding Between Sites
Above Ground Exposed Trunks
Ground Loops
Conclusion
141
141
141
141
142
142
142
Appendix
J:
RF Connector Details
144
Appendix K: Bibliography
148
Appendix L: dBm V-to-Voltage Conversion Chart
150
Index
152
Figures
1-1.
1-2.
2-1.
2-2.
2-3.
2-4.
2-5.
3-1.
3-2.
3-3.
3-4.
4-1.
4-2.
4-3.
4-4.
4-5.
4-6.
4-7.
4-8.
4-9.
4-10.
4-11.
4-12.
5-1.
5-2.
5-3.
5-4.
vii
Broadband Coaxial Network in a Dispersed Facility
Multiple Services Using a Broadband Network
Inverted Tree Network Topology
Bidirectional Communication: Frequency Translation
Bidirectional Communication: Full Duplex Transmission Paths
Unity Gain in a System
Unity Gain of One Stage
Typical Two-Way Cable Trunk Amplifiers
Dual Cable System Amplifier
Interconnecting LANs with Industrial Trunking
Combined Single and Dual Cable Systems
Terminology for Coaxial Trunk Cable
0.500-inch Cable Loss
Cable Attenuation Versus Frequency for Various Sizes of Coaxial Cable
Trunk and Bridging Amplifier
Power Supply Configuration
Insertion Loss Values
Connector Types
Directional Aspects in Cable Systems
The Multi-Tap
The Multi-Tap and Distribution Legs
Amplitude/Frequency Response of Typical Filters
Terminator Symbols
Frequency Allocation Charts
Signal Levels from Trunk to Outlet
A Typical Television Channel
A Typical Data Channel
5
6
13
15
17
22
23
30
32
33
34
42
45
46
49
52
53
55
56
57
58
59
60
69
71
72
73
Contents
5-5.
5-6.
5-7.
5-8.
5-9.
5-10.
6-1.
6-2.
6-3.
6-4.
C-l.
D-1.
J-1.
J-2.
J-3.
Television and Data Carriers ...... .
Noise at a Splitter /Combiner ........ .
A Typical System Level Graph
Headend Configuration Showing Signal Levels
Typical Cable Distribution Scheme ....... .
Detailed Headend Configuration
Trunk Amplifier Configurations ..
Amplifier Alignment Techniques
Forward Path Alignment ...
Signal Levels at a Junction
Broadband Symbols
Frequency Allocation Chart
Terminology for Connector Parts ......................... .
Connector /Equipment Interface
F Connector Types and Parts
2-1.
Bidirectional Single-Cable Systems
74
78
85
90
91
92
101
105
107
109
127
128
144
146
147
Tables
2-2. dBm V /Microvolt Conversion Chart
5-1.
5-2.
5-3.
5-4.
D-l.
F-1.
viii
...................... .
Effect of Cascading Amplifiers on Noise Figure
Picture Quality for C/N Values ......... .
Sample System Design Characteristics .................... .
............ .
Broadband System Specifications
Headend Channel Assignment Reference Table
Test Equipment for RF Network Certification
16
21
76
77
82
97
129
133
1
Background and Applications
In trod uction
Over the past decade, electronic data communication requirements have exceeded the
capacities and capabilities of existing telephone, twisted pair, and similar
communication media. Continuing advances in data processing, new developments in
interactive equipment, and expanding network implementations impose further
strains on conventional wiring schemes. Techniques using standard twisted pairs are
cumbersome when the system requires:
~
Frequent expansion or reconfiguration
~
Complex central intelligence
~
Flexible equipment placement and data flow
In addition, these techniques cannot distribute real time, high quality video signals for
video conferencing and security applications. The office of the future and the wired city
concepts would require multiple, overlaying wiring schemes, if they were to be
implemented using conventional wiring.
The problems faced by many facility managers are further compounded by the
complex wiring necessary to provide each building with contemporary systems for
communication (voice, video, and data), emergency detection and warning, and
equipment control. As a result, ceilings bulge from the weight of many twisted wire
cables, telephone lines, and dedicated coaxial cables installed to provide current and
future services.
Fortunately, advances in cable television technology have led to the development
of a multi mode broadband signal distribution technique. This technique requires only
a single coaxial cable to carry many different signals simultaneously. A broadband
local area network (LAN) provides reliable, inexpensive, wideband communications
within a single building or throughout a dispersed site such as a campus or an
industrial park.
This chapter discusses broadband communications in general, including its origins
and applications.
Broadband Communications
Broadband is a generic term that refers to a type of wide bandwidth communication
network. Some of the main characteristics of a broadband network are listed below and
briefly described in the following sections.
2
1
~
Pi.. broadband network is a communications utility that can be used by many
different services; it provides the backbone for an integrated information system
throughout the area it serves.
~
Information signals modulate RF carrier signals that are transported by coaxial cable.
~
Many different types of information signals can share the cable's wide bandwidth
(several hundred megahertz) by frequency division multiplexing (FDM).
~
Each FOM channel can be subdivided further by using a variety of channel access
techniques, such as time-division multiplexing (TOM).
~
The network is geographically independent: devices can be attached to the network
anywhere and operate correctly.
~
The network is robust. The failure of an interface device attached to the network
cannot prevent the entire system from operating.
One early application of broadband communications was to provide television signals
to remote areas via coaxial cable. The components and network design principles
developed for these cable television systems have been directly applied to broadband
local area networks.
The CATV Connection
Community Antenna Television (CATV) systems began operation in 1949. The CATV
operator built an antenna site and a coaxial cable distribution network. The antenna
site received distant broadcast television signals. The cable distribution network
connected the antenna site (or a separate signal distribution facility) to each customer's
house. Television signals were received, processed, and transmitted along the cables to
viewers. Cable transmission was used in locations that could not receive broadcast
transmissions directly because of distance or because of interfering buildings or
terrain.
These early systems were limited in the services and Signal quality they provided.
Many system operators were satisfied if they could receive, amplify, and deliver a
signal to locations where it could not be received directly. Since most television
viewers were within reception range of at least one broadcast station, CATV systems
were confined to serving small, remote areas of the country.
In the past decade the CATV industry has grown greatly. Operators attracted new
customers by offering additional channels and new services that were not available
from broadcast television stations. In 1982, the cable television industry served 21
million subscribers. By the end of 1989 that figure is predicted to rise to 65 million
(more than thirty percent penetration of the available market). This potential customer
base encourages greater investment in new technologies. New research and
development will bring lower equipment costs and more services to both CATV and
broadband local area networks.
In this overview, the terms CATV and broadband are used synonymously. Both
systems use identical equipment and similar system designs. In general, CATV systems
extend over a wide area (often city-wide), and are franchised operations. Broadband
networks extend over a limited area (one or more buildings), and are privately owned.
Furthermore, broadband networks are usually bidirectional. CATV systems are
3
1
Background and Applications
primarily unidirectional, although two-way transmission capability is becoming more
popular. Bidirectional CATV systems can supply customers with additional services,
such as data communications and interactive videotext.
Coaxial Cable
As a signal distribution medium, coaxial cables are reliable, economical, and can be
installed in existing conduits, underground cableways, and plenums. Coaxial cables
are available in several types, suitable to various environmental conditions. If damaged
or broken, the cable can be spliced or short sections can be replaced quickly by semiskilled personnel using simple tools.
Coaxial cables provide excellent shielding from electromagnetic interference (EMI)
and radio frequency interference (RFI). They are ideal for use in electrically noisy
industrial environments, where twisted pair wires are difficult to use effectively.
Fault isolation in a coaxial network is straightforward using readily available test
equipment. The isolation process can be enhanced by automated statistical recall
systems, status monitoring facilities, and programmable spectrum analyzers and signal
generators.
Frequency Multiplexing
Frequency multiplexing permits simultaneous use of the cable by many different
services. The large bandwidth available on the cable is divided into channels, usually
6 MHz wide. One channel can be used for transmitting video from a local camera,
while another channel can carry data between a computer and a terminal. These
applications can be totally independent yet use the same cable as their communication
medium.
The following list mentions some typical applications that can share a single
broadband local area network. Some require only one-way communication and others
require a two-way capability. Careful planning of equipment to be purchased and
frequency assignments to be made is necessary to ensure proper operation and no
spectrum conflicts. Figure 1-1 depicts a broadband network connecting all the
buildings inside a single facility. Figure 1-2 shows some of the services that could use
this network.
~
Data communications
~
Broadcast television signal distribution
~
Video conferencing
~
Security and safety monitoring
I> Fire alarms
I> Intrusion alarms
I> Closed circuit television (CCTV) surveillance cameras
I> Remote television camera control
I> BUilding and area access control
~
Paging
~
Energy management
4
1
SATELLITE
ANTENNA
r--.
BROADCAST
RECEIVING
ANTENNAS
ADMINISTRATIVE
OFFICES
COMPUTER
CENTER
~
BROADBAND
HEADEND
1
SECURITY
RESEARCH
LABORATORIES
MANUFACTURING
COMPLEX
Figure 1-1. Broadband Coaxial Network in a Dispersed Facility
5
TELEVISION
ORIGINATION
STUDIO
1
Background and Applications
OTHER VIDEO
~
CAMERA
ANTENNA
SENSORS
VTR
VCR
RFMODEM
FOR CONTROL,
BROADBAND COAXIAL COMMUNICATIONS NETWORK BUS
STUDIO TELEVISION
MONITOR RECEIVER
VIDEO CONFERENCING
...-,_ _ _ _ _ _ _ _--, TERMINALS
~
PUBLIC ADDRESS
SENSORS
M = MODULATOR
C = CHANNEL PROCESSOR
D = DEMODULATOR
Figure 1-2. Multiple Services Using a Broadband Network
Geographic Independence
Geographic independence means that a device performs as specified regardless of
where it resides in the network. When a building is fully wired with coaxial cable,
devices can be attached anywhere a standard outlet has been installed. This allows, for
example, a data communication user to disconnect an interface unit, move to a
different office, reconnect the device to the cable outlet, and resume working with no
need for additional wiring changes.
The cost of conventional wiring to provide an equivalent, multiservice capability
with the same flexibility would be much greater than that of a single coaxial cable. For
each of the services to be provided, the conventional scheme would require many
dedicated wires and cables throughout the building, some overlapping others. With
broadband, any type of device for any type of service could be placed wherever the
single coaxial cable was accessible. A complex color television camera, a simple smoke
detector, and many other devices connect to the same coaxial information utility line.
The Business Community
Broadband communications can be applied to problems in businesses of many
different types and sizes. The proliferation of computers and intelligent machines has
6
1
created a need for some method to convey information throughout offices, between
buildings, and to more distant locations quickly and reliably. Three different areas that
could prosper from broadband networking are the office of the future, the work-at-home
concept, and small businesses. Broadband technology provides the necessary resources
and interfaces to meet the local, regional, and national requirements of these and other
business communication applications.
Office of the Future
The aim of introducing advanced technology into the office workplace is to improve
productivity. Productivity in offices has not kept pace with that of factories, where
tremendous improvements have resulted from automation. Office automation is
spreading in large and small companies; it includes the use of word processors,
facsimile machines, intelligent copiers, computer terminals, and video conferencing
equipment.
Broadband networking can provide an integrated solution to the problem of how
to convey all the needed information to all the devices providing these services. Once
the network is implemented, additional services can be added such as electronic mail,
high-speed data transfer between workstations, integrated energy management, and
access control.
Work-at-Home
The work-at-home concept has been popularized over the past several years. However,
the high costs of implementing the required network and providing a home terminal
slowed its growth. Terminal costs are predicted to soon be in the $250 range, making
them cost effective for many users. Product announcements by major manufacturers
indicate the cost of network implementation is also dropping rapidly.
Small Businesses
Networking in the small business sector could proVide the impetus for the greatest
integration of CATV and broadband local area networks. Thousands of companies
throughout the United States could use individual broadband networks to supply their
local requirements. These local networks could interface to CATV industrial trunks
that supply communication links with other buildings (which could be several miles
away).
The costs would be lower and the capabilities greater for a system using a
broadband approach, compared to one using conventional wiring schemes for local
networking.
The Industrial Community
The growth of factory automation has been tremendous over the past several years.
This trend will continue as more intelligence is placed inside machines used directly
on the factory floor, and as robots become commonplace in both large and small shops.
Current facilities have many distinct processors distributed throughout the plant.
These are coupled into a loose network. Information passes between them as physical
7
1
Background and Applications
material, workpieces, and assemblies. Each machine must be programmed, and must
provide various status signals back to the operator. A factory floor status reporting
system can add further data on specific jobs and parts flow through the facility.
Connecting all these devices together with a broadband network can provide the
following advantages.
~
More efficient use of machinery
~
Improved scheduling of work and maintenance
~
Better reporting of results
~
Closer monitoring of machine performance
~
Tighter control of factory performance and costs
A single broadband network can link a factory and an office to the automated data
processing systems that support modern business operations. This integrated approach
can provide timely information for all the following applications.
~
Accounting
~
Personnel management
~
Time and attendance measurement
~
Energy management
~
Assembly line automation
~
Program and schedule. verification
In addition to providing a wide bandwidth medium that can be used by many
different services, broadband coaxial cable is much less susceptible to electrical noise
than twisted-pair wires. Both industrial and office environments contain radio
frequency interference that could degrade the performance of a network and terminal
equipment. It is best to assume the ambient levels of interference will become even
greater, and to design networks protected to work properly under those conditions.
The Educational Community
Educational institutions are finding the broadband approach a reliable, economical,
and efficient way to satisfy their communications requirements. These facilities have
several unique needs that can be met by broadband networks.
~
The wide geographical extent of a large campus can tax any communications
system attempting to serve the whole area. Broadband networks have been
successfully covering wide areas for years.
~
Cost constraints demand a network that is low cost, easy to maintain, and vendorindependent. Broadband satisfies all three of these criteria.
Vendor independence means that the network offers a standard interface to which
equipment from many different manufacturers can be attached.
8
1
1.
The user is not limited to purchasing equipment from only one or a
few vendors.
2.
Many different types of equipment can be connected together over a single
medium. The user can select terminals, workstations, and other devices from a
wide range of available units, and choose those that provide the best cost and
performance tradeoff for a specific application.
~
Real-time video must be provided throughout the campus for closed-circuit
television courses. Several one-way full bandwidth television channels can be
reserved on the cable for this purpose.
~
Other services could use the coaxial cable and minimize the installation and
maintenance costs for communication services throughout the campus.
~
Televised classes in the local community could be conveyed from the campus's
broadband network to the city's local CATV network. A two-way path between
these two networks could provide the campus with additional video signals from
broadcasters and from other educational facilities.
Institutions including Brown, Cornell, Carnegie-Mellon, and the University of
Waterloo have made extensive telecommunication studies, concluding that broadband
distribution is a feasible and cost-effective approach to providing communication
services. *
Network Examples
A fully activated broadband system (one with all return paths active) cannot be
duplicated in a large CATV system covering a major city without special consideration.
However, the industrial, educational, and business communities have seen few
limitations to broadband local area network applications. As an example of possible
broadband network size, a network in a West coast industrial plant covers an area
greater than 3.6 million square feet of floor space on each of its two floors. Only 11
amplifiers support the entire network, which provides an RF connection within 100
feet of any point in the system.
Another system in the Chicago area provides 4000 RF outlets over six buildings,
with no point farther than 20 feet from an outlet. This network also required only 11
amplifiers to support the six buildings, each of which has four floors of distribution.
A third example is a single system connecting six buildings. Within this complex,
each building has from 9 to 23 floors of high-density outlet distribution (an outlet for
each office). Only 21 amplifiers were required in this network. Also, a backup trunk
was installed that can be switched into operation either automatically or manually if
the primary circuit fails.
All three of these examples have operated successfully without a single amplifier
module failure for a total of 12 system-years of operating experience. A broadband
* W.S. Shipp and H.H. Webber, "Final Report of the Study Group on Telecommunications and Networks,"
Brown University, Oct., 1980.
9
1
Background and Applications
network can be counted on for continuous, daily operation. The high reliability of
broadband components makes redundant backup devices necessary only for the most
demanding networks.
If any of these networks had to expand to other facilities miles away, existing or
new industrial trunks could be used. Such trunks could be installed and maintained by
the local CATV company or by contractors working directly for the customer. In
addition, microwave links, satellite links, and other services can provide the necessary
resources for nationwide distribution of video or data.
10
2
Key Broadband Concepts
Introduction
This chapter introduces several important topics that will help in understanding
broadband communication systems. Several of these areas are covered in more depth
in the chapters indicated below.
~
Topology
~
Implementation (chapter three)
~
Components (chapter four)
~
Design issues (chapter five)
Network Topology
The arrangement of a broadband network is often described in two ways.
~
Its physical topology: where the components are located.
~
Its logical topology (architecture): how the components are connected to each other
by the network.
Before discussing these two topologies, the main elements of a network must be
defined. Every broadband network has two main elements, a headend and a
distribution network.
~
The headend comprises the equipment that collects RF signals from transmitting
devices attached to the network, and distributes RF signals to receiving devices
attached to the network. The term headend refers both to the location of this
equipment, and to all the equipment that performs these functions.
~
The distribution network comprises coaxial cables, amplifiers, and other signalcarrying components that provide signal paths between devices attached to the
network and the headend.
Physical Topology
The placement of the headend and the distribution network determines the physical
topology of the system. This topology can be portrayed by a map showing the location
of these network elements. This map resembles a similar plot of other utility
distribution systems (for example, telephone, electricity, water, or gas utilities), and can
be compared to the structure of a tree.
12
2
1.
It has a main trunk line that originates at the headend and traverses central areas.
2.
It has many branch lines that extend from the trunk to outlying areas.
3.
It has individual lines that run from each node on a branch to each connection
point.
The trunk and branches form the backbone network. The backbone can usually support
thousands of connections to user devices, with drop cables (connections to user outlets)
installed as demand dictates (figure 2-1). In a large network, the headend is often
located centrally so that the network can be extended easily in any direction.
The nodes of the network are determined by the location of the devices that use
network services. Branch cables can be laid to connect the nodes in many different
ways. Regardless of the layout chosen, each node connects to the cable system at only
one point, and is independent of all other nodes. This structure eliminates any chance
of multi-path distortion.
The physical layout of a broadband system can be designed to conform to any
building arrangement. The flexibility of the broadband technique exceeds that of any
other wiring scheme in use today.
HEADEND
o
o
BRANCHING POINTS
USER DEVICES
Figure 2-1. Inverted Tree Network Topology
The Headend
The headend is the origin of all RF signals transmitted to the entire network, and the
destination of all RF signals generated by devices connected to the network.
13
2
Key Broadband Concepts
~
In a unidirectional CATV system, the headend transmits all the signals the
network carries.
~
In a bidirectional broadband system, both the headend and user devices transmit
signals over the network.
Any device connected to the network may transmit signals over the network to
other devices. However, these signals do not go directly to the destination device.
Signals transmitted by devices connected to the network first go to the headend,
and are then retransmitted back to the network. Routing all transmitted signals to
the headend prevents interference among signals travelling on the system.
Equipment normally found in the headend of a large network that provides multiple
services are signal processors, modulators, demodulators, signal combiners, data
translation units, and power supplies. These devices are described in references
dealing with CATV systems (see bibliography).
The Distribution Network
The distribution network is the combination of components that transfer RF signals
between the headend and the attached user devices. These components include the
following.
~
Coaxial cables that carry signals between two points.
~
Splitters, directional couplers, and taps that direct signal flow along desired paths.
~
Filters that process signals depending on their frequency.
~
Outlets that connect devices to the network.
~
Amplifiers that increase signal strength.
Chapter four describes these components and their characteristics.
Logical Topology
From the local network point of view, the logical topology or architecture of the
broadband network could be a ring, star, or bus. This level of organization is distinct
from the physical topology, and depends on the LAN interface devices connected to
the broadband network. A single broadband network can support several different
types of local networks, each with a different logical topology.
In a bus network, all interface devices may have equal access to the network's
resources. This structure requires a method to regulate transmissions and to prevent
one device from monopolizing the network. Data communication networks have used
such methods, called channel access protocols, for many years.
In a token-passing ring network, each interface device has permission to transmit
when it receives a unique pattern of data over the network called the token. This
station then transfers its data to the network and passes the token along, when
finished, to the next station in the ring.
In a polled star network, a master controlling device grants permission to transmit
to each connected station, one at a time. All communications traffic passes from the
source device, to the central controller, and then to the destination device.
14
2
Other combinations of topologies and operating rules, or protocols, can be used on
a broadband network. These communication protocols are provided by the interface
devices attached to the network, and not by the backbone network itself. As long as
interface devices can successfully get signals on and off the cable, they can send any
kind of data using any kind of protocols.
Network Implementation: Single and Dual Cable Systems
Several different methods are available to provide two-way communication in a
broadband network, including single and dual cable systems. Single cable systems carry
two-way traffic on one coaxial cable. They provide this bidirectional capability by
dividing the cable's frequency spectrum into two main portions, one for traffic in each
direction. Dual cable systems use two separate coaxial cables, gne for traffic in each
direction.
When using a single cable to carry two-way traffic, the available signal spectrum
on that cable is split into three major portions.
~
~
~
The forward band carries signals from the headend to devices on the distribution
network.
The return band or reverse band carries signals from devices on the distribution
network to the headend.
The guard band carries no signals, and separates signals in the forward and return
bands.
A dual cable system needs no guard band, since separate cables carry the outbound
traffic (from the headend) and inbound traffic {to the headend}.
Figure 2-2 illustrates frequency translation in a single cable system. Devices
attached to the network transmit signals only in the return frequency band. All signals
inside the return band are received at the headend and converted up to signals with
higher frequencies by a device called a translator. These higher frequency signals,
which occupy the forward band, are then transmitted to all receivers on the network.
300 OR 400 MHz
FORWARD CHANNEL
HEADEND EQUIPMENT
INCLUDING
DATA
TRANSLATOR
5 MHz
Figure 2-2. Bidirectional Communication: Frequency Translation
15
2
Key Broadband Concepts
To visualize this concept, compare the broadband cable to a multilane highway. A
highway has lanes going in both directions. The center divider (guard band)
minimizes interference between the two directions and provides a boundary. A single
cable supports two-way traffic exactly like a highway. A guard band of several
megahertz separates the forward and return bands.
A highway has several lanes in each direction. A broadband cable has several
channels for traffic in each direction.
Many different types of vehicles can use the general purpose highway lanes,
including cars, trucks, motorcycles, and busses. Many different types of services can
use the general purpose broadband channels on the network, including low- and highspeed data, voice, and video.
Some highway lanes can be reserved for specific uses only, such as busses or highoccupancy vehicles. Some channels on the cable system can be reserved for specific
uses only. For example, one data communication channel might be reserved for users
of a specific time-shared computer that can be accessed only via that channel.
The frequency spectrum of a single cable system can be divided in one of three
ways. Each provides a different forward and return bandwidth (see table 2-1).
~
The subsplit format is found on many older CATV systems. It has the least return
path bandwidth of the three. It is usually the easiest format with which to upgrade
an existing one-way system to two-way operation.
~
The midsplit format provides more return path bandwidth than the subsplit; it is
often used in contemporary broadband local area networks.
~
The highsplit format is a recent innovation and will become more popular as more
components become available for it. It provides the greatest return bandwidth of
these three formats.
All three of these frequency divisions are further described in chapter three.
Table 2-1.
Bidirectional Single-Cable Systems
Format
Return Frequency Band
Forward Frequency Band
Subsplit
5-30 MHz
54-400 MHz
Midsplit
5-116 MHz
168-400 MHz
Highsplit
5-174 MHz
232-400 MHz
Figure 2-3 shows a simple single-cable network layout including the headend
translator and the distribution network. The arrows show the direction of signal flow
from the transmitting unit, through the headend translator, to the intended receiving
unit. Full duplex operation is possible because different frequencies are used for
transmitting and receiving.
16
2
r
TRANSLATOR
I
I
~-+--,
I
RETURN PATH
I
•
t
I
I
,-+-....1
MODEM
TRANSMIT MODE __...._-'
TERMINAL UNIT
TYPICAL NODE
•I FORWARD PATH
•
I
•
i•
~
•
L·_·_·~·_·_·I
•
t
MODEM
RECEIVE MODE
TERMINAL UNIT
Figure 2-3. Bidirectional Communication: Full Duplex Transmission Paths
17
2
Key Broadband Concepts
Components
The components used in a broadband network can be divided into two categories,
active and passive. Active components require input power to operate properly; passive
components do not. Chapter four discusses components in more detail.
The most ubiquitous component in a broadband network is coaxial cable. It is the
transmission medium for all signals on the network. A coaxial cable is constructed of a
solid center conductor surrounded by a uniform thickness of insulating or dielectric
material. These are thoroughly covered by a second layer of conducting material (the
shield) and the final exterior insulation. Both conductors have a common axis, hence
the name coaxial.
Conductor resistance and dielectric conductance are distributed uniformly along
the length of the cable, and vary with frequency and temperature. Because of these
variations, the frequency response of coaxial cable varies with the following
parameters.
~
Length and diameter of the cable
~
Frequency of the signal
~
Ambient temperature
These variations must be considered when designing and installing a system.
Some of the significant characteristics and capabilities of coaxial cable include the
following.
~
High bandwidth.
Coaxial cables can convey wider bandwidth signals than twisted pair wiring.
Cable bandwidth exceeds that of active and passive RF network components.
~
Shielding.
When properly grounded, the cable's shield (outer conductor) prevents ambient
electrical noise from interfering with signals travelling on the center conductor.
Thus, coaxial cable is a good transmission medium for use in electrically noisy
environments (including offices, laboratories, and factories).
~
Easy connectivity.
A cable can be quickly and easily cut and spliced to repair a break or to attach new
outlets, devices, or cable paths.
~
Characteristic impedance.
Most cables used in broadband systems have a 75-ohm characteristic impedance.
~
Diameter.
Larger diameter cables have less loss than smaller diameter cables and are
therefore used for main trunks and for long cable runs.
Smaller cables have more loss, but are more flexible and easier to manipulate and
install. Branch cables and drop cables (connecting network devices to the system)
are usually smaller than trunk cables.
18
2
~
Power.
Cables with seamless aluminum shielding can safely transport ac or dc power to
amplifiers on the network. By sending power over the cable, each amplifier does
not need a separate connection to a power source.
~
Wide selection.
Coaxial cables have been made for several decades by many manufacturers. The
technology is well known and the product is reliable. Cables with appropriate
characteristics are available for rough environments and other special applications.
Cable attenuation is commonly specified in decibels per 100 feet at the highest
operating frequency (300 or 400 MHz for most systems). A typical cable specification
provided by the manufacturer could be stated in the following manner.
1.63 dB of loss per 100 feet at 300 MHz measured at 68 degrees Fahrenheit.
This represents the specification of one type of 0.412-inch diameter seamless
aluminum coaxial cable.
To calculate the signal loss of a 1000-foot length of this cable, simply multiply 1.63
by 10 which gives 16.3 dB. By using the attenuation factors for each frequency band,
the loss of any given length of cable for those bands can be calculated.
Following chapters contain more information on coaxial cables. Chapter four
provides further details on their characteristics. Chapter five discusses cable selection.
Design Issues
This section introduces three concepts related to signal levels that are important to the
design of broadband networks.
~
~
Using decibels to indicate signal levels simplifies all network signal level
calculations.
The unity gain criterion eases system design and alignment.
~
Transparent system design ensures geographical independence of the network.
This discussion points out some of the advantages of properly designed broadband
networks. Further coverage of backbone network design is provided in chapter five.
Achieving Proper Signal Levels
Once the basic topology is determined, the distribution network can be designed to
supply proper signal levels to each connection. The design engineer selects the
components that meet the system's physical and electrical specifications, and calculates
the signal loss along each path. Where computations indicate signal levels will be too
low, amplifiers cain be inserted to increase them. The calculations are then repeated for
19
2
Key Broadband Concepts
every path that each new amplifier affects, and other components might have to be
changed to achieve specified signal levels. This process is repeated until all signal
levels are within specified ranges.
As an analogy to RF signal distribution, consider a simple water delivery system.
The purpose is to provide a specified water pressure to three destinations, which might
be done in the following manner.
1.
Build a large main delivery line from the reservoir to each location.
2.
Install valves at each location.
3.
Adjust each valve to deliver the required water pressure. If the pressure is too low,
either add pumps or increase the pressure in the main delivery line.
The basic principles of RF design are similar. The main RF trunk cable corresponds to
the main delivery line. Passive taps and couplers act like valves that determine the
signal level at each outlet. Amplifiers increase signal strength where it is too low.
Signal Amplitude: the Decibel (dB)
Calculating signal levels throughout an entire system can be time-consuming,
especially when several channels and amplifiers are involved. This task is made easier
by expressing signal levels in logarithmic units.
The decibel is a unit that expresses the ratio of two levels of power. It can also be
used to express the ratio of two voltage or current values, if they are measured at points
of similar impedance. For example, many network components have an input and an
output connection. The ratio of the signal levels at these two points can be expressed in
decibels, such as an attenuator with 3 dB of loss, or an amplifier with 20 dB of gain,
from input to output.
Number of dB = 1000g (PdP2 )
where
PI, P 2 are power levels
-----------------
Vv V2 are voltage levels
log is the base ten logarithm
A standard unit used in the CATV industry to express signal amplitude is the decibel
referred to one millivolt (dBm V).
Number of dBmV = 20l0g (VI/lmV)
where
VI = the measured voltage level
o dBm V =
20
I mV
=
1000 J.l V across a 75-ohm load
2
Using absolute voltage levels instead of decibels to calculate signal levels requires
calculations that become more complex as more components and channels are added to
a system. Using dBm V to calClIlate signal levels allows easier manipulation of those
values. The gain or loss, in decibels, of a component is added to or subtracted from its
input signal level to obtain its output signal level. Also, fractional values can be
avoided, since any number between zero and one is represented by a negative number
of decibels. For example,
~
A 40-dB amplifier increases its input signal voltage level by 100 times;
~
A 6-dB coupler decreases its input signal level by one-half;
~
A video signal level of 10 dBmV for a specific channel at any given outlet is
3,200
J.Lv.
Because decibels are easy to use, all relevant equipment specifications and signal
requirements are expressed in dBm V or dB. The difference between two dBm V values
is expressed in dB. For example, a typical carrier-to-noise ratio is 43 dB and can be
obtained by subtracting the measured noise floor (in dBm V) from the input signal level
of an amplifier (in dBm V). Both the input level and the noise floor are expressed in
dBm V, but the mathematical result is expressed in dB. Thus, dB is a ratio while dBm V
is an expression of signal amplitude.
Table 2-2 provides a short conversion chart for translating between dBm V and
voltage. Appendix L contains a more detailed table. Note that all levels below 1000J.LV
(0 dBm V) are negative values.
Table 2-2.
dBmV /Volt Conversion Chart
dBmV
80
Voltage
10.0 Volts
70
3.2
60
1.0
50
320,000 J.LV
40
100,000
30
32,000
20
10,000
10
3,200
0
1,000
-10
320
-20
100
-30
32
-40
10
21
2
Key Broadband Concepts
Unity Gain Trunk Design
When designing the trunk portion of the distribution system, the unity gain criterion
should be followed.
Unity Gain Criterion
1. All trunk amplifiers are identical (with respect to noise figure, gain and
equalization).
2. All trunk amplifiers are separated by an identical length of cable.
3. Flat Loss
+
Cable Loss
=
Amplifier Gain.
Result: All trunk amplifier output levels are identical.
Designing the trunk to this standard provides these advantages.
~
The system is easy to design by consistently following this rule.
~
The system is easy to align and maintain, since the output levels of all amplifiers
are identical.
This rule requires that either each trunk amplifier be adjusted to compensate for the
losses between its input point and the previous amplifier:s output point (see figure 24); or that each trunk amplifier be adjusted to compensate for the losses between its
output point and the following amplifier's input point. Thus each amplifier in the
system has the same output signal level, and the system has unity gain throughout (no
increase or decrease in signal level from one amplifier's output to the next).
UNITY GAIN
--.jiot-- UNITY GAIN
--~- UNITY GAIN
HEADEND
CL ~ CABLE LOSS (dB)
FL ~ FLAT LOSS (dB)
G ~ AMPLIFIER GAIN (dB)
V ~ SIGNAL VOLTAGE LEVEL (dBmV)
Figure 2-4. Unity Gain in a System
System Losses
The distribution system's losses can be divided into two main categories, flat loss and
cable loss.
22
2
~
Flat loss, or passive loss, is the attenuation through all the passive components in
the network (not including the cable). The value of this loss is constant across the
entire frequency spectrum.
~
Cable loss is the attenuation of the coaxial cable. This loss increases with frequency,
a characteristic called cable tilt.
To achieve the same amplitude for all signals at all frequencies of interest, it is
necessary to compensate for both flat loss and cable loss. This is accomplished with
equalizers and amplifiers.
An equalizer has an attenuation characteristic that is the inverse of the cable tilt
with respect to frequency. It attenuates low frequency signals more than high
frequency signals. Ideally, the combined cable and equalizer losses produce constant
attenuation across the system's entire bandwidth.
A flat gain amplifier following the equalizer increases signal levels across the
spectrum. Figure 2-5 demonstrates unity gain and shows how one stage of a network
compensates for cable tilt.
I
UNITIGAIN
I
>--..fl5lnr\---I FLAT ...----1
LOSS
CABLE LOSS + FLAT LOSS
=AMP GAIN
(a) Unity Gain
I~
20 dB CABLE
...\
-f",,--)_ _ _ _ _ _ _ _)-dBmV
FLAT INPUT
TILTED OUTPUT
+30
+22
+14
50 MHz
300 MHz
(b) Cable Attenuation
Figure 2-5. Unity Gain of One Stage
23
50 MHz
300 MHz
2
Key Broadband Concepts
-f1""i)~___- =20:=. .: d:.=B~C: : A: : B: :L:!:;E
_____
r-1 EQU~~~ER ~
dBmV
FLAT INPUT
CABLE OUTPUT
- -
__
:"Q':.A~Z:"R ~U2:P~T _
-cs=:=:::::=:::::::: - - - - - - - - - -
1
-
-
-
-
-
-
-
-
-
4-
+30
,-+22
-
+14
I T+6
I
(c) Cable and Equalizer
t---t>FLAT
-f'-')"--___
. :20~d=B. :C:;.A:.:B=L=E
r___i EQ~~~~ZER
____
dBmV
CABLE OUTPUT
FLAT INPUT
50MHz
-------------- - -
300
50
I
300
AMP. OUTPUT
EQUALIZER OUTPUT
--
r
1
50
300
--
---
50
+3o
+2 2
+1 4
+6
300 MHz
(d) Cable, Equalizer, and Flat Gain Amplifier
Figure 2-5. Unity Gain of One Stage (Continued)
Transparent System Design
During the design phase, a standard signal level should be established for all outlets in
a system. When this is done, connectivity requirements for the network are
transparent. Any device can be attached to any outlet and use the same network
services, regardless of its location and function. The input and output circuits of a
network interface device need be adjusted only once for the standard system levels,
after which the device can be used anywhere in the network. Thus, transparency
provides the following advantages.
~
Easy relocation of devices throughout the network.
~
Easy connection of new devices to the network.
24
2
For example, a network interface device can be moved from one office to another, and
plugged into the second office's outlet. At the new location, it can use and provide the
same services that it did at the old location. If the network was not transparent, signal
level differences at the two outlets might require adjustment of the interface device
whenever it was moved.
Many devices designed to communicate over broadband networks are built to
receive at a level of + 6 dBm V, and to transmit at a level of + 56 dBm V (referred to a
television visual carrier signal in a 6 MHz channel). By designing the network to
accommodate these signal levels, any such device could be used at any outlet without
requiring installation adjustments. (Signal levels are covered in more detail in chapter
five.)
Summary
This chapter covered several essential topics in broadband networking. The physical
structure of the network resembles a tree. Attached communication devices see it as a
bus, a ring, or a star. Channel protocols are used to control access to channels that have
more than a single transmitter on them.
Two-way communication can be accomplished with two one-way cables, or with a
single two-way cable that has a separate frequency band for traffic in each direction.
The transmission medium for broadband networks is coaxial cable. Many different
types of coaxial cables are available that have characteristics to match various
applications.
Network design calculations are facilitated by using decibels instead of Volts and
Watts. Signal levels established at the design phase should permit easy alignment and
maintenance. Following the unity gain principle, and designing for transparent
connectivity eases both of these tasks.
25
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3
Single and Dual Cable Systems
Introduction
To achieve bidirectional signal distribution, two basic approaches can be employed.
~
Two-way communications over a single coaxial cable, with different frequency
bands carrying signals in opposite directions.
~
Two-way communications over dual coaxial cables, with each cable carrying
signals in one direction.
This chapter discusses both approaches, shows how they can be be used together, and
compares their capabilities.
Single Cable Systems
Two-way communications can be implemented on a single coaxial cable by dividing
the available frequency spectrum on the cable into two bands. These bands carry
signals in opposite directions, called forward (away from the headend) and return or
reverse (toward the headend). Devices attached to the network transmit to the headend
on the return band, and receive from the headend on the forward band. Currently,
three different frequency divisions are used, called subsplit, midsplit, and highsplit.
Each provides a different amount of forward and return bandwidth.
When specifying CATV equipment for a network, be aware that manufacturers
use slightly different frequencies and bandwidths. Check each component's
specifications to ensure that all equipment in the system is compatible, and that any
filters can pass all the signals being transported.
Subsplit System
Most CATV two-way cable systems now in service use the subsplit format.
~
Forward band 54-400 MHz
~
Reverse band 5-30 MHz
~
Total usable bandwidth 371 MHz
This system has been popular with CATV system operators because it offers the easiest
method to upgrade existing one-way cable systems to two-way operation. It allows
transmission of all 12 VHF television channels on their normal broadcast frequency
assignments, which eliminates the need for special converters at each customer's site.
28
3
However, when more than 12 channels are to be distributed over a single cable system,
separate converters are necessary anyway, eliminating this advantage for many
systems.
The subsplit format has limited utility when information originates from locations
other than the headend. Since only 25 MHz is available in the return direction, only
four television signals or their bandwidth equivalent can be transported to the
head end at one time. The impact of this limitation depends on the type of equipment
used in the network. Microprocessor-based packet communication units available
today permit data networks with thousands of users to operate on a single 6-MHz
channel.
Midsplit System
Midsplit systems are used in many data communication networks.
~
Forward band 168 to 400 MHz
~
Reverse band 5 to 116 MHz
~
Total usable bandwidth 343 MHz
Midsplit is more popular than sub split for local area networks, because of its greater
return direction bandwidth. It can handle high volume two-way interactive
communications including data (both low- and high-speed) and video. The IEEE-802
specification (a standard currently being developed for local area networks) for such
networks endorses this format.
A midsplit system's greater bandwidth can be used by one or more services. For
example, when using a modem with a bandwidth efficiency of 2 bits per Hertz, a T1
channel (which conveys digital data at 1.544 Mbits/s) occupies only 772 kHz. A
midsplit system can provide over 140 such channels in its return path.
Figure 3-1 shows block diagrams of trunk amplifiers for subsplit and midsplit
systems. Appendix C summarizes the symbols used in this and other drawings in the
book.
29
3
Single and Dual Cable Systems
168 TO 300 MHz
IP~ ~u~zffil
~
L ___ -.l
FORWARD AMPLIFIER
TEST
POINT
H
TAUNK
DIPLEX
H
_ _ _~--L_~-I FILTER
L
RETURN AMPLIFIER
(a) Midsplit Format
5 TO 116 MHz
54 TO 300 MHz
FORWARD AMPLIFIER
H
L
L
RETURN AMPLIFIER
(b) Subsplit Format
Figure 3-1. Typical Two-Way Cable Trunk Amplifiers
30
5 TO 30 MHz
3
Highsplit System
This is the newest system of the three.
~
Forward band 232-400 MHz
~
Reverse band 5-174 MHz
~
Total usable bandwidth 337 MHz
A high split system fulfills the need for high return path bandwidth that some large
systems might have. Some amplifiers are now available in the highsplit format, but
standardization among vendors of these units has not yet been achieved.
Converting From A One-Way To A Two-Way System
The following factors should be considered when converting an existing one-way
network into a two-way midsplit network.
~
Use of individual frequency converters at each user device.
~
Expandability of present amplifiers to bidirectional use.
~
Redefinition of system frequency allocations.
~
Modification of existing services and their frequency allocations.
~
Selection of passive components that pass all the required frequencies.
~
Inspection of existing coaxial cables. Inspect them to ensure signal ingress will not
cause problems.
It is not necessary to disrupt service while upgrading a system to support two-way
traffic. Modular amplifier units allow easy installation of return amplifiers, equalizers,
and distribution legs in the field.
One study from the cable industry estimates that existing one-way networks can
be upgraded to two-way subsplit networks at a cost of around $300 per mile. *
Dual Cable Systems
Two-way dual cable systems use two coaxial cables laid side-by-side. One cable
provides the inbound (return) path signals to the headend. The second cable provides
the outbound (forward) path signals from the headend to the attached devices.
~
Outbound band 40-400 MHz
~
Inbound band 40-400 MHz
~
Total usable bandwidth 360 MHz
Not all dual cable networks use this same spectrum. For example, Wang Laboratories'
network uses non-standard amplifiers with a bandpass of 10 to 350 MHz.
* Ellis Simon, "Cable's Business Connection," Cable Marketing Magazine, January, 1982.
31
3
Single and Dual Cable Systems
Each outlet in a dual cable system must have two connections that clearly identify
the inbound and outbound paths. In addition, twice as many amplifier units are
required to implement a dual cable network, compared to a single cable network.
The term amplifier unit refers to the module that contains the gain block, power
supply, equalizers, and any associated circuitry. Currently one vendor supplies all the
necessary circuitry for both paths of a dual cable system inside a single module.
Shielding and isolation requirements dictated against placing circuitry for both
directions inside the same enclosure for many years, until improved isolation methods
were perfected.
Dual cable system amplifiers are used in the unidirectional mode with a bandpass
of 54 to 400 MHz. Figure 3-2 shows the block diagram of a dual cable amplifier.
RETURN PATH
TRUNKA IN
TRUNKAOUT
FORWARD
AMPLIFIER
..
TOWARD
HEADEND
TRUNK BOUT
TRUNK B IN
RETURN
AMPLIFIER
FORWARD PATH
Figure 3-2. Dual Cable System Amplifier
Dual cable systems have no interaction between inbound and outbound signals
except when devices are incorrectly connected to the network. No speCial filters are
required in amplifiers to provide frequency separation. As a result, amplitude and
phase distortion in a dual cable network are less than in a single cable network.
32
3
CATV Dual Trunk Systems
The two-way dual cable systems described in the preceding section are not the same as
CATV dual trunk systems. CATV dual trunk systems are composed of two one-way
trunks laid side-by-side to each subscriber's location. The subscriber selects signals
from only one trunk at a time with an AlB switch. This technique was a simple way
for early CATV systems to double their signal bandwidth. The AlB switch directs the
signals from one trunk to the television receiver and isolates the signals on the other
trunk from the receiver. Some CATV operators have converted one of these trunks into
a two-way system. When the proper trunk is selected by the user, two-way operation is
possible.
Connecting N etwor ks Together
Separate networks connect to a common trunk through devices called gateways (see
figure 3-3). These devices can be set to pass only a selected frequency band so that each
user can be isolated from other users on the common trunk. One well-known system
currently using this approach is in the New York area. Manhattan Cable Television
provides a dedicated commercial trunk to serve the banking community in New York's
financial district. This trunk is primarily used for passing data from one branch of a
bank to another. Other gateways can connect to national data transmission networks
SEPARATE LOCAL AREA NETWORKS
BUS
RING
INSTITUTIONAL!
COMMERICAL
OISTRIBUTION
RESIDENTIAL
CATV
DISTRIBUTION
' - - RESIDENTIAL --.-/
CATV
DISTRIBUTION
GATEWAY BETWEEN NETWORKS
HEADEND FORA BROADBAND
LOCAL AREA NETWORK
Figure 3-3. Interconnecting LANs with Industrial Trunking
33
3
Single and Dual Cable Systems
such as TeleNet and Tymnet and extend the range of a local area network across the
country. This technique (called industrial trunking) is primarily used in CATV systems
to provide dedicated or shared networks for industrial and business users.
Current technology does not permit two-way transmission over the full extent of a
large CATV network. However, one or more portions of a large network can be made
bidirectional to provide wide band communication links between nearby sites, such as
individual local area networks. A device on one local network needing to access a
device on another local network can do so over the CATV trunk (figure 3-3).
Single and dual cable systems can be combined together into useful network
structures. Figure 3-4 illustrates samples of these combinations. At first glance the
structures might appear complicated. However, they are simply expansions or
reconfigurations of basic single and dual cable structures.
TRUNK A
TRUNK A
FEEDERS
TRUNK B
TRUNK B
(a) Dual Trunk/Single Feeder
Figure 3-4. Combined Single and Dual Cable Systems
34
3
TRUNK A
TRUNK A
FEEDERS
TRUNK B
TRUNK B
(b) Dual TrunklDual Feeder
TRUNKA
TRUNKA
TRUNKB
TRUNK B
(c) Dual TrunklDual Feeder With Crossover
FEEDERS
Figure 3-4. Combined Single and Dual Cable Systems (continued)
35
3
Single and Dual Cable Systems
Comparing Single and Dual Cable Systems
This section discusses several advantages and disadvantages of single and dual cable
systems to assist in determining where one approach is more suitable than the other.
System Bandwidth
Although it is difficult to generalize for all networks, 60 MHz of bandwidth in each
direction has proven adequate for many applications. Several systems use only about
35% of the total bandwidth available.
Where wide bandwidth is necessary, there are two alternatives.
~
A dual cable system can be implemented.
~
Two separate single cable systems, each carrying different services, can be
implemented side-by-side.
Two single cable systems provide all the advantages of the single cable method
including simpler design, maintenance, and installation, in addition to increased
bandwidth and system redundancy. The disadvantages of this approach are cable
identification, trunk switching, and more complicated documentation requirements.
When applications require more than about 100 MHz in the return path, it is more
cost effective to install two midsplit cables instead of one dual cable system. This
provides 222 MHz in the return path and 464 MHz in the forward path.
Multiple Cable Systems
One alternative to a large cable system throughout a facility is to build several distinct
and complete systems that serve various subdivisions of that facility. At first glance,
this scheme might seem to offer several advantages over either a dual cable or two
single cables where large bandwidth is required in the entire facility. However, it has
some serious drawbacks.
~
Confusing layout
~
Redundant cabling and equipment
~
Duplication of resources
~
Diffused responsibility for maintenance
A better solution is to have an intra-plant cable trunk that provides the backbone for
shared services throughout the entire complex. Connected to this trunk are branches
that feed each building. Inside each building, one cable network provides the
necessary services. This approach is simpler to design, easier to maintain, and allows
better management of system noise.
Amplifier Capacity
One disadvantage of using high-bandwidth systems is that the capacity of the
amplifiers might not be great enough to handle the maximum load. For example,
36
3
assume that one 6-MHz channel contains 56 separate data carriers that use subchannels
of 96 kHz bandwidth each. If each carrier is transmitted at a level of + 56 dBm V, the
amplifiers are loaded with the equivalent of 56 separate 6 MHz-wide channels at a + 56
dBm V signal level. All the power capacity of the amplifiers could be consumed by the
signals from only 6 MHz of the spectrum, preventing any other signals from being
transmitted. In addition, this would distort each amplifier's output signal excessively,
and interfering signals could be generated at harmonic frequencies throughout the
spectrum.
To prevent overloading the amplifiers, narrow bandwidth data signals should be
transmitted at lower levels than wide bandwidth video signals. The amount of this
signal level difference depends on the number of carrier signals inside a 6-MHz
channel. Chapter five contains examples showing how to calculate what is called the
derated carrier level for this situation. These calculations are straightforward, since
Single cable networks using the same CATV technology and components have been
successfully applied for several years. CATV amplifiers are capable of full channel
loading, and the proper operating levels are well known in the industry.
Amplifier loading and data carrier derating can be a problem in dual cable systems
where the amplifiers must pass a much wider bandwidth. The proper operating levels
for such wideband networks have not been clearly established yet.
Components
Amplifiers for dual cable systems cost less than those for single cable systems. This is
because each dual cable amplifier housing does not hold the extra components (filters
and a second amplifier) used by the single cable system. However, twice as many
amplifier housings can be required in a dual cable system. Two separate housings are
commonly used in most systems.
Most amplifiers for single cable systems operate in standard frequency ranges and
with standard bandpasses. Subsplit and midsplit amplifiers are available from many
vendors. However, some dual cable systems use different frequencies and bandwidths
than others. Amplifiers for such systems might not be available from several sources,
which could lead to supply and repair problems.
Dual cable systems do not require the translator that single cable systems need.
Interface equipment for dual cable systems must keep signals from the two paths
isolated from each other while extracting the data from the RF carriers. Circuits that
achieve the necessary isolation can add to the cost of the interface device.
Installation
Dual cable systems take up more space than single cable systems because they use
twice as much cable, and they need twice as many amplifiers, passive components, and
other hardware. When retrofitting an existing facility, mounting space can be critical
and the smaller requirements of the single cable system are an advantage.
In addition, single cable systems are easier to install because all the directional
components face the same way. In a dual cable network, these components must be
placed in different directions, depending on the cable to which they attach.
37
3
Single and Dual Cable Systems
Maintenance
Repairing both single and dual cable systems is easier if the network has been designed
and built with proper components. Maintaining and troubleshooting the single cable
network is easier because of its simpler implementation with fewer cables and
components. It is easier to follow drawings and trace cables through the single cable
plant because there are no directional markings that could be incorrect.
The abundance of cables at every branching point can be confusing when trying to
trace the loop from source to destination in a dual cable system. Each piece of cable
must be marked correctly throughout the network to minimize the chance of mistakes
during connection.
Single cable systems also offer better reliability since they use fewer components.
Interface to Other Networks
Most industrial and institutional trunks provided by the CATV industry are single
cable systems. An in-house single cable system can be connected to such trunks more
easily than can a dual cable system. It can be desirable to use such services, since
bandwidth rental costs average from 10% to 20% lower than equivalent services from
the local telephone company.
Redundancy
The purpose of redundancy is to provide a backup for important network services
when the primary network is damaged. Single cable networks can provide
redundancy by installing a second cable system throughout the facility. A breakdown
on one cable leaves a functional system albeit with lower performance. On the oth.er
hand, losing one cable of a dual cable system would eliminate traffic in one direction
entirely, making two-way communication impossible.
Running two single cable systems side-by-side lowers the cost of installing the
second network. However, two cables along the same path do not provide as much
protection against failure as two totally different paths would. Alignment of both
routes of the system is straightforward, and tracing problems is only slightly more
difficult. Each outlet would have two connectors that could provide identical services.
A redundant dual cable network requires twice the cable, amplifiers, hardware,
and other components of a redundant single cable network. There are four cables in
the system for each drop and four separate connectors at each outlet. Each of four
cables at every location must be identified clearly.
~
The primary in bound
~
The primary outbound
~
The backup inbound
~
The backup outbound
Trying to trace a signal in a ceiling at a four-way splitter could be cumbersome. This
maze of cables is exactly what local area networking is intended to eliminate.
38
3
Outlets
Since the outlet is the one component that the user sees and might have to manipulate,
it should have a clear and simple design. Many different types of single cable outlets
are available, including self-terminating devices. These provide protection from signal
ingress and egress and maintain proper matching in the network. They can also meet
military and government requirements for network security. The use of standard RF
connectors enables repairs to be made quickly, using inexpensive tools.
Dual cable outlets require two separate connectors at each outlet. To prevent
installers and users from crossing the paths at the outlet, two different types of
connectors are recommended. When a second, redundant path is required, four
connectors must be installed at each outlet. If the installation uses nonstandard RF
connectors, maintenance problems could arise. When a connector breaks, Murphy's
law will prevail and there will be every type of connector on hand except the one that
is needed.
Future Developments
The development of cable communications over the past years has dictated some
changes in user equipment and system choices. When precise wide band equipment
was not available, the only way to obtain additional signal bandwidth was to use a
second coaxial cable. As better components and techniques were employed in the
design of RF data modems, more signals could be placed in narrower frequency bands.
Also, many more users could be connected to a network without degrading its
response. These factors enable more users to operate on less cable spectrum than in the
past. This means that a modern single cable system can provide more capacity at a
lower cost than an older dual cable system.
Most current broadband local area networks use only seven channels (42 MHz) or
less for video conferencing and video security applications. Increased bandwidth will
be necessary as new video applications requiring interactive control arise, such as the
following.
~
Videotext servers that provide individually addressable video frames.
~
Video switchers that route signals throughout a network.
~
Remote-controlled video tape recorders and video disk players that can be used
interactively for teaching and entertainment.
By expanding a single cable network from 300 to 400 MHz, 16 additional 6-MHz
channels are available for such services. Upgrading a network in this way can be more
economical than either installing a second cable or installing a dual cable network
originally. Advances in bandwidth compression could bring such services to single
cable networks without requiring extensive modification.
39
3
Single and Dual Cable Systems
Conclusions
Several studies have concluded that the single cable system is the best method with
which to implement broadband local area networks. Cox Cable Communications
found that both one-way and interactive two-way services can best be provided by a
packetized data service on a single cable network. Other Multiple Station Operators
(MSOs) have tested frequency agile modems in networks with over 20 amplifiers in
forward cascade. The results have been data bit error rates from a worst case of one in
108 to one in 109 bits transmitted, with a noise level between 15 and 27 dB below peak
video carrier on the system. A one-way network upgraded to two-way operation can
supply a much better 44 dB carrier-to-noise ratio for the return path with about 200
miles of cable plant activated. *
In general, technology, cost, and installation considerations favor the use of single
cable systems for the distribution of RF signals for data processing, audio, video, and
control applications. However, each application should be analyzed critically with all
the above points in mind before choosing between a single or dual cable system.
* Claude Baggett, Paul Workman, Michael Ellis, "Upstream Noise and Bit-Error Rates Analysis of an
Operational One-Way System Converted to Two-Way Operation," Cable '81 Technical Papers: The Future of
Communications, ed., NCTA (1981).
40
4
Broadband Components
Introduction
This chapter describes various broadband system components. All passive and active
components discussed have 75 ohms impedance and provide 100 dB or better shielding
from radio frequency interference (RFI), electromagnetic interference (EMI), and signal
ingress. Appendix G lists some of the tools needed to install and maintain these
components.
The Coaxial Cable
Chapter 2 described coaxial cable and briefly mentioned some of its features. This
section provides a more detailed examination of some specific properties and uses of
coaxial cable for broadband networks. The types of cable used in broadband networks
can be divided into three application categories: trunks, feeders, and drops. These can
be installed with or without conduits, depending on the insulation material, the
environment, and local building codes. Finally, the variations of cable attenuation with
frequency and temperature are covered.
Because it has been useful in so many applications, coaxial cable has developed
greatly over the past decades. Even the least expensive cable can provide 80 dB
shielding effectiveness and low loss across a 400 MHz bandwidth. Figure 4-1 shows the
composition of a typical coaxial trunk cable.
D.
A.
A.
CENTER CONDUCTOR: Center most feature of coaxial
cable. it consists of solid cooper or copper clad aluminum
wire.
C.
B.
DIELECTRIC: Electrical insulation utilized to maintain
position of the center conductor. It is composed of foamed
polyethylene. This insulator Ipositioner may also be evenly
spaced polyethylene discs.
D.
FLOODING COMPOUND: (OPTIONAL) A viscous substance
placed between the outer conductor C and the jacket E to
maintain a protective seal should the jacket E contain or
develop any cuts or openings.
E.
JACKET: (OPTIONAL) A black polyethylene coating over the
aluminum outer conductor to provide a weather-tight seal.
diameter.
Figure 4-1. Terminology for Coaxial Trunk Cable
42
OUTER CONDUCTOR: Is constructed of an aluminum tube.
The cable size (412. 500. 750 & 1000) is derived from its outside
4
Types of Cables
Cables used in coaxial cable networks can be divided into three layers.
~
The first layer, the trunk cable, transports signals between amplifiers.
~
The second layer, the distribution or feeder cable, connects the trunk cable to the
vicinity of the subscriber or office.
~
The third layer, the drop cable, links the feeder cable to the outlet.
Trunk Cables
Trunk lines come in six sizes, ranging from 0.412 to 1.000 inches in diameter. These
cables exhibit attenuations from 1.6 to 0.5 dB per 100 feet (at 300 MHz). Their
construction includes a rigid aluminum shield of seamless tubing with a bending
radius of 10 times the diameter, covered with a strong polyethylene jacket. In some
cables, a flooding compound is injected between the aluminum shield and the outer
jacket to provide protection in underground installations. Inside the shield a foam
dielectric surrounds the solid copper or copper clad aluminum center conductor.
Generally, all trunk lines should be 0.500 inches or larger in diameter. Cables run
outside buildings or mounted to poles are usually jacketed. Cables buried or placed in
conduits should use corrosion-resistant flooding gel between the outer jacket and the
aluminum shield. The gel protects the aluminum from corrosion if the jacket is cut or
damaged. Armored cable with floodiryg gel is mandatory where the cable is buried
underground without further protection, and where it is mounted in underground
vaults and might be damaged by water or rodents.
Cables with messengers are available for suspension between buildings or on
poles. This feature provides protection and eliminates the need for strand lines.
Feeder Cables
Smaller-sized trunk cables are used for feeder cables. These indoor cables are selected
according to the following criteria.
~
The physical constraints of the building: smaller cables are easier to install.
~
The required signal level for the distribution network: larger cables have less
signal loss.
~
Local and national building codes.
In general, jacketed or unjacketed 0.500-inch aluminum cables are used for trunks and
feeders.
Drop Cables
Drop cables connect feeder cables to network outlets. These cables need not be very
large, since only one cable is used for each outlet. They range from RG-U and RG-6 to
RG-59. Each type incorporates foil and braid shielding to prevent radiation and pickup
of RF energy. The outer jacket is made of an insulating material.
43
4
Broadband Components
Drop cable lengths vary from 10 to 50 feet. They can be installed above ceilings
and through walls. A drop can connect directly to a wall outlet or to a device such as a
television receiver, modulator, demodulator, or data modem. To minimize pickup of
noise and broadcasted RF signals, the best quality and best shielded cable should be
used for drops.
Installation Considerations
Cables can be installed throughout a facility with or without conduits. This choice
depends on the type of insulation used by the cable, and on building codes. Fire codes
might prohibit PVC-coated cables in ceiling plenums or computer floors because toxic
fumes could be circulated if a fire occurred. In these cases, PVC cable can be placed
inside a conduit, or cable with a fire-retardant jacket (such as Teflon TM) can be used.
To compare the cost of using Teflon-insulated cables to installing conduit, consider
the following approximate data.
~
Teflon cable.
The cost per 1000 feet of Teflon cable is higher than that of PVC cable (1984 price
quotes range from $0.50 to $1.50 per foot).
Installation of Teflon cable is more expensive than PVC cable because of
special connectors and longer labor time to install these connectors.
~
Conduit.
A rule-of-thumb states conduit installation cost, including materials, to be around $1
per foot.
Conduit provides additional physical protection for the cable, and additional
shielding from radiated signals.
During construction, buildings can be piped with conduit for cables. Passive
components and amplifiers should be installed inside enclosures appropriate for the
environment. Each enclosure should be located to provide access for alignment and
maintenance.
Proper ventilation should be provided for components mounted inside enclosures.
It is important to maintain the temperature inside the amplifier hOUSing as close as
possible to the temperature experienced by the cable. Enclosures installed inside
buildings might require fans to prevent heat build-up. Enclosures installed outdoors
might not need fans, depending on the amplifier's operating temperature range and
environmental conditions.
Although coaxial cable is durable, it is not invulnerable. When transporting and
installing coaxial cables, always handle them carefully. The cable should be left uncut
and fastened securely until it is required for installation. During installation the cable
should not be kinked or bent beyond the specified limits.
In overhead installations, several factors relating to safely mounting coaxial cables
must be considered. A good source of information is the article "Guidelines for
Handling Trunk and Feeder Cables"* from Times Fiber Communications. This article
is easy to read and highly informative on cable-handling techniques including tension
factors, cable reels, and pulling limitations.
Teflon™ is a registered trademark of DuPont.
* Coaxial Cable Catalog, Times Fiber Communications, January, 1982.
44
4
Cable Attenuation
The attenuation of a coaxial cable is often quoted as a single number, such as 10 dB per
100 feet. This is the attenuation of the cable at the highest frequency of interest for the
system (usually 300 or 400 MHz). However, cable attenuation is not constant and
changes with both frequency and temperature.
Frequency Variation
Cable attenuation increases with increasing frequency in a nonlinear (exponential)
manner. This characteristic is due to the composition of the cable and is called cable
slope or cable tilt, and it must be considered when designing a distribution network.
Figure 4-2 shows the attenuation change with frequency for a 20 dB length of
0.500-inch cable. Figure 4-3 shows attenuation per 100 feet versus frequency for several
different sizes of cable. This graph shows that as cable diameter increases, cable loss
decreases, which is why larger cables are preferred for long cable runs. The smaller
coaxial cables have more loss, and only short lengths of them are used in drop cables.
20
'1
J(
1/
Channel
7
15
m
"0
Z
2
~
10
-I
I
I
CJ)
n
Ii V
1
~
j, V ......
5
50
60
70 80 90 100
150
FREQUENCY IN MHz
Figure 4-2. O.SOO-inch Cable Loss
45
200
250
300
4
Broadband Components
10
10
8
8
L..-
6
5
4
. / 1'1.0-'
. / ./
--'"
3
j::'
u.
2
a:
w
Q..
./
1.5
m
~
Z
0
i=
«
::::>
zw
....
....
«
t/' V
1,..-1,..-
0
0
r.~C
1.0
.8
.6
.5
.2
1.0-'
.1
3
2
.... . /
V V
1.5
l /1/
1.0
.8
I.'"
V 1,.0' ....
~
~ ~ ~~~ ~
)P'
~
I-" 1Cb9.,. ~
~
1-""'-
5
~
10'
./
...C:Y
....1-"
/
1,..-1.-
j.;
L\\.\~ ~
/
u~ ~~ l,....~ )\~ P-\~ ~
~
~
~
~I,..-
V
-,1".;0
./
.4
.3
~~
V
V V
j.;
~
6
5
4
./
.6
I-"
.5
.4
17
.3
.2
1,.0'
10
20
30
50
100
200
400
700 900
.1
FREQUENCY (MHz)
Figur~
4-3. Cable Attenuation Versus Frequency tor Various Sizes ot Coaxial Cable
Temperature Variation
Cable attenuation is also directly affected by temperature variations. The attenuation
of coaxial cable increases with temperature at the rate of 0.11% per degree Fahrenheit.
This amounts to an overall change of about 15% over the temperature range of -40 to
+ 120 degrees Fahrenheit. The accepted rule-ot-thumb is 1% change in cable attenuation
for every 10 degree Fahrenheit change in temperature at a given frequency.
The network must function properly despite any RF signal level changes caused
by frequency and temperature variations. In bidirectional systems there are two
different variations that must be considered: one for the forward path and one for the
return path. The design must take into account cable tilts for both of these frequency
bands. These variations affect the operation of the amplifiers (because they determine
equalizer selection and setting) and the overall peak-to-valley response of the system.
Amplifiers must compensate for the combination of cable loss, tilt, and temperature
variations experienced in daily operation. Proper compensation ~eeps system gain and
signal levels reasonably constant under all possible conditions.
Amplifiers
Because of the parameter variations caused by frequency and temperature changes,
only small systems can successfully transport signals without requiring compensation
46
4
circuitry. Modular amplifiers in the distribution system can include various equalizers,
gain blocks, filters, and other circuits that make up for cable-caused variations. Signal
level gain corrects for attenuation caused by the cable and by other components.
Frequency compensation (equalization) corrects for cable tilt.
Amplifiers used in a broadband network can be divided into four different
categories.
~
Trunk
~
Bridging
~
Line extender
~
Distribution
Each of these types is discussed in the following paragraphs after a description of
general amplifier characteristics.
General Amplifier Characteristics
Amplifiers are differentiated by their cost and performance. The cost factor is
straightforward: the more expensive units usually provide better performance. The
cost of a unit can depend on the following characteristics.
~
Gain is the increase of signal level occurring from input to output of the amplifier.
~
Output level is the maximum signal level that the amplifier can deliver.
Noise figure is the amount of noise contributed by the amplifier.
Distortion is the amount of unwanted modification of the input signal done by the
Amplifier gains are usually 20 to 30 dB.
~
~
amplifier (this includes inter modulation products, which are often specified
separately).
~
Gain control of an amplifier can be either manual or automatic. Automatic gain
control units are more expensive.
When evaluating the gain specification of an amplifier, the design engineer should
include any associated passive loss. The usable gain of an amplifier is the gain available
from a fully configured unit and it equals total gain less the insertion loss. A configured unit might contain additional modules such as filters and equalizers. These
modules have an insertion loss that must be subtracted from the total gain of the
amplifier. This loss is generally between 1 and 3 dB, depending on the configuration of
the amplifier.
Amplifier Gain Control
Two different types of gain control for an amplifier are available, manual and
automatic.
~
47
The gain of a manual gain control (MGC) amplifier is mainly adjusted by hand.
Variations caused by temperature changes can be accommodated automatically by
an internal thermal compensator circuit that changes the amplifier's gain. This
4
Broadband Components
combination of MGC with thermal compensation suits broadband local network
applications, because most of them do not exist in harsh environments.
(Amplifiers for outdoor CATV networks must accommodate wider signal level
changes found in that environment.)
MGC amplifiers are primarily used on short-range, high-signal-level
distribution trunks. Their cost per unit is less than that of automatically-controlled
amplifiers.
The thermal compensation provided by MGC units is desirable only when the
amplifier is subjected to the same temperature variations as the cable. A problem
can arise, for example, if the amplifier is installed in a pedestal above ground and
the cable is underground. In that case, the amplifier would probably experience
greater temperature variations than the cable. To eliminate possible
overcompensation, the thermal circuit should be removed from the amplifier. In
most cases, doing so requires no special tools or training.
~
The automatic gain control (AGe) amplifier maintains a relatively constant output
level regardless of input level variations. It can accommodate changes of 3 dB
above or below the nominal input value. When used properly, AGC amplifiers can
provide constant signal levels to all outlets in facilities that experience varying
environmental conditions.
A general rule-oj-thumb is that a 6-dB change in the input signal of an AGC
amplifier causes a I-dB change in its output signal.
Types of Amplifiers
Four main types of amplifiers are used in broadband systems: trunk, bridging, line
extender, and distribution amplifiers. Each offers different characteristics, performance, and features, and each is appropriate in different applications. These amplifiers
are further described in the following paragraphs.
Trunk Amplifiers
Trunk amplifiers are high quality, low distortion units capable of being cascaded into
long chains to distribute signals throughout a large geographic area. Amplifiers are
cascaded, or connected in series along the trunk cable, to make up for losses and
variations encountered in long cable runs.
Trunk amplifiers are typically operated at 22 dB gain, with input levels of 8 to 10
dBm V and output levels of 30 to 32 dBm V for 35-channel systems with 20 amplifiers in
cascade. Where fewer amplifiers are cascaded, output levels can be increased up to +45
dBm V. When any amplifier is to be operated above its suggested output level, the
manufacturer should be consulted for advice.
A rule-oj-thumb for cascading amplifiers is that each time the number of amplifiers
in series is doubled, the output level of each unit must be reduced by 3 dB from its
rated output. For example, the output level of each amplifier in a cascade of two units
should be 3 dB below the rated output of the amplifiers, or less. Doubling the number
of amplifiers in series up to four requires that the maximum output level be reduced to
6 dB below rated output, or less. With eight units in cascade, the maximum output
level should be 9 dB below rated output, or less.
48
4
Bridging Amplifiers
The bridging amplifier, or bridger, provides high level signals for distribution on the
branch or feeder lines. They can be installed inside the same housing as the trunk
amplifier. The output signal level of a bridging amplifier is usually +47 dBm Vat the
highest operating frequency.
A bridging amplifier receives its input signal from the tap of a directional coupler
connected to the output of a trunk amplifier. One to four output lines are available for
distribution.
In a broadband network, a common trunk line can feed several buildings. The
bridging amplifier can drive distribution cables that feed the individual buildings.
With this approach, trunk amplifier levels can be adjusted to CATV standards,
allowing easy cascading and future expansion.
Figure 4-4 illustrates an example of signal levels for a trunk and bridging amplifier
combination. Return path signal levels, when not given on the drawings, are equal to
or slightly greater than the forward path signal levels.
FORWARD
+10 dBmV IN
FORWARD
+32 dBmV OUT
+37 dBmV FORWARD 1 - + 1 - - +23 dBmV RETURN
+32 dBmV OUT
RETURN
FEEDER
MAKER
0
+14 dBmV IN
RETURN
+6 dBmV
ASSUME 22dB trunk spacing
31 dB loss between point A and the user connection
54 d8mV signal at the user connection injected into the return path
Figure 4-4. Trunk and Bridging Amplifier
Test points are provided at both the input and output points of the amplifier. The
input test point is used to sample the signal before the input filter, pad, and equalizer
modules. The output test point is used to sample the signal after the amplifier,
directional coupler, and filter sections.
The terms input side and output side are relative to the direction of signal
transmission. The input side of an amplifier in the forward path corresponds to the
output side of an amplifier in the return path, and vice versa.
49
4
Broadband Components
Line Extender Amplifiers
Line extender amplifiers, or line amplifiers, are used when the signal level provided by
the bridging amplifier is insufficient to drive receiving devices. These amplifiers cost
less but have higher distortion and noise figure specifications than trunk and bridger
units. Line extender amplifiers should be limited to a maximum cascade of three to
provide acceptable quality signals to the users.
Some smaller two-way networks use line extenders as their only amplifying
device. Such systems have the following characteristics.
~
Cascades of three or less
~
Many outlets located within a small area
~
Coverage of a limited geographical area
Line extender amplifiers are available in the subsplit, midsplit and highsplit formats
for two-way applications, as well as in dual cable versions.
Internal Distribution Amplifiers
Internal distribution amplifiers are high gain units used for signal distribution. They
can be used where several high level feeder legs are required, for example, over several
floors within a building. Cascading is not recommended because of their higher gain.
One advantage of such amplifiers is that they have built in lIO-Volt ac power
supplies and do not require ac power to be transmitted over the cable. Currently these
amplifiers are available in subsplit and midsplit versions only.
Amplifier Module Additions
There are several additional circuit modules that can be included inside amplifier
housings. These modules can provide signal attenuation, return channel gain,
frequency equalization,. and remote control.
Attenuators
When the input signal amplitude is too high, a pad can be installed inside the
amplifier module, in series with the amplifier's input, to reduce the level.
Bidirectional Amplification
All four types of amplifiers can be used in bidirectional networks by adding
appropriate filters and a second amplifier module for the return path. Return path
amplifiers usually have less gain (19 to 26 dB) than forward path amplifiers since cable
attenuation at the lower frequencies (return direction) is less than at the higher
frequencies (forward direction).
Equalizers
Variable equalizers to compensate for cable tilt can be installed in each amplifier
housing. These circuits provide a frequency response that is the inverse of the response
50
4
of the coaxial cable. The combined effect of the cable and the equalizer is to provide
equal attenuation to all signals regardless of their frequency.
Adjustable equalizers can accommodate different cable lengths. A single equalizer
circuit can be used inside all amplifier housings, which only needs to be adjusted for
the length of cable between it and the previous amplifier.
Feeder Disconnect
A feeder disconnect circuit can be added into an amplifier housing, usually a bridging
amplifier. This circuit permits disconnection of a feeder line from the trunk, either
remotely or locally, which can help when troubleshooting and repairing the network.
~
When noise or unwanted signals are entering the system from an unknown point,
disconnecting one feeder line at a time can help to isolate the source of the
problem.
~
When aligning the system, disconnecting branches from the trunk can help to
check and match the signal levels coming from all branches.
Control signals for feeder switching originate at the headend and are generated by
status monitoring systems and intelligent amplifiers. The state of each module (on, off,
or remote), can be selected individually. Placing the module in the on or off state
connects or disconnects the feeder and trunk. In the remote state, a signal sent from the
headend controls feeder switching.
Power Supplies
All amplifiers require ac power. The internal distribution amplifier contains its own
power supply, and can connect directly to a lI~-Volt ac outlet. All other units run on
either 30 or 60 Volts ac that is delivered over the coaxial cable by power supplies.
Distributing power in this manner eliminates the need for lI~-Volt ac outlets at each
amplifier location and allows greater flexibility in amplifier placement.
Thirty-Volt power is used mostly in older systems. Sixty-Volt power is used widely
in modern broadband communication networks.
AC power is coupled to the coaxial cable through devices called power combiners.
These devices permit the injection of power in either or both directions with little
effect on the radio frequency signals.
Once power is delivered to the cable, multi-taps and amplifiers can control its
distribution.
~
For safety reasons, multi-taps pass current along the trunk connections but
prevent it from reaching the outlets. Each outlet is electrically isolated from the
main network and from other outlets, reducing the possibility of total system
failure from accidental or malicious causes.
~
Amplifiers can pass or block ac power travelling on the cable. Power can be passed
to other amplifiers or stopped at either the input or the output of each unit.
51
4
Broadband Components
Observe these precautions when sending ac power over coaxial cable.
~
AC power should not be injected through multi-taps or couplers incapable of
passing power. Typically, units unable to pass power have F-type connectors.
~
Use only cables with seamless aluminum shields to convey power.
~
Consider the current-passing capabilities of each device in the network to ensure
that limits are not exceeded.
In system design, a general rule-at-thumb has been one power supply for every three
amplifiers and cable spans. This quantity depends on cable resistance and amplifier
operating current and voltage. Large networks require calculations using power supply
voltage and current capacity, amplifier current draw and required input supply
voltage, and cable loop resistance. The power available for each amplifier can be found
by Ohm's law. These calculations will show where additional power is required.
One result of calculating power requirements is learning that amplifiers are
voltage dependent and not current dependent. If the voltage drop across the cable
between an amplifier and its power supply is too great, the amplifier operates poorly if
at all. To solve this problem, another power supply can be added to the system, or a
cable with less resistance can be installed.
Standby power units can be incorporated into any broadband network. These units
provide power when the network's main ac input line fails. Surge protection gives each
unit some protection from high voltage spikes that can occur when power is applied.
Clean and reliable ac power should be used when available. One source of such
power is the source used for a computer room, which is also a good location for the
broadband network's headend equipment and its power supplies. Such sources often
have backup units to ensure that the computer does not fail during a general power
failure.
Figure 4-5 shows a typical power connection scheme for a power supply in a
broadband network.
.6dB
RFOUT
RFIN
REMOVABLE FUSES
POWER COUPLER
ACIN
r--------,
:
~
110VAC
,~8LJ,
L ______ -1
BACK-UP POWER SUPPLY
POWER SUPPLY
Figure 4-5. Power Supply Configuration
52
4
Grounding
Grounding the system at every amplifier helps to ensure long, reliable service. When
ground connections deteriorate, amplifiers can be damaged because of high shield
current developed from the electrical system's neutral line. All ground points should
be checked at least once each year and ground resistance readings taken to ensure
system integrity. Make no assumptions about the quality of existing grounds. Verify
that the cable system has at least one good ground by measuring existing grounds or
by making your own ground. For details on grounding aspects, the Grounding
Principles* booklet by Copperweld is recommended. Further information on
grounding can be found in appendix I.
Passive Components
This section describes the passive components used in broadband networks. Passive
components require no power to operate and include connectors, couplers, splitters,
taps, filters, outlets, and terminators.
All passive components cause a certain amount of signal loss when they are
inserted in a network. Unlike cable attenuation, which varies exponentially with
frequency, this RF signal loss is constant across the entire frequency spectrum and is
called insertion loss or passive loss. The amount of insertion loss is different for each type
of device and is specified by the manufacturer. For example, couplers and taps can have
insertion losses between 0.4 to 2.9 dB, depending on the tap value. Sample insertion
loss values are shown in figure 4-6.
3.5 dB EACH LEG
.5 dB
___~~g==
~
TAP VALUE
--~(
MULTI-TAP UNIT
)
SYMMETRICAL OUTPUTS
SPLITTER
.6dB
~
TAP VALUE
DIRECTIONAL COUPLER
Figure 4-6. Insertion Loss Values
* Copperweld Bimetallics Group, Robinson Plaza 2, Pittsburgh, PA 15205,412/777-3000.
53
4
Broadband Components
Connectors and Hardware
Solderless 75-ohm connectors designed for each type of cable are the most important
components in a broadband system. Industry experience has shown that 75% of all
system failures are directly or indirectly caused by connector failure or by poor
connector installation.
Connectors come in many varieties from many manufacturers. It requires care to
select connectors suitable for the coaxial cable being used and for the environment in
which they are used.
A wide range of mounting hardware for the cable and for other components is
available from many vendors. Connectors are weak points in the distribution system,
and should not be subjected to physical stress by supporting equipment with only the
cable. Equipment should never be physically supported by the coaxial cable. The cable
and the system components (amplifiers, multitaps, directional couplers, and power
supplies) should be separately and securely fastened as the situation dictates. Shrink
tubing should be used where connectors enter equipment to ensure the integrity of the
connector and to prevent corrosion. Where splices are made in underground
installations, materials are available that can be installed over the splice to provide a
watertight seal.
Figure 4-7 shows some of the types of connectors currently on the market.
Appendix J provides further details on connectors, their parts, and how they attach to
equipment.
Directional Couplers
The directional coupler has three ports:
~
Trunk Input (placed toward headend)
~
Trunk Output (placed away from headend)
~
Tap
The main cable attaches to the two trunk connections in the appropriate direction. The
branch cable connects to the tap point, and receives only a portion of the signal at the
trunk input terminal. Signals originating from devices connected to the tap point are
also attenuated by the directional coupler, and always flow toward the headend (to the
trunk input terminal of the directional coupler).
The directional coupler provides a means for both dividing and combining RF
signals while maintaining the system's 75-ohm impedance and isolation
characteristics. The directional characteristic ensures that signals being transmitted
from any network device go only toward the headend, and minimizes the reflection of
RF energy back to its source.
Four parameters describe the RF performance of a directional coupler:
~
Insertion loss
~
Tap loss
~
Isolation
~
Directivity
54
4
FEED THRU (VSF)
©
PIN TYPE (STINGER)
SPLICE
A connector that seizes both the outer and center
conductors. This device has an additional feature
not found in the feed thru type consisting of a
solid brass pin which seizes and retains the cable
center conductor. The pin then extends thru the
body and is retained within the equipment
housing.
This connector is utilized to join together two
cables. It seizes both the outer and center
conductors (as in the pin-type).
F FEMALE
A device used when an F type female port is
required at the end of the cable. This connector
seizes both the center and outer conductor of the
coaxial cable.
F MALE
This device is used when it is necessary to have
an F type male connection at the end of the cable.
This connector seizes both center and outer
conductors of the cable.
CABLE TERMINATOR
This connector is used in a cable system where it
is necessary to terminate both RF signal and
60 Hz AC power. This device seizes the center
and outer conductors.
Figure 4-7. Connector Types
55
A device that seizes only the outer conductor of
the coaxial cable. The cable center conductor
extends thru this type connector and is retained
within the equipment housing.
4
Broadband Components
Figure 4-8 illustrates the applicable parameters for the directional coupler and for the
splitter. In broadband applications, high isolation alone is not the most important
parameter. Directivity, which is the difference between isolation and tap loss, is the
significant parameter.
All coaxial devices that combine or split signals use the principle of the directional
coupler. During installation the directional coupler must not be installed backwards.
Doing so would create reception problems at that point and transmit signals in the
wrong direction. Arrows indicating signal flow direction are stamped on all devices.
TAP LEG
8.5 dB Loss
8dBmV
3.5 dB loss per leg
RFOUTPUT
RF INPUT
20dBmV
14.8 dBmV
- ' - - - - - 1 6 . 5 dBmV
1.7 dB loss
(a) Typical Signal Levels
DIRECTIVITY = ISOLATION - TAP LOSS
INSERTION
LOSS
INSERTION
---.
LOSS
TRUNK IN
TRUNK OUT
ISOLATION
TAP
LOSS
TAP
(b) Splitter
(c) Directional Coupler
Figure 4-8. Directional Aspects in Cable Systems
Multi-taps
A multi-tap combines a directional coupler with a signal splitter, and allows the
connection of several drop cables to the system. Multi-taps are placed along the feeder
cables to proVide connections to outlets, or to provide access to the network for strings
56
4
of single-port taps which loop from office to office. Insertion loss between the trunk
input and trunk output connections is low. A greater attenuation exists between the
trunk input and the tap output lines.
There can be two, four, or eight separate tap connections, called ports, from a
multi-tap. By changing the tap assembly, the number of ports or their attenuation
value can be easily changed without disconnecting the entire multi-tap from the cable.
Knowing the length of cable between the port and the outlet it feeds, the attenuation
of the port can be selected to allow matching the total attenuation of each path from
the headend to each outlet in the system. In addition, multi-taps provide isolation so
that all outlets stand alone. Connecting or disconnecting a device on one tap does not
affect the operation of the overall system.
Components in distribution legs are usually adjusted to deliver relatively flat
levels at the middle of the branch, so that most outlets receive a relatively flat signal.
Cable tilt makes it difficult, if not impossible, for all outlets to have a flat signal level
across the entire frequency band. Most systems operate properly with a signal level
variation of 3 dB above and below the nominal level at any outlet.
Figures 4-9 and 4-10 show a multi-tap and its use in a typical distribution system
and signal levels at each tap.
TAP VALUE = 20 dB
.6 dB INSERTION LOSS
~
TO HEADEND
ATTENUATION
FROM INPUT
TO OUTPUT
I - I
I GI--+I_C_O_N_NE_C_T_IO_N_S
RF INPUT
-~
DROP CABLE
ATTENUATION FROM INPUT
TO OUTLET CONNECTIONS
TYPICAL WALL OUTLET
,--
--I
TRUNK OUT
I
-20 dB
I
I
L_
I
I
_-1
FOUR WAY
SPLITTER
TO OUTLETS
Figure 4-9. The Multi-Tap
57
4
Broadband Components
+33 dBmV @300 MHz
+29.5dBmV
+29.5dBmV
+28.5
+8.5dBmV
+28
+27
+7 dBmV
+26.5
+25.5
+8.5
+25
+24
+7
+23.5
ASSUME 1 dB loss at 300 MHz for interconnecting cables
+ 7 dBmV signal level at 300 MHz required at each tap port
0.5 dB insertion loss for each tap
Calculations represent forward path only
+4 dBmV
Figure 4-10. The Multi-Tap and Distribution Legs
Programmable Taps
Manually-programmable four-way directional taps featuring security traps are now
available. Unlike normal and computer-controlled taps, the manual tap accepts any of
three plug-in modules to program each port for:
~
Basic Service
~
Full Service
~
Termination
58
4
Modules available in the future will allow each port to be individually programmed
for one, two, or three pay channels. These taps are useful in CATV networks, but are
rarely used in broadband local network applications.
Filters
Filters are used in several applications in most RF systems. Most filters located at the
headend combine or separate frequency bands associated with antennas, channel
processors, and two-way filter/combiners.
Bandpass filters (BPF) pass an aSSigned portion of the spectrum and attenuate
signals at frequencies above and below that passband.
Bandstop filters (BSF) attenuate signals in a given frequency range and pass all
others.
Diplex filters (diplexers) are used in two-way single cable networks. They direct
Signals in the two frequency bands (high and low) to the correct processing
equipment, and have three connections. One is the common port and the other two are
for the frequency bands associated with the return and the forward paths (high port
and low port). Signals of both these frequency bands appear at the common port.
Diplex filters are used at the input and output of each amplifier housing used for
two-way communication over a single cable. Signals from the headend are passed by
the diplexer to the forward amplifier and isolated from the return amplifier. Signals
from network-connected devices are passed from the diplexer to the return amplifier
and isolated from the forward amplifier. The diplexer limits the amount of crossmodulation and intermodulation associated with two-way interactive systems.
Envelope delays associated with diplex filters are easier to handle in the midsplit
format than in the subsplit format.
Figure 4-11 gives typical frequency response plots for these filters.
5 MHz
116MHz
(a) Low Pass Filter Response
iJ
168 MHz
5 MHz
(b) High Pass Filter Response
300/400 MHz
LOW
5 MHz
116MHz
168 MHz
(c)
300/400 MHz
Diplex Filter Response
Figure 4-11. Amplitude/Frequency Response of Typical Filters
59
4
Broadband Components
Outlets
The wall outlet can be a a single gang plate with two female F connectors mounted
back-to-back (called a barrel connector), or with a self-terminating outlet that
automatically terminates when the attached device is removed from it. Alternatively,
the outlet can be a directional coupler which has only one tap. This allows offices to be
wired through a looping coaxial cable, with the tap available to the user and the
through portion available to connect to adjacent office taps.
All unused outlets should be terminated by manual insertion of a 75-ohm
terminator, or by using self-terminating outlets. Terminating all unused outlets can
significantly limit the ingress of undesired signals in the return path. When using
manual termination, ensure that the terminator is attached to the outlet plate with a
chain or a stainless steel cable. Manual termination is the most positive means to
control ingress.
Terminators
Termination of distribution lines or unused tap ports is important to provide proper
matching, to maximize power transfer, to limit reflections, and to minimize ingress of
undesired signals. Simply stated, a terminator converts RF energy to heat.
Seventy-five-ohm terminators come in several varieties. They are available for
both indoor and outdoor applications. Where 60-Volt ac is on the coaxial cable, the acblocking type must be used.
Figure 4-12 shows the two symbols that are used to designate terminators.
I~>---AC BLOCKING TERMINATOR
Figure 4-12. Terminator Symbols
60
RESISTIVE TERMINATOR TYPE
5
Factors Affecting System Design
Introduction
This chapter provides details on planning, designing, and implementing a vendorindependent broadband local area network (LAN). This information covers the tasks
performed and the vocabulary used by the designer. It includes guidelines to assist the
network manager during the design phase.
The network manager must be aware of the factors considered when a broadband
network is designed. However, it is not necessary for a manager to have a detailed
understanding of these factors; the broadband design engineer brings that expertise to
the project. The manager should know enough about networks to communicate
effectively with the technical personnel. The engineer then designs a system that
solves the problems described by the manager. This chapter does not provide a list of
procedures to follow to design a network. It does identify and describe important
design factors. More specific information can be obtained from the cable design group
with any major broadband LAN equipment vendor.
Any broadband or CATV-based network should be designed by a qualified coaxial
system engineer. When a company uses its own telecommunications department t.o
design a network, a broadband system consultant should review the design and
equipment specifications. Such a review is recommended because system layout is a
craft that relies on repeated trials and comparisons, rules-of-thumb, and experience.
A major advantage of a properly-designed network is that it can be easily
expanded. Systems built according to sound principles several years ago can be
upgraded smoothly to support new services, greater traffic, and new communication
devices. Without proper design, it is much more difficult to change and expand the
existing system.
The contents of this chapter cover five main areas.
~
The initial approach to a network, including structure, layout, and frequency and
bandwidth requirements.
~
Sample design calculations including signal levels, noise levels, and distortion.
~
Reliability factors to be considered in network design.
~
Headend design.
~
Typical system specifications.
Initial Considerations
Consider the following factors at an early stage in the design of the local area network.
62
5
~
The information utility. Keep in mind the idea that a broadband network is an
information distribution utility. Like any utility, it should be planned, designed,
and installed independently of any specific application that it might support.
~
Future expansion. When construction is underway or when new services are being
added to a facility, consider installing coaxial trunk cables, even if there is no
current need for a network. These cables can be used to build a broadband system
in the future.
~
Geographical coverage. Determine the maximum geographical extent of the system,
including a good estimate of future requirements.
~
Building survey. Inspect underground vaults, building access points, ceiling
construction, and wall composition. This inspection can aid in designing the
layout and can help a contractor provide a realistic cost estimate of the project.
Local building contractors can do the building survey. Specialized Engineering
Contractors (ECs) can provide total turnkey service including proposals, surveys,
design, installation, alignment, and maintenance. Some or all of these services can
be used when considering any large broadband network.
~
Network architecture and topology. Determine the general plan for the system.
t>
t>
t>
Single or dual cable.
For single cable select subsplit, midsplit, or highsplit.
Main trunk routing.
Star, ring, or bus architecture for data transmission devices.
~
Frequency allocations. Select services to be carried and make frequency assignments
so that future expansion can be easily done.
~
System headend. Locate the headend centrally within the system. Criteria such as
serviceability and network management could dictate a different placement.
Adequate space should be available for signal processing, data translation, and test
equipment.
~
Antenna siting and cabling. When broadcast television is part of the system,
carefully plan the location of receiving antennas and the routing of cables. The
construction and alignment of antennas can be difficult and dangerous. Safety is
the main concern; operating and performance characteristics are secondary.
~
Trunk design. Make a preliminary layout of system trunk cabling. The best
arrangement for a multiple trunk distribution system is to use several trunks
radiating from the central headend.
~
Redundancy, reliability, and repair. Consider reliability and repair requirements
during system design. This will minimize the need for expensive redundant
components to ensure network availability.
Status monitoring systems and redundant components can be installed in any
network. However, redundancy often does not pay for itself unless the system was
poorly designed or poorly installed in the first place. Systems that have been
properly designed and certified after installation have had few problems.
Some of these factors are discussed in greater detail in this and other chapters of this
overview.
63
5
Factors Affecting System Design
System Structure
The structure of a network must be determined early, for this has an effect on many
other network design factors. Most commercial and industrial broadband systems have
a tree architecture. This versatile form of organization permits data to move between
any two points in the system. It readily accommodates all coaxial cable based systems
and does not require switching to establish connectivity between any subset of users.
System structure also involves selecting a single or a dual cable system. If single
cable is chosen, the frequency division scheme can be subsplit, midsplit, or highsplit.
The interface devices that operate together determine the network topology of a
channel, which can be bus, ring, or star. Chapter 3 of this overview discusses the major
aspects of system structures.
System Frequency Considerations
The highest operating frequency of the system can be 300, 400, or 450 MHz. Today,
most systems are designed for 300 MHz, although 400 MHz systems are becoming
more widespread. Regardless of the system's design bandwidth, all passive
components should be able to pass signals from 5 t'J 400 MHz. Amplifiers should be
able to pass any signals in the system's spectrum.
To determine the necessary bandwidth for a system, draw up a frequency allocation
chart. Such a chart shows the services occupying each frequency band and helps to
ensure that there are no conflicting frequency assignments. The frequency allocation
chart in appendix D shows frequencies used in over-the-air transmission and in CATV
transmission. This chart can be a good starting point when assigning frequencies for a
broadband network.
A general procedure for making a frequency allocation chart includes the
following steps.
1.
Identify the needed services that can be provided by the network.
2.
Estimate the number of users of each service, and calculate the required resources:
Operating speed and throughput.
Number of channels.
Bandwidth required for each service including guard bands between different
services.
3.
Evaluate interface equipment from all appropriate vendors that can provide the
desired services. Select those devices that satisfy your needs.
4.
Assign network frequencies to those services whose interface equipment operates
on fixed frequencies and cannot be easily or inexpensively changed.
5.
Assign network frequencies to the remaining services whose interface equipment
can be used on different frequencies.
Additional bandwidth for the expansion of each service should be reserved. This
enables the growth of each service without the need to change existing frequency
assignments. Commonly used expansion factors are
64
5
~
Subsplit: 40%
~
Midsplit: 30%
~
Dual cable: 40%
For example, the initial design of a midsplit system with 100 MHz of bandwidth
available in the forward path should not allocate more than 70 MHz of that bandwidth.
If less than 30 MHz is available for the future expansion of such a system, some
method for obtaining more bandwidth should be considered at the outset.
Regardless of the system's design bandwidth, all passive components should pass
all signals between 5 and 400 MHz. Amplifiers need pass only the required bandwidth
for the frequency spectrum and allocation for which they are being used.
Bandwidth Requirements
The bandwidth requirements of a system are based on current and projected services.
These services can be divided into the following categories.
~
Closed circuit television
~
Data communications
~
Special services
~
Broadcast television distribution
Each of these services is examined in the following paragraphs, especially regarding
bandwidth requirements.
Closed Circuit Television (CCTV)
Each security monitoring device or educational television channel requires 12 MHz of
bandwidth, 6 MHz in each direction.
A typical CCTV application would include a television camera generating a video
signal and an optional microphone generating an audio signal. Both of these signals
are sent to a modulator. The modulator produces a composite RF signal on a channel in
the return band. This RF signal is transmitted to the headend where it is translated
into an RF signal at a higher frequency in the forward band. The higher frequency
signal is distributed throughout the system. A television receiver can be connected at
any point in the network to monitor these video and audio signals.
Data Communications
The bandwidth needed for data communications depends on the services needed, the
number of users, and the equipment providing the services. This section provides
some details on the types of data communication devices available for broadband
networks.
A common use of broadband data communications is to support moderate speed
data traffic between many remote terminals and one or more host computers. A
bandwidth of 6 MHz can accommodate over 4000 terminals operating on several
distinct subchannels with equipment from one vendor. A different application might
require continuous access to a data channel. This can be accomplished with fixed
65
5
Factors Affecting System Design
frequency, point-to-point RF modems. Such devices made by one manufacturer create
28 dedicated, two-way ports inside a single 6-MHz bandwidth.
Most manufacturers specify the number of data subchannels that occupy a 6-MHz
bandwidth. The 6-MHz channel proVides a common reference for allocating
frequencies on the network. The number of subchannels and their bandwidth inside a
6-MHz slot is different for equipment made by each manufacturer; no industry-wide
standards exist in this area as of this writing.
The following paragraphs discuss two categories of interface devices.
~
Fixed frequency, point-to-point or multi-point modems.
~
Frequency agile, multiplexed, interactive packet communication units.
Each type of interface device is compatible with existing two-way broadband systems.
When the broadband system is to be used as a local area network, additional
bandwidth might be necessary to allow both types of modems on the same cable. The
major differences between them are
~
the bandwidth of each subchannel;
~
the tuning ability of the frequency agile unit;
~
the method used to access the cable.
These differences determine which device is best suited to an application. Given the
number of users of each service and knowing the devices to be used to prOVide that
service, the necessary bandwidth can be calculated.
1. Point-to-Point Modems.
A point-to-point modem operates only on its assigned channel. The operating
frequency of each modem is fixed. Only one transmitter can be used on any given
channel without interference. Full duplex communication between two points in a
translated system requires four distinct frequencies: two for transmitting (one from
each location) and two for receiving (one at each location).
Point-to-point devices are best suited for applications requiring a dedicated
channel and continuous access to the network. One manufacturer provides such
devices using carrier signals every 96 kHz. A 6-MHz bandwidth can contain up to 28 of
these dedicated two-way ports.
2. Multiplexed Modems.
Devices that need not have continuous access to the network can be multiplexed
together to share a single data subchannel. All units can continuously monitor the
channel, but only one device at a time can transmit. Channel access protocols
determine which station transmits.
Multiplexed interface units are well-suited for applications where traffic is slow to
medium speed and bursty. Bursty traffic is characterized by short periods of
transmission separated by long periods of inactivity, which is typical of people
interacting with computers. Many devices operating in this manner can effectively use
a single channel without overloading it, since each transmission is brief.
66
5
One type of currently available unit uses carrier signals every 300 kHz. This
spacing provides 20 distinct subchannels inside a 6-MHz bandwidth. These units are
specified to support 200 asynchronous ports at 9600 bits per second on one such
subchannel. A network of 4000 ports would require only 6 MHz.
Special Services
A wide range of services can use the network, with each one consuming some portion
of the network's bandwidth.
~
Local origination signals, such as those from a television studio, require 6 MHz for
the forward path and 6 MHz for the return path for each video channel.
~
High resolution video for closed-circuit television (CCTV) applications can use
from 4 MHz to 14 MHz of bandwidth in each path.
~
Digital control signals using dual frequency coding (frequency shift keying)
usually require from 300 kHz to 1 MHz depending on the number of signals and
the transmission speed.
~
A telephone system using frequency division multiplexing (FDM) can
accommodate 300 simultaneous conversations by using about 6 MHz in each
direction.
The manufacturer's specifications list the bandwidth requirements of each device that
uses the network.
Television Distribution
Television signal distribution needs a separate book to be described adequately. For this
overview, only a brief discussion is possible. Broadcast signals are captured with
antennas and are filtered, amplified, and converted with channel processors.
Understanding the placement of towers, the selection of antennas, the interaction of
signals, and the specification of filters and signal amplitudes can take years of
experience. Designing multiple channel headends that include large arrays of
interactive equipment and control circuits is a complex task, best left to experienced
broadband design engineers.
In general, 6 MHz is required for each television channel transmitted in the
forward direction. An additional 6 MHz in the return path is necessary for each link to
a remote studio whose signal must be transmitted over the network.
When the frequency of a received television channel must be changed for cable
distribution, care must be taken to be sure that the conversion is not a prohibited type.
Some channel assignments cannot be converted to others because of co-channel
interference and interaction with adjacent channels. These restrictions can also apply
to translating some return channels to forward channels. Consequently, an allocation
chart must be made and all frequency assignments must be checked for compatibility.
When television signals are to be transported on the coaxial network, the reports
made by the Television Allocations Study Organization (TASO) in 1959 should be
consulted. These reports provide details on signal quality and how it is specified.
Copies can be obtained by writing to any cable television publication or to the
National Cable Television Association (see chapter seven).
67
5
Factors Affecting System Design
Bandwidth Allocation
Integrating a communication system consisting of data, video, and voice equipment
manufactured by several companies can be a formidable task. The combination of
multiple services can require large amounts of system bandwidth. It is important to
estimate the total bandwidth required when designing the network.
When assigning operating frequencies, those subsystems that cannot be obtained
with selectable frequencies should be assigned first. Tunable subsystems are then
assigned frequencies in the remaining spectrum. Managers making channel allocations
should always consider current and projected requirements.
Federal Communications Commission (FCC) regulations must be investigated to
avoid using prohibited frequency bands. Although broadband systems are not directly
regulated by the FCC or by the Federal Aviation Administration (FAA), certain rules
about radiation, restricted bands, and similar matters are generally followed. The
network should meet or exceed the technical standards set forth in the FCC rules,
Section 15, under the heading MATV systems.
When possible, the use of aircraft frequencies at locations within 60 nautical miles
of an airport or a transmitter site should be avoided. This regulation currently applies
only to CATV operators with over 1000 subscribers. Designing systems to comply with
this restriction can prevent expensive redesign or reconfiguration if similar regulations
are imposed on broadband local area networks at a later date. There is no current
indication that FCC/FAA compliance will be required for local networks in the near
future.
Figure 5-1 shows frequency allocation charts to illustrate bandwidth
considerations. Such a chart can be made by first identifying the channels on the
system, and then dividing the spectrum into forward, return, and guard bands. When
a channel is allocated to a specific service, mark the block directly below that channel
number on the chart (as are TS and no in the return path of figure 5-1(a)). Appropriate
guard bands should be maintained between channels allocated for different services.
Information on necessary guard bands and proper filtering should be available from
broadband equipment vendors. When a channel is to be translated from the return
path to the forward path, connect the corresponding frequency allocations with a
dotted line as shown.
Physical Layout
In a multiple building system, the design of the physical layout should never loop a main
trunk cable through one building and into another. Instead, the trunk should be run
alongside each building. Directional couplers connect the trunk to branches that run
inside each building. If one branch is damaged, the other parts of the network can
continue to operate. Overall network operation is not affected since each branch is
isolated from the others.
Signal distribution inside a building should be divided into independent sectors.
A user in one sector should be able to function despite a failure in another sector.
The number and locations of outlets should be carefully planned. Allow connections
for all potential services including data, television, video, voice, and control. An
advantage of coaxial cable is that a single drop cable can support all these services at
68
5
RETURN PATH
5 MHz
54 MHz
30 MHz
300/400 MHz
I
FORWARD PATH
t
_______________ .J
Ic.~=-9~----!t ~'1.151'1 ~8~1O~1211,'1
FM
I}
-----------------~
t~
TRANSLATED
CHANNELS
(a) Subsplit Format
5 MHz
116 MHz 168 MHz
I
300/400 MHz
I
I
i 12~.151·1 tf 1718~:~;~~rl ~t S
RETURN PATH
I
I
I
L _ _ _ _ _ _ _ _ ..J _ _
l _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .J
L ____________________
~
I
_ _
.J}
TRANSLATED
CHANNELS
~
(b) Midsplit Format
CHANNELS ARE GENERAL IN PLACEMENT
AND IN DESCRIPTION
Figure 5-1. Frequency Allocation Charts
one location. When several different devices are to be connected in one office, greater
signal strength or multiple outlets must be supplied. For example, if a television
receiver and a computer terminal in each office are to connect to the network, two
separate outlets should be installed.
Many networks quickly grow beyond the size envisioned by their designers,
despite comprehensive planning from the start. In many cases, most of the resources
reserved for future expansion were consumed shortly after the initial installation was
completed. A limited network forces users to compete for access, creating additional
problems. To avoid such problems, the system should be designed to provide full tap
coverage (capable of supporting an outlet in each office) and include a 25% expansion
factor. This can be done, for example, by reserving one RF port on each 4-port tap for
future use. When such expansions are to be done in staggered phases, multi-taps with
interchangeable circuit boards are used.
Network branches can also be made expandable. The best approach is to place a
splitting device at each major branching point. One port can feed the current network;
the other port can be terminated to reserve it for future use. To increase the coverage of
the network, new distribution wiring can be installed, checked, and connected to the
69
5
Factors Affecting System Design
previously-reserved port. After the new segment has been certified, it can be attached
to the expansion port with little or no interruption in network service.
In two-way networks, the layout of the return paths is more demanding than that of
the forward path. Since the output signals from all return amplifiers eventually
converge at the headend, so does all the noise contributed by those amplifiers. The
forward path from the headend to any outlet has only the noise build-up of the
amplifiers in cascade with one another in that particular path.
In addition, the return path signal levels from all feeder legs at the input of any
amplifier must be as equal as possible. A distribution system meeting this criteria can
be equalized effectively.
For these and other reasons, the cable loss in each distribution leg should be the
same. Note that cable loss was specified here. The passive components can vary in
value, number, and location, but the cable lengths should be as equal as possible.
Signal Levels
A reference signal level for the system must be determined. This is the amplitude of
the RF signal that appears at each outlet and is measured at the network's highest
frequency. It is the value used to design the distribution network and to determine
amplifier gain. Maintaining this standard reference amplitude in the entire network
produces a transparent distribution system, as described in chapter two. This section
describes using the video carrier level as a reference, and how signals with different
bandwidths can be related to that reference level.
The Video Reference Level
The most common reference level for a broadband system is the video carrier signal
measured in dBm V within a 6-MHz passband.
The video reference level specifies both input and output signal levels:
~
The amplitude of video signals received at any outlet
~
The amplitude of video Signals injected into any outlet from a transmitting device
By designing the network to convey television signals properly, any other signal level
can be related to the video reference level and the proper receive and transmit levels
can be calculated for them. Any interface device can be attached to the network,
aligned for proper transmit and receive levels, and operated successfully. Each
broadband equipment manufacturer uses different signal amplitudes and bandwidths
for their interface devices. Establishing a reference level eases the design of a network
that uses equipment from several different manufacturers.
The following statement can be used to specify a transpareht system.
The broadband system shall provide for the distribution of color or monochrome
television Signals to any outlet in the system. In addition, the capability to originate
television signals from any outlet in the system shall be considered in every design.
Other non-television signals shall be compatible with the system for distribution,
using a variety of transmission techniques and access schemes. Signals such as data,
control, and audio shall conform to the standards set forth for television distribution.
70
5
The following values refer to a 6-MHz video signal and satisfy the transparency
criteria. These values are used in many operating networks.
~
The distribution system supplies a signal level of + 6 dBm V to each outlet. This is
the device receive level.
~
The interface device supplies a signal level of + 56 dBm V to the distribution
system. This is the device transmit level.
~
The forward path loss is about 50 dB. (This is the loss from the headend to a typical
outlet.)
The receive level is near the middle of the input signal range of a typical television set,
which is from 0 to + 15 dBm V. Normally, it is desirable to have the receive level
between 6 and 10 dBm Vat each visual carrier frequency, with the aural carrier level 15
dBm V below the visual carrier level to minimize interference and to provide good
reception. The cable network should be designed with all outlet signal amplitudes
within 3 dB of each other. This means that the variation from the lowest signal level to
the highest Signal level at the outlets must be 3 dB or less.
The transmit level of +56 dBmV also originated with television equipment
specifications. This is a standard output level for television modulators. The forward
loss of 50 dB is found by subtracting the receive level from the transmit level.
Figure 5-2 shows a small portion of a distribution system and how the desired
receive level can be obtained at an outlet.
*20 dBmV
@ 216 MHz
'42 dBmV
'39.8 dBmV
TRUNK CABLE
r----{>-----
22dB ~-+---------+
200' of 0.5" cable
loss = 1.1 dB!100'
@216MHz
8 dB (actual loss = 9.2 dB)
30.6 dBmV
FEEDER CABLE
(negligible loss)
+10 dBmV
TAP
Select Tap Value
(20 dB tap)
=
20.6
100' of drop cable
loss=4dB
OUTLET
+6 dBmV
Figure 5-2. Signal Levels from Trunk to Outlet
A different network, providing 35-channel service with a cascade of 20 amplifiers, uses
the following levels. (Both network designs are based on a video carrier reference
level.)
~
+8 to + 10 dBmV amplifier input level.
~
+ 33 to + 35 dBmV amplifier output level.
This network has 66-dB rejection of second-order beat frequencies, which is within the
range required for good network performance.
71
5
Factors Affecting System Design
Bridging amplifiers for the same channel capacity usually have higher output
signal levels (about +45 to +47 dBm V), but fewer such units can be connected in
cascade.
Narrow Bandwidth Carrier Levels
Broadband amplifiers are specified in terms of visual carrier levels and 6-MHz video
channels (see figure 5-3). When data communications devices with many carrier
signals in that same 6-MHz bandwidth are used on the system, the transmission level
of each data sub carrier must be lower than the video reference level; otherwise, the
amplifiers in the system could be overdriven. This would distort their output Signals,
and create interfering harmonic signals across the entire frequency spectrum of the
cable. The necessary carrier level can be calculated with the following carrier derating
formula.
DCL = VCL - IOlog(NC)
where DCL = carrier level for the desired data signal
carrier level for a 6 MHz video signal
VCL
=
log
= the base ten logarithm
NC = the maximum number of data carrier signals that can occupy a 6-MHz
assignment
The level difference between the video and data signal levels depends on the
number of data subchannels within a 6-MHz channel. As more subchannels are
squeezed into a 6-MHz bandwidth, less gain is available for each signal. The derating
formula can be used to determine the maximum signal amplitude for each data carrier,
and the manufacturer's specifications should be checked to ensure that this level is
adequate for the application.
Two examples using the specifications from two different broadband data
transmission units are provided to show how to calculate data carrier levels for a
transparent system.
AMPLITUDE
(dBmV)
56
VIDEO BANDPASS
38
VIDEO CARRIER
COLOR SUBCARRIER
o~
- - +____________________________
- - - - - - +3.58
________
~~
I
2
.
I
3
FREQUENCY (MHz)
Figure 5-3. A Typical Television Channel
72
______
~
·1
+4.5
o
SOUND
~
4
CARRIER
15-18dB DOWN
5
Example 1
Distribution system specifications:
Typical output VCL
+56 dBmV
Typical input VCL
+6dBmV
Manufacturer's specifications:
Data subchannel bandwidth
= 300 kHz
Number of carriers inside 6 MHz
= 20 carriers
Find the typical input and output DCL.
DCL = VCL - 1OIog(20)
= VCL - 13 dB
Typical output DCL
Typical input DCL
+56 - 13
=
+6 - 13
=
+43 dBmV
=
-7 dBmV
=
Given the distribution network's reference levels and this particular interface
device, the interface device's transmitter should supply +43 dBmV to the network,
and its receiver should expect a -7 dBmV signal from the network. If the interface
device cannot be adjusted for these levels, pads could be used or the reference levels
could be changed.
Figure 5-4 shows the spectrum of the device used in this example referred to the
same 6 MHz scale as the television channel.
13dB DOWN
AMPLITUDE
(dBmV)
56
43
300kHzCARRIER
/
'/c '"'""'"'''"'
~ nm /.
'\
I 1
I 1
f2'"
f2
;:::
;:::
;::: ;::'
.....
111 2
1 1
3
4
5
7
1«>1 a>
0
1.... 1
~
~
N
....
«>
;::'
....
'" ~'"
~
8
9
10
11
I
I
a>
N
'"....
"' ;!
'"
;! ;!
....
"'
12
13
14
16
~
1
..
....
~
it:
FREQUENCY (MHz)
17
18
19
CHANNEL NUMBER
~
1 1
0
-I
6
300 kHz CHANNEL
(4-- BANDWIDTH
!
I
I
FREQUENCY (MHz)
Figure 5-4. A Typical Data Channel
73
15
5
Factors Affecting System Design
Example 2.
Distribution system specifications:
Typical output VCL
+56 dBmV
Typical input VCL
+6dBmV
Manufacturer's specifications:
Data subchannel bandwidth
= 96 kHz
Number of carriers inside 6 MHz
=
56
DCL = VCL - 1000g (56)
=
VCL -18 dB
Typical output DCL = + 56 - 18 = + 38 dBm V
Typical input DCL = +6 - 18
-12 dBmV
=
Comparing these two examples confirms that when more subchannels are used in
the same 6-MHz bandwidth, the signal levels must be lower for equivalent quality
transmission. Using the interface device's specifications, proper signal levels for use on
a transparent system can be computed.
Figure 5-5 shows the relationship between signal levels and bandwidths for a
television channel and for the two data channels discussed in/the two examples.
TELEVISION
DATA
TELEVISION
DATA
/,.---~I\,---....,,/,....----,I\,-----..,/,.---~I\'-------..·v~----1I\~_---..,
VIDEO
- - -
-
-
-
-
-
AUDIO
-
-
-
-
-
-
-
-
-
-
-
-
VIDEO
- - -
-
-
-
DATA CARRIERS
-
-
-
-
-
-
AUDIO
-
COLOR
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
56 dBmV
DATA CARRIERS
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
43 d8mV
38 dBmv
COLOR
--:-----,----,-1~~...I....I...I..L.J..I-U.-.----'---~I~-~
I...- 6MHz ~
6MHz
.1.
20 CARRIERS
300 kHz
6MHz
~
6MHz
OdBmV
---l
56 CARRIERS
96 kHz
Figure 5-5. Television and Data Carriers
Narrow Bandwidth Advantages
The previous two examples show that many narrow bandwidth data subchannels can
be transmitted in a 6-MHz channel without interference by reducing their carrier
amplitudes below the video carrier reference level. Two consequences of using narrow
bandwidth signals arise.
74
5
~
A narrow bandwidth signal can operate successfully with a lower signal-to-noise
ratio. This is because noise increases with bandwidth. The noise floor of the 300kHz-wide data channel is -70 dBm V, while the noise floor of a 4-MHz-wide
television channel is -59 dBm V.
~
Lower amplitude data signals create less intermodulation distortion than higherlevel television signals.
Basing the system's design on video signal levels allows the construction of a
transparent network that supports a wide range of interface equipment. Using the
carrier derating formula provides the proper signal levels for operating each device.
In addition to channel bandwidth, noise is an important factor in determining
operating levels. The following section discusses thermal noise and how it affects
signal levels.
Noise Level
Thermal noise is generated by any device operating above a temperature of absolute
zero. The amount of thermal noise generated is a function of bandwidth and
temperature. For a television system with a 75-ohm impedance and operating at 68
degrees Fahrenheit, a channel with a 4-MHz bandwidth has a noise floor (the minimum
noise level possible) of -59 dBmV. A 300-kHz-wide channel on the same system has a
noise floor of -70 dBm V. The following equation gives the noise floor of a system with a
bandwidth B.
En
=
-125
+
1OIog(B) dBmV
Appendix E has details on the source of these figures and of other calculations used
in this section.
Noise Figure
The noise figure of an amplifier is the amount of noise that it contributes to signals that
it amplifies. It is a property of the amplifier and cannot be changed by alignment.
Increasing the signal level at the amplifier's input changes the carrier-to-noise (C/N)
ratio, but does not change the amplifier's noise contribution. An amplifier with a noise
figure of 7 dB raises the noise floor of the 4-MHz system from -59 to -52 dBm V. In a
cascaded system, the noise contributed by the amplifiers increases by 3 dB every time
the number of amplifiers is doubled. The following table lists noise figure for several
values of cascaded amplifiers.
75
5
Factors Affecting System Design
Table 5-1.
Effect of Cascading Amplifiers on Noise Figure
Amplifiers in Cascade
System Noise Figure (dB)
1
Catalog spec. of amplifier
2
Catalog spec.
4
Catalog spec.
8
Catalog spec.
16
Catalog spec.
+ 3 dB
+ 6 dB
+ 9 dB
+ 12 dB
32
Catalog spec.
+ 15 dB
This relationship can also be expressed in the following equation:
+ IOlog(N) dB
F
=
Fa
where F
=
noise figure of the system including all amplifiers
=
noise figure of one amplifier
Fa
N = number of amplifiers in cascade
Also, the noise floor of a cascaded 4-MHz system can be computed with
En
=
-59
+ Fa +
IOlog(N) dBmV
System Carrier-to-Noise Ratio
Carrier-to-Noise (C/N) ratio is the difference between the input signal level and the
noise floor. For an input signal of 10 dBm V, the C/N ratio of the single-amplifier
example system (with amplifier noise figure of 7 dB) is 62 dB.
To calculate the C/N of a cascaded trunk system, use the following rule: the C/N
ratio decreases by 3 dB every time the number of amplifiers is doubled. This can be
stated mathematically in the following formula (derived in appendix E).
C/N
=
C/NO - IOlog(N)
where C/N
=
the C/N ratio of the system including all cascaded amplifiers
C/NO
=
the C/N ratio of one amplifier
=
-Noise Floor
=
the number of amplifiers
N
76
+ Input Level- Noise Figure
5
For a system with 32 amplifiers in cascade, a 59 dBm V noise floor, a 10 dBm V input
level (typical), and a 7 dB noise figure, its C/N can be calculated as follows.
C/N
62 - 1010g(32)
62 -15
= 47 dB
For convenience, the CATV /broadband industry has related the following C/N
values to subjective evaluations of picture quality (table 5-2). A system's C/N ratio
should be greater than or equal to 43 dB. The worst case value is measured at the
farthest point of each branch.
=
=
Table 5-2.
Picture Quality for C/N Values
C/N (dB)
Picture Quality Rating
45
Excellent, no distortion
35
Fine, distortion just perceptible
29
Passable, distortion perceptible
25
Marginal
Noise at a Splitter/Combiner
The effects of noise in a bidirectional network are not identical for both directions. One
factor contributing to the difference is the passive signal splitter. This component
divides one signal travelling in the forward direction into two or more signals for
distribution along different paths. In the return direction, this device works as a
combiner. It combines signals from two or more paths into a single signal for
transmission to the headend. The following discussion describes what happens at a
splitter.
~
Forward Direction
Figure 5-6 (a) shows two carrier signals at the common port of a splitter, and the
resulting signal and noise levels that appear at each leg. The same output spectrum
appears at both output legs. The carrier Signals at frequencies Fl and F2 are
decreased in amplitude by 3 dB. The noise level is also reduced by 3 dB.
~
Return Direction
Figure 5-6 (b) shows two different carrier signals, one at each of the two input legs
of the combiner. Each carrier signal is attenuated by about 3 dB after passing
through the device on its way to the headend.
Noise, however, is a different matter. To calculate the worst-case condition,
assume that noise at input port A is in phase with noise at input port B. The noise
from each leg is attenuated by 3 dB after passing to output port C. Since the
components making up the noise in both paths are in phase, they add together at
port C. The result is an output noise level that is the same as the input noise level.
The result is a decrease in C/N ratio of 3 dB.
77
5
Factors Affecting System Design
For combiners with more than 2 input legs, the attenuation is greater, and
there could be a decrease in the noise level at the output leg. However, the decrease
in signal level is still greater than that of the noise, so the C/N ratio could be
degraded.
For this reason, noise effects in the return path must be considered separately in the
design of a cable system. Be sure that the C/N ratio at the headend from the farthest
transmitter is adequate for good receptIon.
FORWARD SPECTRUM
v
N
TO HEADEND 4 - -
-
-------fC. . .). . ___~:.. V/2
®
@
N/2
~i£lU,Q.w.w.w.LC..l1fJd
F1
(a) Splitter
RETURN SPECTRUM
v
V/2
N
N
@
TO HEADEND ..... -
-
------l()
:
~~-~---
®
v
N
(b) Combiner
Figure 5-6. Noise at a Splitter/Combiner
78
F2
5
Amplifier Selection
Selection of the type of amplifier to use depends on the application, the performance
and features desired, the signal level requirements, and the number of amplifiers to be
cascaded.
~
Trunk amplifiers, fitted for midsplit applications, are used in most situations.
~
Combined trunk/bridger amplifiers are used when the distribution system calls
for multiple legs with equal cable loss.
~
Line extender amplifiers are used only in small networks with little need for
future expansion.
Another criteria for selecting an amplifier is the type of gain control used, automatic
gain control (AGC) or manual gain control (MGC). Chapter 3 contains descriptions of
both types. In small systems, there is little justification for the additional expense of
using AGC amplifiers. In larger systems, the expense can be justified and their use
improves system performance significantly.
A combination of both AGC and MGC amplifiers is often employed in a large
distribution system. When too many amplifiers in cascade have AGe, a noise spike can
cause overcompensation problems. If the noise spike is large enough, it can decrease the
gain of all AGC amplifiers. When this occurs, the amplitude of all desired signals
passing through the amplifier is reduced, and the signal-to-noise ratio is severely
degraded. A noise pulse lasting several microseconds can cause a loss of desired signals
for several seconds
One approach to accomplish AGC has a pilot carrier signal transmitted over the
network at a single frequency. The pilot signal is processed by the AGC circuit in each
AGC amplifier, which determines the gain of the amplifier. If a noise spike at or near
this frequency occurs on the system, amplifier gain drops and data errors or loss of
desired signals results. For this reason, AGC amplifiers are often placed at every third
to fifth unit in cascade with MGC units. This configuration minimizes
overcompensation and provides control over system variation.
AGC is sometimes used in the return path. Special blocking filters and trunk
combining considerations require careful design and component selection to control
return path gain. Use AGC in the return direction only when suggested by the
manufacturer of the broadband equipment.
Design and Performance Calculations
This section provides sample calculations of system parameters including amplifier
gain, output signal level, C/N ratio, and inter modulation distortion. Although these
examples are presented simply, the design of any broadband system requires a great
deal of care and RF signal distribution knowledge. Never attempt to design a large or
complex system without first obtaining training or help from a qualified RF Systems
Engineer. Usually, the design is performed by a qualified engineering contractor, or
through the design services offered by the network equipment manufacturer.
79
5
Factors Affecting System Design
Amplifier Gain
Amplifier gain is the one factor that allows the designer to overcome the loss caused by
the coaxial cable. Usable gain is the amount of amplification the device can supply, less
any flat loss associated with its internal modules, and less any reserve gain.
~
All signals on a single cable system pass through two diplex filters for each
amplifier module encountered. The loss through an equalizer module is
proportional to the length of cable being equalized; 2 dB is an approximate figure
to use in estimates.
~
Reserve gain is a small amount of amplifier gain set aside during the design process
to accommodate signal level variations that can arise when implementing and
using the network. This gain can be used when the length of an installed cable run
exceeds the estimated value used for design calculations.
Amplifier Gain (Forward Path)
Minimum full gain (catalog specification)
26.5 dB
Diplex filter loss (2 x 0.6)
1.2 dB
Equalizer loss
2.0 dB
Reserve gain
Total loss
Usable amplifier gain
5.2 dB
21.3 dB
The distribution system will be designed with an amplifier gain of 21.3 dB in the
forward path (the return path needs less gain, but the calculation is similar.
Amplifier Cascade
Amplifier cascade is the number of amplifiers connected in a series configuration (one
after another) in a trunk system. Since each amplifier contributes some noise to the
system, there is a practical limit to the maximum number of units that can be cascaded.
To determine the maximum cascade, several factors must be considered such as:
~
Output level of the amplifiers
~
System bandpass
~
The amount of cable loss between each amplifier
. In general, the more amplifiers in a cascade, the lower each amplifier's output level
should be.
The following calculations show how to determine the minimum number of
amplifiers needed to compensate for the signal loss of the longest cable run of a
system. The amplifier cascade to use for designing the system (the design cascade) is
then found by doubling this calculated minimum value (to provide room for
expansion).
80
5
Amplifier Cascade
Longest cable run (estimated)
Cable loss (O.5-inch cable @ 300MHz)
1880 feet x 1.31 dBIlOO feet
1880
feet
24.6 dB
Required cascade over longest run
Cable loss / Usable gain
24.6 dB / 21.3 dB
1.1 amplifiers
Minimum required cascade
2 amplifiers
Design cascade
2 x Minimum required cascade
4 amplifiers
Amplifier Output Level
Knowing the design cascade allows the calculation of the maximum amplifier output
level permitted in that cascade. The amplifier's rated output level (the highest signal
level it can deliver without exceeding distortion specifications) should be reduced by 3
dB for each doubling of the number of amplifiers in cascade.
Cascaded Amplifier OutP.llt Level
5 c = 50 - 1000g(N)
where 5c = maximum permitted output level of each amplifier in cascade (dBm V)
50= rated output level of one amplifier (dBmV)
N = number of amplifiers in cascade
For example, using amplifiers whose rated output level is +48.0 dBmV, the
permitted output level in a cascade of four units can be no more than 42 dBm V.
5 c = 48 - 10l0g(4)
= 48 - 6
= 42 dBmV
Based on the above parameters and the design discussion, the sample system has
the following design characteristics (table 5-3).
System Noise
Random noise occurs over a wide bandwidth and is associated with a network's
amplifiers. Typical noise figures for amplifiers range from 7 to 11 dB. A previous
section showed how the noise for a cascaded system can be calculated.
Noise can also be injected from outside sources such as radio transmitters and
electrical motors. This type of noise is usually associated with specific frequencies or a
particular frequency range.
81
5
Factors Affecting System Design
Table 5-3.
Sample System Design Characteristics
Single cable network, midsplit implemented
Cable
Diameter
0.500 inches
Loss per 100 feet at 300 MHz
1.31 dB
Amplifier
21.3
Gain
dB
Output level
+36.0
dBmV
Input level
+14.7
dBmV
+6.0
dBmV
Minimum outlet level
across specified spectrum ( ± 1.5 dB)
When properly installed, the system should provide a noise floor 40 dB below the
video carrier level. This means that the level of noise in the system should be 40 dB or
more below the level of the nominal video carrier. When the noise floor is higher, an
amplifier might be contributing more than its share of noise, or a portion of the
distribution system might not be properly terminated.
Each amplifier contributes some noise to the system depending on its noise figure.
To determine if an amplifier is more noisy than it ought to be, the carrier-to-noise ratio
of the system can be measured at various points to find the faulty device. This
procedure is best done in a logical progression, either beginning at the headend or at
the farthest point from the headend, and moving toward the opposite end until the
fault has been isolated.
A specific type of noise found in many electrical systems is hum, which is noise at
the ac power line frequency (60 Hz). The suggested limit for the carrier-to-hum (C/H)
ratio for a full system is 40 dB or more. Cascading amplifiers in a system causes the
system's C/H ratio to decrease by 6 dB for every doubling of the number of amplifiers.
The following example shows how the C/H ratio of a system is calculated, given a C/H
ratio for each amplifier of -70 dB and a cascade of 20 amplifiers.
Carrier-to-Hum Ratio
Suggested limit
C/Hc = C/HO + 20l0g(N)
where C/Hc = C/H ratio of the cascaded system
C/HO = C/H ratio of one amplifier
N = the number of amplifiers in cascade
82
> 40 dB
5
For example,
C/Hc = -70 + 20log(20)
= -70 + 26
= -44 dB
Intermodulation Distortion
Intermodulation distortion (IMD) occurs when desired signals on the system interact
to produce undesired signals. The primary causes are amplifiers operating at improper
levels and defective amplifier stages. Either of these situations might cause the unit to
operate in a non-linear fashion, which can create IMD.
If Fl, F2, and F3 represent frequencies of carrier signals on the system,
intermodulation distortion can occur at the following second-order beat frequencies:
~
Fl ± F2
~
Fl ± F3
~
F2 ± F3
Interference can also occur at the following third-order beat frequencies:
~
Fl ± F2 ± F3
The recommended limit on second-order inter modulation distortion is 60 dB
below video carrier level. In a cascaded system, second-order beat frequency
components increase by 6 dB for every doubling of the number of amplifiers in
cascade. However, these same interference Signals decrease by 6 dB for every 3-dB drop
in amplifier output level. As a result, lowering amplifier output level by 3 dB every
time the cascade is doubled maintains the same second-order distortion specification.
This practice coincides with the recommendation of dropping amplifier output levels
when cascading amplifiers to keep system noise levels within specifications.
Another measure of inter modulation distortion on the system is composite triple
beat (CrB). CTB is caused by the combination of all possible third-order beat
frequencies that occur on the system. Its source is nonlinear effects of system
components on transmitted carrier signals. For example, if a system has five carrier
signals at frequencies Fl, F2, F3, F4, and FS, the possible triple beats on the system are
the follOWing.
~
Fl ± F2 ± F3
Fl ± F2 ± F4
~
Fl ± F2 ± FS
~
F2 ± F3 ± F4
~
F2 ± F3 ± FS
F3 ± F4 ± FS
~
~
83
5
Factors Affecting System Design
The combination of all frequencies represented by these triple beat frequencies is the
composite triple beat. The recommended limit for CTB is 51 dB or more below video
carrier level. CTB increases by 6 dB with every doubling of the number of amplifiers in
cascade.
Composite Triple Beat
CTBc = CTBO
+ 20l0g (N)
where CTBc = CTB ratio of a cascaded system
CTBO = CTB ratio of one amplifier
N
=
number of amplifiers in cascade
The System Level Graph
A system level graph summarizes many of the design calculations for a broadband
network in a graphical manner. This section discusses how such a graph can be made,
and describes the sample shown in figure 5-7.
This graph shows the relationship between amplifier specifications (noise figure,
rated output level) and system specifications (noise floor, carrier-to-noise ratio,
distortion level, and amplifier cascade). It also shows the operating window inside
which the amplifier's input and output levels must reside.
A graph like this can be used to check that amplifier operating levels are within
acceptable ranges and to show what signal margins exist between the operating values
and the limits. It can be generated with the following procedure.
1.
Starting with a typical amplifier's noise figure (e.g., 8 dB), compute the equivalent
noise input (ENI) for the system for cascades of 1, 2, 4, 8, and 16 amplifiers. Plot the
resulting values of ENI versus cascade.
a.
For a 75-ohm system with a 4-MHz bandwidth, the noise floor is -59 dBm V.
b.
Adding a single amplifier with an 8 dB noise figure gives an ENI of -51 dBmV.
c.
Each doubling of the number of amplifiers in cascade raises the ENI by 3 dB.
For example, with two amplifiers, the ENI moves up to -48 dBmV; with four
amplifiers, it becomes -45 dBm V.
2.
Select a signal-to-noise ratio for the system (e.g., 45 dB). Add this value to the plot
of ENI versus cascade to obtain the plot of minimum acceptable input level.
3.
Compute the maximum allowable output level for each value of cascade.
84
a.
Find the rated output of a single amplifier for the desired distortion level from
its specifications. This example uses a device with a rated output of + 48 dBm V
with a CTB of -57 dB.
b.
Plot output level versus cascade; the output level drops 3 dB with every
doubling of the cascade to maintain the CTB specification.
5
4.
The operating window for the system using these amplifiers is the area above the
minimum input line and below the maximum output line.
5.
Compute the maximum allowable gain for each cascade value by subtracting input
from output. This value drops by 6 dB with each doubling of the cascade.
dB
+54 ~~------r--------.--------r--------r--------'
~--------r--------r------~~~----_+--------1
MAXIMUM ALLOWABLE
GAIN
+30 ~--------r--------r--------r-----~~~~----1
+24
-r--------r-------_+--------+--------+------~~
dBmV +48 ~=-------~------~--------~-------+------~
+42
+36
MAXIMUM AMPLIFIER OUTPUT
FOR -57 dB COMPOSITE
TRIPLE BEAT (CTB)
+30
+24
+18
+12
:;;....____-1
+6
MINIMUM AMPLIFIER INPUT
FOR +45 dB
SIGNAL·TO-NOISE RATIO
-3
-6
-36
+-----t----+---+--+-----t----::;:::ooo/
-t---------r--------r---;~:::oo'.....:;;....----_+----__i EQU IVALENT NOISE INPUT
-r-------r-----~~~------+_----_+----__i(BASEDONAN8dB
-48 +---~
__t-==:::.----+--------f_---_+----_l
NOISE FIGURE)
-51
16
32
AMPLIFIER CASCADE
Figure 5-7. A Typical System Level Graph
Using the System Level Graph
A graph developed with the preceding procedure can be used in several different
ways. This section provides some ideas.
When designing the network layout, cable lengths cannot always be the desired
length for amplifiers to be used at the exact design values. Once amplifier placement is
85
5
Factors Affecting System Design
estimated, the loss for which it must compensate can be calculated, and the input level
it sees can be found. Plotting this input level on the system level graph provides an
estimate of its margin above the minimum input level.
If the calculated input level is below the minimum input level line, the system's
SIN specification will not be met at that point in the distribution system, and the
design must be modified to provide a higher signal level to this amplifier (e.g., shorten
the distance between it and the previous amplifier).
After an adequate input level is delivered to the amplifier, its output level can be
determined by adding the amplifier's gain to its input level. The output level should be
below the maximum amplifier output level line on the graph; the difference between
it and the maximum output value is the margin; 3 dB is commonly used as a minimum
margin value.
If the output level is beyond the maximum amplifier output line, the design must
be changed to avoid excessive distortion. Possible corrective measures include the
following.
~
Decreasing the input signal level feeding the amplifier input level by introducing
attenuators or by inserting more cable between it and the previous unit.
~
Lowering amplifier gain (but not too far, since trunk amplifiers operate best near
their maximum level).
~
Using fewer amplifiers in cascade.
Reliability and Redundancy
Properly laid out systems cannot have a total system failure unless the headend is
disabled. Broadband networks are reliable because their design is simple and their
components are sturdy. The basic design of a network is very straightforward. The
components have been proven in many harsh environments over years of service. For
systems that require maximum reliability, redundant components and trunks can be
installed to be activated whenever a failure occurs. These backup systems can be
switched into the network either manually or automatically by equipment that
monitors network performance. This section discusses network reliability and
redundancy aspects that can be considered during the design of a broadband network.
The most common causes of system failure in CATV and broadband networks are
physical breaks in the cable and faulty connections. Other major causes of failure
include power interruptions, processing equipment failures, and interface device
failures. Networks inside buildings usually have fewer physical breakdowns but
experience the same low number of electronic failures as their outdoor counterparts.
The following paragraphs briefly describe some methods and systems that can
increase the probability of the network being operational, including
~
Installing backup network trunks and amplifiers
~
Making regular system performance tests
~
Using automatic monitoring systems and backup switching systems
86
5
Broadband Component Quality
The high quality of broadband components is evident from their mean-time-betweenfailure (MTBF) specifications. The life expectancy of properly installed passive
components is estimated to be 30 to 40 years. The MTBF specification of amplifiers,
data translators, and video processors is around 18 years.
The reliability specifications of any customer equipment to be attached to the
network should be checked, and the appropriate selection should be made based on
the network's requirements. Most failures occurring on broadband networks have
been related to customer equipment rather than the backbone cable and signal
distribution system.
Periodic Maintenance
Passive components require only annual physical inspection and Signal tests to insure
proper functioning. Once active components are adjusted for proper input voltage,
signal level, and equalization, no further adjustment should be necessary. It is
recommended to check signal levels at the amplifiers once a year. Usually, no work or
adjustment is done, but these yearly checks can provide warnings of possible
component failures. CATV alignment and maintenance procedures are much different
and are not discussed in this overview.
Equipment Replacement and Repair
Once a device breaks, the failure must be located, the defective unit replaced, and
signal levels readjusted when necessary. When a backup device is not available, then
replacement of faulty subassemblies should be employed. Enough key components
should be kept on hand for this purpose. On-site repair of broken modules is difficult
without the proper test equipment and trained personnel. It is often easier and less
expensive to send the modules to the distributor or manufacturer for repair.
Redundant Trunks and Components
A break in the cable can usually be quickly repaired at a fraction of the cost of
designing and installing a redundant trunk system. When such a short interruption to
service cannot be tolerated, a backup trunk cable can be installed. Ideally the primary
and secondary trunks would follow different routes. This proVides greater reliability
but costs more. The secondary cable must be run far enough away from the primary
cable to minimize the probability of damage to both cables by a single accident.
Remote-controlled coaxial switches can switch from a primary system to a
secondary system if a failure occurs. The secondary system must be checked regularly
to ensure that it is working properly. Checking can be done manually or automatically
depending on the available test equipment.
If the distribution network is fully redundant but the connected devices such as
television monitors, computers, terminals, and cameras are not, then no real
redundancy exists. Emergency power sources for key items of equipment should be
considered.
Redundancy concerns can be carried too far. For example, a major network not
using any form of system redundancy except backup data translators has operated for
over five years without a single amplifier, cable, or passive component failure.
87
5
Factors Affecting System Design
Status Monitoring Systems
One approach to controlling backup components and cable trunks is with status
monitoring systems. Status monitoring systems that are currently available consist of a
computer-based control center located at the headend, and transponders within each
amplifier and in key trunk arteries. Each amplifier housing also contains backup
amplifier modules. The control unit includes a video display terminal, microprocessor,
message modem, and a printer. Such a system can be installed directly into a network,
providing the network was designed with status monitoring in mind and uses
compatible components.
The control unit interrogates each transponder throughout the day and evaluates
the replies received. Any failure triggers the following actions.
~
The defective area and component are reported on a status board and on the
control center's monitor.
~
The broken module is bypassed and replaced with a standby module. This
switching can be done either automatically or on command from the control
center. It prevents the loss of communications and allows repairs to be done at a
convenient time.
Hardware for this equipment is modular, allowing many existing networks to be
upgraded to include status monitoring.
Technical Control Systems
Technical control systems are microprocessor-based control systems that use automated
test equipment for system analysis, fault isolation, carrier allocation, and remote feeder
switching. A failure in the trunk system can be detected by analyzing responses from
control units placed throughout the facility. Once a failure is detected services can be
switched from one cable to another, restoring the communication link.
These units differ from status monitoring systems in that they must be custombuilt for the network. Status monitoring systems and compatible components are
readily available, and can be installed directly into many existing networks. Technical
control stations provide remote monitoring and control facilities for any network that
has no integrated status monitoring system.
Test equipment found in technical control systems includes programmable
spectrum analyzers, signal generators, and sweep oscillators. All these instruments are
linked together on a standard IEEE-488 interface bus. A computer program could use a
mapping scheme that would show any change in the system's performance based on
the original settings and measurements. Detecting an excessive change could trigger
alarms and automatic switches if desired.
Total redundancy for two-way systems is not now available, although
development seems to be heading in that direction. Both status monitoring systems
and technical control stations provide the network manager with valuable information
about the network and the ability to control its operation.
88
5
Headend Design
The headend is the origin of all RF signals sent to receiving devices connected to the
network, and the destination of all signals generated by all transmitting devices on the
network. It is the heart of the network, and must be designed carefully so that new
services can be added without having to recalculate signal levels and realign
amplifiers.
It is a good rule to design the headend and its associated components before
designing the remainder of the network. All the passive loss associated with
combiners, splitters, and directional couplers can be computed. These computed values
then are used to find the signal strength available for distribution throughout the
network.
The folloWing paragraphs describe the equipment found in a typical headend.
Some aspects of large CATV headends are discussed briefly, since they could be
applicable to broadband local area networks in the near future.
System Diagrams
Diagrams of an RF network are provided to show how various needs can be satisfied
by a broadband network. Figure 5-8 illustrates a typical headend that can support data
communications and one local origination television channel. Signal levels in dBm V
are shown at several points. This example is only a guideline but is typical of many
headends in use today.
Figure 5-9 shows a typical distribution scheme through a single floor complex.
Each office can have its own RF outlet. In addition, the locations of computers,
terminals, receivers, and cameras are included.
Standard Headend
A standard headend must be able to support the large bandwidth necessary for a
broadband network. Many designers miss this point when designing smaller systems
or test bed systems that will be expanded later to support more services. The deSigner
should always plan for the eventual loading of the system. Economizing at the outset
by planning for a small network often produces systems that cannot support all the
services needed in the near future.
The following paragraphs describe the salient features of figure 5-10, which is a
detailed view of a head end.
89
5
Factors Affecting System Design
-16 dB
-16 dB
+42
EIGHT-WAY
COMBINER/
SPLITTER
DATA TRANSLATOR
FROM OTHER
~~---. HEADEND
SPLITTER
+33.9
TAPS~
POWER COUPLER
+28.8
1 dB OF CABLE LOSS
AT 300 MHz
POWER
~--------~SUPPLY
+28.8
1 dB OF CABLE LOSS
AT 300 MHz
(See Figure 5-9 for Continuation)
Figure 5-8. Headend Configuration Showing Signal Levels
90
COMPONENTS
5
~
[gJ
="d
y
n
dl
EJ
fr\
'e'
1"[J
u
~
,,
=:/\
HEADEND
l~.7
~
CPu
,, £]
17
...
~
...
~
,, £]
17
£]
dJ
~
14
,,
I~
-
,dJ
,
TYPICAL
•
DROP CABLE
~
SYMBOLS
TERMINAL
RECEIVER
TAP (4 PORT)
Figure 5-9. Typical Cable Distribution Scheme
91
dJ
CAMERA ~
1001
TYPICAL . .
OUTLET
0
5
Factors Affecting System Design
ANTENNAS
VHF/UHF CH 11
CH36 CH54
CHANNEL PROCESSOR
CHANNEL IN
FORWARD AND RETURN TEST POINT
DIRECTIONAL
COUPLER
FUTURE EXPANSION
EXPANSION TIE
REVERSE TRUNK TEST POINTS
TO THE SYSTEM TRUNKS
Figure 5-10. Detailed Headend Configuration
First, the top of the drawing shows two eight-way combiners. These combiners connect
equipment for the various services to the forward and return paths of the system. The
example system shown can connect four such combiners in the forward path and two
such combiners in the return path. This configuration allows several different devices
to be attached to the system at the headend, such as data translators for different data
systems, and television signal translators. The number of input connections provided
by the combiners is usually proportional to the bandwidth of the path. Using
combiners allows the introduction of more services in the future without having to
reconfigure the network or to adjust signal levels.
92
5
Next, the forward and return frequency bands are processed by a system filter. The
type of filter used depends on the frequency split chosen.
As with any well-designed communications network, an accessible test point is
required for test equipment such as a spectrum analyzer.
Two directional couplers are placed back-to-back to support non translated devices.
Non-translated devices are simpler, require no frequency translator, and are less
versatile than translated devices.
~
Non-translated devices are usually restricted to point-to-point or multipoint
communications between the headend and remote points.
~
Non-translated devices connected to different branches cannot communicate
directly with each other.
~
A special headend extension is required if two-way communications to a new
location is required; this can cause wiring and signal level problems.
Non-translated systems are not covered further in this book.
A terminated directional coupler provides an expansion point for additional
headend devices, or a point for trunk expansion.
The amplifier close to the headend provides needed return path signal gain and
return band frequency equalization. At least 75% of all systems require this amplifier
to obtain proper signal level at the translator and television processor input
connections. This amplifier also brings the return path signals to the input of the data
translators at levels that are as equal as possible.
The user should be able to measure the return path signal from each trunk
independently. This is accomplished by installing taps in each trunk to monitor the
return path's signal. These tap ports are connected to a test patch panel where signals
from each trunk can be analyzed separately. Both automatic and manual fault isolation
are made easier with this configuration.
The broadband system's power supply and power combiner can be located almost
anywhere in the system, although most are placed at the headend. Redundant power
supplies and automatic power supply switching units are available if needed.
Additional protective devices should be used at the headend. A surge protector and
an RF filter should be installed on the incoming ac power line. These circuits can
protect the equipment from lightning surges, noise, and ground loops that cause hum
in the video.
The standard ac voltage level for two-way system design has been 60 Volts. ThirtyVolt systems are less able to supply the proper voltage level to the amplifiers because of
voltage drop in the cable. The CATV community has converted most of their systems
to 60-Volt operation.
The headend described here is typical of many systems. The flexibility of
broadband allows many different system configurations. Just as no two electrical
distribution systems are designed exactly the same, no two broadband systems are
likely to be identical.
93
5
Factors Affecting System Design
Large CATV Multichannel Headends
Extending the bandpass of coaxial networks to 400 and 450 MHz and increasing the
number of available television channels has created unique operating situations. A
400-MHz, 52-channel television system provides significantly more bandwidth than
existing 300-MHz 35-channel networks. Using such systems creates new technical
problems in deSign, equipment, and maintenance. These problems apply mainly to
CATV networks, but are discussed here since broadband local area networks are
planned to provide multiple services and channels.
Composite Triple Beat Distortion
Recall from the previous discussion that composite triple beat (CTB) distortion is the
combination of all possible third-order beat frequencies that occur on the system. In a
300 MHz, 35-channel system, analysis of all possible F1 ± F2 ± F3 (where Fl, F2, and
F3 are RF carrier frequencies) triple beats shows that the greatest number of beats
falling on one channel is 334 on channel 11 or 12. Because there are so many frequency
components so close together, they tend to add on a power basis and appear as
narrowband noise on a spectrum analyzer. This impairment can be seen visually on a
television picture when the ratio of visual carrier level to CTB level is 51 dB or less.
In a 52-channel system using standard frequency assignments, the greatest
number of beats is 842 for channels 0 and P. To maintain the same -51 dB CTB as in a
35-channel system, all amplifier output levels would have to be lowered by 5 to 6 dB.
The combination of lower amplifier gain and higher cable losses at 400 MHz can
require the use of more amplifiers and increase network cost. However, by using nonstandard frequency assignments, CTB problems can be alleviated. The harmonically
related carrier and interval related carrier techniques can be used to control carrier
phasing at the headend, and are discussed in the following sections.
Harmonically Related Carrier
One technique used to control large multichannel headends is called harmonically
related carrier (HRC). HRC headends use a master 6 MHz generator to phase lock all
channels together to retain exactly the same frequency spacing from each other. This
generator is called the comb generator. With its use, each channel must be shifted by
-1.25 MHz, except channels 5 and 6 that must be shifted by + 0.75 MHz, from their
original frequency assignments. This shift results in an additional channel between
channels 4 and 5. Implementing the HRC frequency relationship provides an
additional 3 to 4 dB of output level and reduces effects caused by CTB. A 44-46 dB CTB
would be used as the limiting distortion factor.
The 6-MHz oscillator, however, creates interference problems for frequencies with
strong broadcast stations. In the past this problem was solved witb phase locking at the
channel processor. Phase locking cannot be used in HRC headends because of the
frequency shift caused by the comb generator. A similar conflict also arises with
frequencies used for aircraft navigation and communication. In some instances,
channels on the cable system might have to be abandoned to satisfy offset
requirements of the FCC and FAA.
94
5
Interval Related Carrier
Another technique used to control multi-channel headends is called Interval Related
Carrier (IRC). With an IRC headend, all channels except 4 and 5 operate on the
standard NCTA-defined frequencies. (The National Cable Television Association
(NCTA) has defined standard frequencies for conveying broadcast television signals
over cable systems.)
Channels 4 and 5 are shifted to fall within the spacings of other channels. All but
these two channels can operate coherently (on the same frequen-0' as the broadcasted
signal) with off-air television channels as part of an IRC cable system. The result is that
an IRC system has fewer channels affected by the ingress of off-air signals. An IRC
system still allows the same 3-4 dB increase in amplifier output level as an HRC
system.
A table in appendix D provides a listing of frequency assignments for each
channel for both HRC and IRC systems.
Drawing Standards
Drawings are important tools when building a local area network. Without them,
troubleshooting can be extremely difficult and expansion can be a troublesome chore
instead of the simple matter that it should be. The following list includes some of the
items that should appear on every broadband system drawing to ensure consistency
throughout the industry.
1. Symbols used and the part number(s) of the associated equipment. Appendix C
provides a recommended symbol chart.
2. System design frequencies, both forward and return.
3. Cables selected and the loss associated with the forward design frequency and the
return design frequency.
4. Details of the headend including the calculated output levels of the television
processors and data translators, and any equipment limitations regarding power
levels.
5. Trunk routing throughout the facility.
6. Calculated length and associated loss for the forward and return directions of each
segment of cable.
7. Suggested attenuator values.
8. Suggested equalizers and other necessary equipment for each amplifier.
9. Details regarding ac power distribution and blocking at the headend and at all
amplifiers.
95
5
Factors Affecting System Design
10. Distribution system parameters such as tap values, number of ports, and calculated
signal levels at each tap.
11. Locations of all manholes, raceways, conduits, trays, and similar structures.
12. Locations of any extra cables for system redundancy or for spares.
13. Suggested trunk and distribution amplifier levels for both the forward and return
paths.
14. Minimum outlet and tap levels at the design frequency.
15. A note to terminate all unused outlets when not in use.
16. Recommended alignment method: flat output, flat input, or flat midspan. The flat
output method is considered the standard for broadband LANs. All three methods
are described in chapter 6.
17. Notes describing any special considerations or special areas of design.
18. A bill of materials.
19. The designer's name, address, and telephone number.
20. A letter that explains the design objectives and parameters, including any special
areas needing clarification.
System Specifications
Table 5-4 summarizes the specifications of a typical transparent system. All signal
levels are referred to the 6-MHz video carrier level. Each manufacturer should be
consulted to ensure that the signal levels are consistent with the requirements of any
interface devices to be used on the network.
96
5
Table 5-4.
Broadband System Specifications
Carrier-to-noise ratio, forward and return
>43
dB
Peak carrier to p-p hum ratio
< 3
%
Peak-to-valley response, full system
Peak-to-valley response, n-amplifier cascade
± 3.5 dB
1
+ n/lO
dB
± 1 dB
Peak-to-valley response, any 6-MHz channel
Carrier-to-second order beat ratio
>-60
dB
Carrier-to-composite triple beat ratio
>-51
dB
Outlet receive level, nominal
+ 6 dBmV ± 3.5 dB
Tap receive level, min. {assuming a 2 dB drop cable loss)
+ 8 dBmV ± 3.5 dB
Radiation (measured with a tuned dipole)
5 MHz to 54 MHz (measured at 10 feet)
15 f.LV/m
54 MHz to 216 MHz (measured at 100 feet)
20 f.LV/m
216 MHz to 400 MHz (measured at 100 feet)
15 f.L V / m
Forward path loss, nominal
50
dB
Return path loss, nominal
46
dB
Headend processor transmit level, nominal
+56
dBmV
Headend processor receive level, nominal
+10
dBmV ± 3.5 dB
Data translator transmit/receive levels
Vendor-specified
Outlet transmit level
+56
dBmV
Meeting the following additional specifications helps to ensure proper network
operation.
1.
Refer all signal levels to the 6-MHz video carrier level.
2.
All headend modulators and channel processors should produce an output level of
+ 56 dBmVat the input of the forward path combiners.
These devices should receive an input level of + 10 dBm V from the return
path combiners, within a tolerance of ± 3.5 dB.
3.
In office environments, maintain an outlet level of +6 dBmV ±3.5 dB. The tap
level is slightly higher than the outlet level to account for drop cable loss).
4.
In industrial environments, maintain an outlet level of + 9 dBm V ± 3.5 dB. This
allows the use of a two-way splitter to provide an on-line test point at every
network connection. This test point can be used to connect audio modems for
maintenance communications (useful when working in remote areas of a large
facility).
97
5
Factors Affecting System Design
5.
The RF signal level difference between any two adjacent channels should be less
than 2 dB.
6.
Isolation between any two outlets in the system should be at least 28 dB over the
total range of 5 MHz to 400 MHz.
7.
For 20 data subchannels within a 6 MHz assignment (300 kHz each), the individual
carrier levels should be 13 dB below the peak video level (assuming all carrier
signals on).
8.
For 56 data subchannels within a 6 MHz assignment (96 kHz each), the individual
carrier levels should be 18 dB below the peak video level (assuming all carrier
signals on).
9.
Any unused outlet should be terminated with a 75-ohm resistor to minimize
reflections. This can be accomplished either by attaching a terminating connector
to the outlet, or by using self-terminating outlets.
A distribution system including tap values, passive splitter values, and the number
and type of connectors, can be designed based on these recommended specifications.
Summary
This chapter has provided information relating to the design of broadband networks.
The reader should have gained an understanding of the design process from this
material. This understanding should encourage effective communication between the
network designer and the network user. The network user should have also gained an
appreciation for the complexity of network design, and an awareness of some of the
compromises and tradeoffs that require experience and judgment. The next chapter
provides details on aligning the completed network so it can provide its specified
services.
98
•
6
System Alignment
Introduction
Once a broadband communications system has been installed, it must be aligned and
certified before.it can be used. If the network's performance is not certified, significant
operating problems could occur as communication services are added to it. This
chapter provides information to aid the alignment and troubleshooting of both single
and dual cable systems. This discussion assumes that the system has been designed
properly for two-way communications. The reader should have an understanding of
broadband communications as provided by the pre<:eding chapters of this book.
The Need for Alignment
There are two basic reasons for aligning a network.
~
To provide a flat signal response across the network's bandpass.
~
To provide equal signal levels at each outlet.
The alignment process consists of adjusting the gain and equalizer controls of each
amplifier to achieve consistent and desired signal levels throughout the system. Twoway trunk amplifiers are typically adjusted to produce 18 to 22 dB of gain at the
system's highest design frequency (300 or 400 MHz), and correspondingly lower gain
at lower frequencies. Two-way line extender amplifiers provide higher gain at the cost
of greater noise. Amplifier selection depends on system design factors covered in
chapter five.
Figure 6-1 shows schematic illustrations of broadband amplifiers for single cable
subsplit and midsplit systems and for dual cable systems. These drawings show the
relationship of test points, pads, adjustable equalizers, and gain blocks; this
information is useful during alignment.
100
6
168-300/400 - - - .
FIL TER
1--+---'--+-
(aJ Midsplit Amplifier
54-300/400
------
(bJ Subsplit Amplifier
54-300/400 MHZ ------
(c.) One-Way Amplifier
Figure 6-1. Trunk Amplifier Configurations
Before aligning the system, it helps to become familiar with the insertion loss of
the broadband components used and the expected attenuation characteristics,
including tilt, of the coaxial cable(s). Also, it is important to examine the system design
drawings, to ensure that they reflect the current installed configuration, and to note
any changes made to the design during installation.
101
6
System Alignment
Test Equipment Required
This section discusses the test equipment required to align and maintain the system.
About $15,000 (at 1983 prices) is required to purchase the appropriate test equipment,
including the following items.
~
RF spectrum analyzer
~
RF sweep generator
~
Field strength meter
~
Multimeter
~
Cable reflectometer
~
RF radiation monitor
Appendix F lists some currently available devices that can be used in these
applications.
The RF spectrum analyzer provides a graphic display of the frequency spectrum and
is useful in system alignment, troubleshooting, carrier analysis, carrier-to-noise
measurement, intermodulation distortion measurement, and many other performance
tests. The spectrum analyzer should have a 75-ohm impedance, and show signal levels
in dBm V from 5 to 500 MHz.
An RF sweep generator is a signal source. It should have the following
characteristics.
~
RF output signals from 5 to 500 MHz with a tunable sweep function.
~
A fixed-frequency output signal at any frequency in the bandpass range. Thebandwidth of this signal should be adjustable.
~
Adjustable output signal amplitude.
~
75-ohm impedance.
~
Calibrated for dBm V.
The field strength meter (FSM) is a popular test item. This is a tuned RF voltmeter used
for determining the amplitude of a signal at a specific frequency. This unit is easy to
operate and can be used for measuring performance, aligning amplifiers (once they are
equalized), verifying signals, and troubleshooting. It also is calibrated in dBm V and
some units can read signals up to 800 MHz.
A multimeter is used primarily for checking power supplies and evaluating ground
loops. Such units should be portable and able to select several voltage and resistance
ranges.
The cable refleetometer is used in locating cable faults caused by physical breaks in a
given span of cable or by reflections caused by bends or kinks. These instruments can
indicate the location of the fault to within a few inches. Cable system troubleshooting
time is minimized by the combined use of a reflectometer and accurate, scaled
drawings of the cable layout.
A sniffer or a bloodhound are instruments that measure RF radiation. These
instruments pinpoint areas where the system radiates RF energy, either through a poor
connection or a damaged cable. When radiation occurs, one can also assume that the
102
6
system is more susceptible to signal ingress. Signal ingress can hinder the proper
operation of the network. It is important to ensure that the system provides maximum
isolation from external signals, including electromagnetic interference (EMI) and radio
frequency interference (RFI).
There are many other types of specialized test equipment that the CATV industry
uses. These include devices to do the following functions.
~
Status monitoring
~
Feeder disconnection
~
Automatic fault isolation
~
Remote interrogation of pre-calibrated RF detectors
These devices are not covered in this overview.
Documentation
Documenting the process of cable alignment is important. A record should include at
least the follOWing items.
~
The tests made.
~
The equipment used.
~
The settings on the test equipment.
~
The signal levels measured.
~
Miscellaneous items including changes made to the system during alignment, and
special notes about any unusual situations.
Producing good documentation will simplify the diagnosis of system problems later.
Proper documentation provides a normal pattern for comparison with the abnormal
pattern encountered during a breakdown. Noting simple items, such as test equipment
settings and readings during alignment, helps maintenance personnel understand the
task at hand, and shows what signal levels are expected at various points in the system.
Good records allow easy replication of alignment conditions, which can provide a good
starting point for troubleshooting.
Coaxial Cable Certification
Separate tests can be done on the cable and the distribution system to help ensure
proper performance, and to identify possible problem areas before the network is used
for data transmission.
The structural return loss test measures cable loss before and after installation.
~
Sweep testing the cable before installation minimizes the chances of installing
defective cable, which could be costly to replace.
~
Checking the cable after installation points out any frequencies having excessive
loss due to handling and attaching connectors.
103
6
System Alignment
Small discontinuities in the cable can cause reflections or standing waves. These
periodic disturbances can make a very effective filter, resonating at the frequency
whose wavelength is twice the spatial period of the discontinuities. This can produce a
pronounced spike in the return loss at that frequency. Each individual discontinuity
might have a return loss of 60 dB, but the accumulated effect could be enough to cause
a narrow-band dip in the transmission characteristic of the cable. When the dip occurs
at a desired carrier frequency, the additional attenuation can degrade that channel's
response. Sweep testing the cable before and after installation can detect structural
problems when poor return loss readings are obtained.
The cable sweeping or component performance test can be performed during or after
installation. A sweep generator at the headend and a portable sweep receiver make up
the system. Measurements can be made at any point in the network. This test analyzes
and plots system attenuation, cable exponential characteristics, and spectrum integrity
through the passive and active components. The performance of the return path in
bidirectional networks can be verified in this way. With a sweep test one can compare
the characteristics of the installed system to specifications of the designed system.
Alignment Methods
There are three common methods used to align a broadband system.
~
Flat amplifier output
~
Flat amplifier input
~
Flat midspan
Each approach is briefly described in this section. The flat amplifier output method
applied to a single cable, midsplit format system is covered in more detail in the
following section.
All discussions pertain to MGC amplifiers. Read the manufacturer's manuals on
the amplifiers being used, and pay close attention to the sections covering
configuration and alignment. These manuals give suggestions on alignment
procedures for AGC amplifiers, and advice on proper output levels.
Since cable systems have been designed around the transportation of television
signals, the amplitude of television signals are to be used when aligning a system. A
system aligned to convey only data carriers can be overdriven and caused to operate in
a nonlinear mode when video carriers are introduced. These effects can render the
network unusable for television signal distribution. Therefore, it is always
recommended to design and align broadband systems to convey television signals.
Flat Amplifier Output
In the flat amplifier output method, each amplifier is adjusted to give equal amplitude for
signals of all frequencies at its output connection. As a result, each amplifier
compensates for the tilt of the cable between it and the preceding amplifier (figure 62(a». This alignment technique is recommended for most broadband systems, since it
can be done by one person. Flat output alignment results in a flat response across the
spectrum at taps immediately following each amplifier, with increasing variation (tilt)
in signal response as the distance from the amplifier's output point increases.
104
6
LEVEL REFERENCE
(a) Flat A mplifier Output Alignment
(b) Flat Amplifier Input Alignment
(c) Flat Midspan Alignment
60 dBmV
LEVEL REFERENCE
AMPLITUDE
OdBmV
5 MHz
300/400 MHz
FREQUENCY
Figure 6-2. Amplifier Alignment Techniques
Flat Amplifier Input
In the flat amplifier input method, each amplifier is adjusted to give equal amplitude for
signals of all frequencies at the input of the following amplifier in cascade. Each
105
6
System Alignment
amplifier compensates for the tilt of the cable between its output point and the input
point of the next amplifier (figure 6-2(b». This alignment method requires two people,
one at the amplifier being adjusted and one at the measurement point (the following
amplifier in cascade). Flat input alignment results in a flat response across the
spectrum at taps immediately preceding each amplifier, with increasing tilt in signal
response as the distance from the amplifier's input point increases.
Flat Midspan
In the flat midspan method, each amplifier is adjusted to give equal amplitude for signals
of all frequencies at a point midway between it and the following amplifier in cascade
(figure 6-2(c». This method also requires two people: one at the amplifier being
adjusted and one at the measurement point (a tap at the midpoint of the cable between
the amplifier being adjusted and the following amplifier). Flat midspan alignment
results in a flat response across the spectrum at taps in the middle of each cable span
between two amplifiers, with increasing tilt as the distance to either amplifier
decreases.
Flat midspan is the best alignment technique because it provides a flat signal
response at more outlets than the other techniques provide. The flat output method is
used more often because it is easier to do. An alternative combines aspects of both the
flat output and midspan procedures (called the modified flat midspan method).
1.
First, the system is aligned using the flat output method.
2.
Then, the signal is monitored at the midpoint between any two amplifiers or
between any two multi-tap cable runs (which approximates the location of a
typical outlet). The slope control of only the amplifier nearest the headend is.
adjusted to provide a flat response at the monitored point. If the amplifier has no
slope control, the equalizer control is adjusted.
The modified flat midspan method provides flat response near the midpoints of all
pairs of amplifiers. Unless all the cable spans are identical in length, some variation
from flat response occurs at these midpoints.
Summary
The flat output method is recommended for all trunking situations (that is, for cable
spans that have no taps between amplifiers). The flat output method is also
recommended for aligning the interior or feeder (as opposed to trunk) distribution
system. It could be enhanced by adjusting the slope of the the first amplifier in the
system, as described for the modified flat midspan method.
Alignment of Single Cable Two-Way Systems
Alignment of a bidirectional single cable system requires two operations: aligning the
forward path and aligning the return path. The following paragraphs describe both
procedures done on a midsplit system. The alignment of a subsplit system is similar
except for the frequencies involved.
106
6
Non-amplified System
Non-amplified system refers to a network that does not use an amplifier, or to the
portion of a larger network that is supported only by the headend processing
equipment and does not include an amplifier. Such systems normally include the
headend taps or local distribution taps connected to the headend.
The non-amplified portion of a network must be aligned for proper signal levels
before the rest of the distribution network is aligned. This is because amplifier gain can
be increased or decreased to achieve proper levels, while non-amplified sections do not
have this additional dimension of control. Headend devices, such as data translators
and television channel processors, must be adjusted to provide the proper input Signal
levels to all devices in the non-amplified part of the network.
Forward Path Alignment
In both subsplit and midsplit systems, the forward and return paths occupy a wide
bandwidth. To properly align an amplifier requires observation of several frequencies.
The technique described here uses a sweep generator placed at the headend. Its output
level is adjusted to the video carrier level listed on the system design drawings or in the
system specifications. A spectrum analyzer measures the response of the system at any
point. Figure 6-3 illustrates forward path alignment described in the follOWing
procedure. Follow proper RF measurement techniques throughout this procedure.
RF BANDPASS
SIGNAL SOURCE
HEADEND
168 300/400
-II
SIGNALS ARE ATTENUATED
AND TILTED
-n
SIGNALS ARE AMPLIFIED
AND FLAT
-II
AMPLIFIER
FORWARD PATH ONLY
Figure 6-3. Forward Path Alignment
107
I----<.-n
6
System Alignment
1.
Attach the sweep generator to a combiner input port feeding the forward path at
the headend.
2.
Adjust this generator to supply the desired video carrier level in the forward band
(168-300 MHz).
3.
Connect the spectrum analyzer to the output of the first amplifier in cascade.
The attachment of the spectrum analyzer to the amplifier is critical. Some
amplifiers have built-in test points whose signals are 20 dB below system levels.
Others require the use of a test probe that contains a 30-dB isolation pad. Use the
proper technique for the amplifiers in the system. Consult the appropriate catalog
or manual to find the exact amplifier configuration being used.
4.
Observe the variations in signal levels between the low and high frequencies of
the forward band. Follow the amplifier manufacturer's procedure to set the
equalizer controls of the amplifier properly to obtain a flat response across the
forward band.
5.
Adjust the amplifier's gain until the measured output level matches the specified
design level. When additional attenuation is needed, insert an appropriate pad in
the forward path until the desired signal level is achieved. A good rule-at-thumb is
to select the pad value that allows the gain adjustment control to be within 1 dB of
its maximum setting. This leaves some headroom that can be used when
additional gain is required because of system reconfiguration or cable degradation.
6.
Move the spectrum analyzer to each amplifier in turn and repeat this procedure.
Where the output level of the amplifier is not noted on the drawing, adjust the
amplifier until the desired signal strength is obtained at a typical outlet.
It is possible (but not recommended) to align a system with a field strength meter
instead of a spectrum analyzer. Since the meter monitors a single frequency at a time,
the sweep signal can be replaced by two or three fixed-frequency pilot signals. The
meter can then be used to align the network. This technique usually requires two
people: one to change the frequency of the signal source and one to align the
amplifiers. Communicating over two-way radios in a large facility can speed this
process dramatically. The system should be equalized by comparing the response at
low frequencies to the response at high frequencies (170 and 300 MHz).
Return Path Alignment
Aligning the return path is similar to aligning the forward path with a few exceptions.
Move the signal source from the headend to appropriate taps or outlets in the network.
Adjust this generator to the frequency and amplitude of the transmitting devices that
will be connected to the network. Connect the spectrum analyzer to the output of the
return amplifier being aligned. Then, adjust the return path equalizer to obtain a flat
response at the frequencies of the return path (5 to 116 MHz). The procedure is similar
to that for the forward path.
108
6
q:1-_ +27.5 dBmV
5 dB LOSS
+10 dBmV
+32 dBmV OUTPUT
13 dB LOSS
HEADEND
12 dB LOSS
34.5 dBmV
_--'~
3-WAY SPLITTER INSERTION LOSS
NOTE. ALL LOSSES NOTED ARE CALCULATED
AT116MHz
Figure 6-4. Signal Levels at a Junction
One significant difference between forward and return path alignment is the
addition of signals from different branches on their way to the headend. The unity gain
principle cannot be followed when designing or aligning the return path. Figure 6-4
shows three return path branches being combined, and that the return signal levels
from each branch must be the same. Since the length of cable in each branch is
probably different, the gain of amplifiers in the three paths must be adjusted to
compensate for a different loss. Adjust the gain of each amplifier so that the following
amplifier receives equivalent signal levels from each branch. This procedure can be
tedious, especially for large systems. However, it is a standard procedure that ensures
compatible performance for outlets throughout the network.
Return amplifiers have less gain than forward amplifiers, because there is less
cable attenuation at the lower return frequencies.
For example, 100 feet of a typical 0.500-inch diameter cable has a loss of 1.32 dB of
signal level at 300 MHz. The same length has a loss of only 0.8 dB at 116 MHz. A
section of cable 1000 feet long, would have attenuations of 13.2 dB and 8 dB at 300 MHz
and 116 MHz respectively.
109
6
System Alignment
Trunk and bridger amplifiers have return path gain controls that operate like the
forward path controls. Line extender amplifiers might not have a gain control. When
using amplifiers without gain controls, select an appropriate pad to reduce the signal
level to the desired value.
An alternative to measuring signal levels throughout the network is to calculate
the losses associated with the return path and then make corresponding adjustments of
each transmitter connected to the system. If there are large differences in transmitter
levels, moving devices throughout the network is difficult; adjustments would be
necessary for each device. This approach is not recommended.
Ideally, proper alignment permits the use of pre-adjusted communication devices
anywhere in the network. Network services can then be physically moved from one
location in the network to another without readjusting the interface device or
realigning the network components. In addition, signals introduced into the return
path from any location in the network arrive at the headend within 3 dB of each other.
A system designed and aligned to these specifications can accommodate all originally
installed services and make the addition of new services easy.
Alignment of Dual Cable Systems
Dual cable systems comprise two separate cables. The inbound cable carries signals to
the headend from the distribution network; the outbound cable carries signals from the
headend to the outlets. The bandpasses associated with both paths are from 54 to 300 or
400 MHz. A common design practice is to make the outbound path identical to the
inbound path, except that the direction of the amplifiers is reversed.
There are some special dual cable systems that uSe different bandpasses than those
listed above. Stocking extra amplifier modules for such systems is recommended,-since
spare parts cannot always be purchased off the shelf from standard suppliers.
Inbound and Outbound Cables
Alignment techniques for dual cable systems are similar to those listed for single cable
systems, except that each cable path is unidirectional. Usually, the inbound cable is
identical to the outbound cable.
The most accurate alignment is done by placing a test signal generator at the
remote end of the inbound path. Check the signal at each amplifier and adjust its gain
until the test signal is received at the corresponding outbound path outlet at the
desired level.
Amplifier equalization can also be performed in this manner. Transmit a signal on
the inbound cable and monitor it at each amplifier in the distribution system. The
frequency response at any location can be checked and adjusted easily by this round
robin method.
This procedure can also be performed on a single cable system, by following the
signals translated by the central retransmission facility. This unit receives return path
Signals, converts them to the higher forward frequency band, and transmits them on
the forward path. The round robin signal level can be monitored at any point in the
system.
110
6
An alternative to measuring signal levels throughout the network is to calculate
the attenuation in each span and adjust the amplifiers to compensate for this computed
loss. A span is the cable between the two outlets (for inbound and outbound signals) at
anyone location. As with many alternatives, it is less accurate and should be used only
when no test equipment is available. Never use this technique on any large system.
Potential Problems and Solutions
Problems can arise during alignment. Typical symptoms are a loss of signal or an
inability to adjust the amplifiers to achieve the desired output. Problems can be
divided into the following categories.
~
Design errors
~
Installation defects
~
Hardware failures
~
Vandalism
~
Test equipment defects
Design errors can be minimized by following these rules.
~
Have the system designed by a qualified broadband CATV /LAN engineer.
~
Double-check all mathematical calculations (this is the most frequent cause of
design errors).
~
Verify that the system's components are used within their design speCifications.
~
Completely test any hardware intended for use in an abnormal or non-standard
scheme.
The most common installation defects have the following causes.
~
Poor connections
~
Components installed in the wrong direction
~
Defective amplifiers
~
Improperly configured amplifiers
~
Power supply failures
~
Blown fuses
~
Cable damage
A common symptom of all the above problems is little or no RF signal. RF signal loss
often results from improper tightening of the cable's center conductor in internally
seized cable components. RF signals can radiate across an open path between the two
conductors, but the greater attenuation (because there is no solid connection) results in
a much lower output signal level. If ac power is on the cable to feed the amplifiers, a
good indication of an open circuit is the loss of ac power on the device's output.
If the center conductor is connected too tightly, other problems can occur.
Overtightening can cause the center conductor to weaken and eventually break.
111
6
System Alignment
Incorrect orientation of passive taps can permit signals to pass through the unit with
little or no signal strength at the tap ports. On all components, the direction of Signal
flow (from trunk input to trunk output) is shown by an arrow stamped on the
housing. The arrows on directional coupler taps should point away from the headend.
In dual cable systems, the arrows should also point away from the headend, for both
the inbound and outbound cable paths.
~
Output ports of forward path amplifiers should point away from the headend.
~
Output ports of return path amplifiers should point toward the headend.
~
Output ports of amplifiers in dual cable systems point away from the headend on
the outbound path and toward the headend on the inbound path.
Another common installation defect is a short circuit. Usually, fuses in the power
combiner blow when ac power is applied to a distribution branch containing a short
circuit. Another possible symptom is high current draw. This can pull the voltage
supplied to the amplifiers down below the required level. Test the supply voltage at
each amplifier with a voltmeter. When the amplifier has less than adequate supply
voltage, its performance degrades and affects network operation.
Passive equipment is less subject to failure. M05t failures concerning these devices
are associated with damage during installation. The directional couplers used in trunk or
branch connections have no tap ports. To verify that ac current is passing through the
device, it is necessary to open the cover and measure it with a meter.
One system component that is often overlooked is the outlet drop cable. Usually,
this is a short piece of coaxial cable with type F fittings on both ends. It connects the
interface device to the outlet. Since the cable is directly exposed to the user, or to a
potentially hostile environment, damage can occur. A common problem is physical
stress placed on the cable by the user or by furniture near the cable. This stress can
push the center conductor back into the connector, and cause intermittent or total loss of
signal. The connected device can fail totally or operate intermittently. The resulting
problems can drive the maintenance crew crazy. Such problems can also cast doubt on
system operation in general, when in fact most of the system is operating properly.
Broadband network problems can be found quickly by one who is familiar with
their effects on the system. However, when several problems occur simultaneously,
isolating each one can be difficult. Analyze each problem separately and in a logical
sequence. Don't stop once a problem is discovered and fixed. Continue with the cable
certification test until it is successfully completed.
Radiation and Signal Ingress
It is important for a system to reject outside interference. When the system leaks or
radiates, some external signals must also be entering the system. If these signals are
suffiCiently large, they can disrupt or inhibit normal network operation.
Measurement of signal ingress is straightforward. Install a signal source at the
headend and set it to an amplitude that simulates the level of the strongest signals on
the system. Then monitor all points in the system for excess radiation using an
instrument for measuring leakage (a sniffer or a bloodhound).
Radiation usually occurs where the network is not mechanically sound, because of
poor connections, broken or missing shielding, or extreme kinks and bends in the
cable.
112
6
The return path is susceptible to unwanted signal ingress that is more difficult to
measure. The amount of return path ingress can be related to the amount of radiation
measured in the forward path. Excessive signal ingress is often caused by poor
connector installation and loose or open equipment housings. Another major cause is
the shielding characteristics of the drop cable itself. The highest quality coaxial cable
with several shields should be used. The shielding should consist of alternating foil
and braid shields. Finally, all unused tap ports and drops should be terminated.
Checking an Outlet
During alignment and troubleshooting, it might be necessary to ensure that an outlet
is working. Use the following round robin test to verify that an outlet is functional.
Simulate the operation of a data modem with the following connections. Attach a
combiner (two-way splitter) to the outlet to be tested. Connect a signal generator to one
side of the splitter and a field strength meter to the other side. Inject a test signal into
the system. This signal should be 3.5 dB higher than the output amplitude of the
network interface device usually connected to that outlet (to account for the 3.5 dB loss
through the splitter). The frequency of this test signal should be within the input
bandpass of the data translator (that is, it should be within the return frequency band).
Monitor the translated signal from the original injection point with the field
strength meter. The meter should show a signal level that is 3.5 dB lower than the
usual level at that outlet, because of the splitter.
Monitoring System Performance
Once alignment is completed and the network is put into operation, system
performance should be monitored continuously. Inject a pilot carrier signal at the
headend on a reserved frequency. Set its output amplitude to a value which can be
monitored throughout the network with a field strength meter or a spectrum analyzer,
and install a monitoring device. This reference signal can be checked on a regular basis
at a typical outlet in the system, as part of a preventive maintenance program.
Summary
This chapter covered the alignment and troubleshooting of broadband networks in
general. Once a network has been designed and installed, its performance must be
certified before it can be used with confidence to convey RF signals. Certifying and
aligning the network properly will identify any broken, misplaced, or poorly-installed
components. Certifying the network at the outset aids later troubleshooting work by
providing a record of correct signal levels and frequencies at critical test points for later
verification.
This section listed test equipment and described how to use it when aligning and
testing the network. The differences between forward and return path alignment were
mentioned. Finally, possible network problems and solutions were described,
including RF radiation and signal ingress.
113
7
For Further Reading
For more information about broadband communications consult the following
resources.
~
The appendices following this chapter contain specific details on some topics that
were covered in this book. The bibliography lists reference sources in several
areas.
~
The National Cable Television Association (NCTA) can provide information on
technical standards used in the CATV industry.
National Cable Television Association
918 16th St., NW
Washington, D.C. 20006
202/775-3550
~
The following trade publications can provide useful information in many areas;
this list is not comprehensive.
Cable Age
1270 Avenue of the Americas
New York, NY 10020
Cable Marketing Magazine
Jobson Publishing Corp.
16th Floor
352 Park Avenue South
New York, NY 10010
Cable News
Suite 1200N
7315 Wisconsin Ave.
Bethesda, MD 20814
Cable Vision
2500 Curtis St.
Denver, CO 80205
CATJ, The Official Journal
for the Community
Antenna Television Association
4209 N.W. 23rd
Oklahoma City, Oklahoma 73107
116
Communications Engineering Digest,
The Magazine of Broadband Technology
2500 Curtis St.
Denver, CO 80205
Communications News
124 South First St.
Geneva, Illinois 60134
Data Communications
42nd floor
1221 Avenue of the Americas,
New York, NY 10020
Micro Communications
Miller Freeman Publications
500 Howard St,
San Francisco, CA 94105
••
Appendices
Appendix A: Definition of Terms
Allocations The assignment of specific broadcast frequencies by the FCC for various
communication uses (e.g., commercial television and radio, land-mobile
radio, defense communications, microwave links). This divides the available
spectrum between competing services and minimizes interference between
them. The manager of a broadband network must allocate the available
bandwidth of the cable among different services for the same reasons.
Amplifier A device which increases the power or amplitude of an electrical signal.
Amplifiers are placed where needed in a cable system to strengthen signals
weakened by cable and component attenuation. Two-way, Single-cable
systems use a forward and a reverse amplifier inside one enclosure to boost
signals travelling in both directions.
Balancing Adjusting the gains and losses in each path of a system to achieve equal
signal levels (usually to within 3 dB) at all user outlets. A balanced network
also provides near equal input signal levels to the headend from transmitters
connected anywhere in the network.
Bandwidth The frequency range that a component, circuit, or system can pass. For
example, voice transmission by telephone requires a bandwidth of about
3000 Hertz (3 kHz). A television channel occupies a bandwidth of 6 million
Hertz (6 MHz). Cable systems occupy 5 MHz to 300 or 400 MHz of the
electromagnetic spectrum.
Branch An intermediate cable distribution line in a broadband coaxial network that
either feeds or is fed from a main trunk. Also called a feeder.
Broadband In general, wide bandwidth equipment, or systems that can carry signals
occupying a large portion of the electromagnetiC spectrum. A broadband
communication system can simultaneously accommodate television, voice,
data, and many other services.
Cable Loss The amount of RF signal attenuation by a given coaxial cable. Cable
attenuation is mainly a function of signal frequency and cable length. Cables
attenuate higher frequency signals more than lower frequency signals
according to a logarithmic function. Cable losses are usually calculated and
specified for the highest frequency carried (greatest loss) on the cable.
Cable Powering Supplying operating power to active CATV equipment (for example,
amplifiers) with the coaxial cable. This ac or dc power does not interfere with
the RF information signal.
Cable Tilt The variation of cable attenuation with frequency. The attenuation of a
length of cable increases with frequency; therefore, the amplitude of an RF
sweep signal measured at the end of a length of cable is greater at low
frequencies than it is at high frequenCies. (When viewed on a spectrum
analyzer, the waveform tilts downward).
118
Cable TV Previously called Community Antenna Television (CATV). A communication
system which simultaneously distributes several different channels of
broadcast programs and other information to customers via a coaxial cable.
Carrier Sense, Multiple Access with Collision Detection A communication medium access
technique that allows many separate transceivers to share a single channel.
All units monitor the channel (carrier sense), and do not transmit while
receiving a signal. Whenever the channel is idle, any unit can transmit
(multiple access). If two or more units begin transmitting during the same
break, their signals collide and they realize that a problem occurred (collision
detection). They cease transmitting and wait for a short time before trying to
retransmit the data.
Cascade The number of amplifiers connected in series in a trunk system.
CATV Community Antenna Television (See Cable TV).
Central Retransmission Facility (CRF) The location of the equipment that processes RF
signals for network retransmission. Also called headend.
Coaxial Cable A single cable with two conductors having a common longitudinal axis.
The center conductor carries information signals; the outer conductor
(shield) is grounded for those signal frequencies to prevent interference. This
shield is often made of a flexible foil or braid, or solid aluminum. The two
conductors are separated by an insulating dielectric.
Composite Triple Beat (CTB) The combination of all possible third-order beat frequencies
(Fl ± F2 ± F3) that occur on a system.
Composite Video Signal The complete video signal. For monochrome systems it
comprises the picture, blanking, and synchronizing Signals. For color
systems it includes additional color synchronizing signals and color picture
information.
CRF Central Retransmission Facility.
Cross Modulation A form of signal distortion in which modulation from one or more
RF carrier(s) is imposed on another carrier.
CSMAjCD Carrier Sense, Multiple Access with Collision Detection.
Data Communication Equipment (DCE) Equipment that links a user's data terminal
equipment to a common carrier's line (for example, a modem).
Data Rate The rate of information transfer, expressed in bits per second (bps).
Data Terminal Equipment (DTE) Equipment that is the ultimate source or destination of
data.
119
Appendices
dB Decibel.
dBm V Decibels relative to one millivolt. Zero dBm V is defined as 1 millivolt across 75
ohms.
Number of dBmV
=
20 10glO (V/1 mY)
DCE Data Communication Equipment.
Decibel A unit used to express the ratio between two power or signal values:
Number of dB = 10 10glO (PdP 2 ) = 20 loglO (VdV2 )
where log is the base ten logarithm;
PI! P 2 are measurements of power at two points;
VI! V2 are measurements of voltage at two points having identical
impedance.
Directional Coupler A passive device used in cable systems to divide and combine RF
signals. It has at least three connections: trunk in, trunk out, and tap. The
trunk signal passes between trunk in and trunk out lines with little loss. A
portion of the signal applied to the trunk in line passes to the tap line, in
order to connect branches or outlets to the trunk. A signal applied to the tap
line is attenuated and passes to the trunk in line, and is isolated from the
trunk out line. A signal applied to the trunk out line passes to the trunk in
line, and is isolated from the tap line. Some devices provide more than one
tap output line (Multi-taps).
Distribution Amplifier A high gain amplifier used to increase RF signal levels to
overcome cable and flat losses encountered in signal distribution.
Drop Cable A flexible coaxial cable which connects a network tap to a user's outlet
connector. Also called Drop Line.
Drop Line Device Any external device attached to the coaxial network through a drop
line (e.g., RF modem, television set, audio modulator).
DTE Data Terminal Equipment.
Echo See Reflection.
Equalization A technique used to modify the frequency response of an amplifier or
network to compensate for distortions in the communication channel. The
ideal result is a flat overall response. This slope compensation is often done
by a module within an amplifier enclosure.
F Connector A standard, low cost, 75-ohm connector used by the CATV industry to
connect coaxial cable to equipment.
120
FDM Frequency Division Multiplexing.
Feeder Cable See Branch.
Feedermaker A splitting device used to provide multiple outlet connections from
distribution amplifiers.
Filter A circuit that selects one or more components of a signal depending on their
frequency. Used in trunk and feeder lines for special cable services such as
two-way operation.
Flat Loss Equal signal loss across the system's entire bandwidth, such as that caused by
attenuators.
Flooded Cable A special coaxial CATV cable containing a corrosion resistant gell
between the outer aluminum sheath and the outer jacket. The gell flows into
imperfections in the aluminum to prevent corrosion in high moisture areas.
Forward Direction The direction of signal flow in a cable system that is away from the
CRF or headend.
Frequency The number of times a periodic signal repeats itself in a unit of time, usually
one second. One Hertz (Hz) is one cycle per second. One kilohertz (kHz) is
one thousand cycles per second.
Frequency Division Multiplexing (FDM) Dividing a communication channel's
bandwidth among several subchannels with different carrier frequencies.
Each subchannel can carry separate data signals.
Frequency Response The change of a parameter (usually signal amplitude) with
frequency.
Frequency Translator See Translator.
Harmonic Distortion A form of interference caused by the generation of signals
according to the relationship Nf, where N is an integer greater than one and
f is the original signal's frequency.
Headend The facility that contains a cable system's electronic control center, generally
the antenna site of a CATV system. It usually includes antennas,
preamplifiers, frequency converters, demodulators, modulators, and other
related equipment which receive, amplify, filter, and convert broadcast
television signals to cable system channels. It might house a host computer in
broadband data communication systems.
In two-way broadband systems, the headend holds at least the frequency
translator.
High Frequencies Frequencies allocated for transmission in the forward direction in a
midsplit broadband system, approximately 160 to 400 MHz.
121
Appendices
Highsplit A frequency division scheme that allows two-way traffic on a single cable.
Reverse path signals come to the headend between 5-174 MHz; forward path
signals go from the headend between 232-400 MHz. No signals are present
between 174-232 MHz.
Hub Same as a headend for bi-directional networks except that it is more centrally
located within the network.
Inbound The cable carrying signals to the headend in a dual cable system.
Insertion Loss The loss of signal level in a cable path caused by insertion of a passive
device. Also called Thru Loss.
Isolation Loss The amount of signal attenuation of a passive device from output port to
tap outlet port.
Low Frequencies Frequencies allocated for transmission in the return direction in a
midsplit broadband system, approximately 5 MHz to 116 MHz.
Main Trunk See Trunk Line.
MATV Master antenna television system. A small cable distribution system usually
restricted to one or two buildings.
Mid Band The part of the electromagnetic frequency spectrum that lies between
television channels 6 and 7, reserved by the FCC for mobile units, FM radio,
and aeronautical and maritime navigation applications. This frequency
band, 108 to 174 MHz, can be used to provide additional channels on cable
television systems.
Midsplit A frequency division scheme that allows two-way traffic on a single cable.
Reverse path signals come to the headend between 5-116 MHz; forward path
signals go from the headend between 168-400 MHz. No signals are present
between 116-168 MHz.
Modem A modulator-demodulator device. The modulator codes digital information
onto an analog carrier signal by varying the amplitude, frequency, or phase
of that carrier. The demodulator extracts digital information from a similarly
modified carrier. It allows communication to occur between a digital device
(for example, a terminal or a computer) and an analog transmission channel,
such as a telephone voice line.
Multi-tap A passive distribution component composed of a directional coupler and a
splitter with two or more output connections. See Tap.
Noise Any undesired signal in a communication system.
Noise Figure A measure of the amount of noise contributed to a system by an amplifier.
122
Noise Floor The minimum noise level possible on a system. A 4-MHz-wide channel on
a 75-ohm cable system operating at 68 degrees Fahrenheit has a noise floor of
-59 dBmV.
Outbound The cable carrying signals away from the headend in a dual cable system.
Outlet See Tap Outlet.
Packet Communication Unit (PCU) A device that connects a terminal or computer to a
broadband packet-switched communication network.
Pad A passive attenuation device used to reduce a signal's amplitude.
Receiver Isolation The attenuation between any two receivers connected to a cable
system.
Reflections Secondary signals caused by the collision of the transmitted signal with
structures or objects in its path. Such echoes can be created in a cable system
by impedance mismatches and cable discontinuities or irregularities. Also
called echoes.
Reserve Gain An amplifier has a maximum amount of available gain. When designing a
network, amplifiers are specified to supply this maximum amount of gain
less some amount of reserve gain. This reserve gain can be used to
accommodate signal level variations that can occur during installation. A
common figure used for reserve gain is 2 dB.
Return Loss A measure of the degree of impedance mismatch for an RF component or
system. At the location of an impedance mismatch, part of the incident
signal is reflected back toward its source, creating a reflected signal. The
return loss is the number of decibels that the reflected signal is below the
incident signal.
Return Path Reverse direction; towards the headend.
Signal Level The RMS voltage measured at the peak of the RF signal. It is usually
expressed in microvolts referred to an impedance of 75 ohms or in dBm V.
Slope The difference between signal levels at the highest frequency and at the lowest
frequency in a cable system. Also called spectrum tilt.
Slope Compensation The action of a slope-compensated gain contllOi. The gain of the
amplifier and the slope of amplifier equalization are changed simultaneously
to provide equalization for different lengths of cable; normally specified in
terms of cable loss.
Spectrum Tilt See Slope.
123
Appendices
Splitter A passive device that divides the input signal power from the forward
direction into two or more output signals of less signal power. Input signals
from the reverse direction are combined into a single signal and passed
toward the headend. Splitters pass through 60 Hz power to all1ines.
Subsplit A frequency division scheme that allows two-way traffic on a single cable.
Reverse path signals come to the headend between 5 and 30 MHz; forward
path signals go from the headend between 54 and 400 MHz. No signals
occupy the 30 to 54 MHz band.
Surge Arrestor A device that protects electronic equipment against surge voltage and
transient signals on trunk and distribution lines.
Tap A passive device, normally installed in line with a feeder cable. It removes a
portion of the signal power from the distribution line and delivers it to the
drop line. The amount of power tapped off the main line depends on the
input power to the tap and the attenuation value of the tap. Only the
information signal (and not 60 Hz power) goes to the outlet ports. See also
Multi-tap.
Tap Outlet A type F connector port on a tap used to attach a drop cable. The
information signal is carried through this port. The number of outlets on a
tap usually varies from two to eight.
Termination A 75-ohm resistor that terminates the end of a cable or an unused tap port
with its characteristic impedance to minimize reflections.
Thru Loss See Insertion Loss.
Tilt See Slope (spectrum tilt) or Cable Tilt.
Time Division Multiplexing (TOM) Sharing a communication channel among several
users by allowing each to use the channel for a given period of time in a
defined, repeated sequence.
Translator In a two-way broadband system, an active device at the headend which
receives RF signals coming to it from devices connected to the network,
converts them to signals at a higher frequency, and sends them back to the
network in the forward direction.
Trunk Amplifier A low distortion amplifier that amplifies RF signals for long distance
transport.
Trunk Cable Coaxial cable used for distribution of RF signals over long distances
throughout a cable system. Usually the largest cable used in the system.
Trunk Line The major cable link(s) from the headend (or hub) to downstream branches.
Also called Main Trunk.
124
Unity Gain A design principle wherein amplifiers supply enough signal gain at
appropriate frequencies to compensate for the system's cable loss and flat loss
(Cable Loss + Flat Loss = Amplifier Gain). It implies the use of identical
amplifiers separated by identical lengths of cable.
Usable Gain The amount of gain an amplifier can supply after subtracting any loss due
to its internal modules, and any reserve gain.
125
Appendices
Appendix B: Abbreviations and Acronyms
amplifier input level in dBm V
alternating current
automatic gain control
amplitude modulation
amplifier
bandwidth
amplifier cascade factor;
1Olog(NC)
carrier-to-hum ratio in dB
C/H
carrier-to-noise ratio in dB
C/N
CATV
Community Antenna
Television
CCTV
Closed-Circuit Television
central retransmission facility
CRF
CRT
cathode-ray tube
CSMA/ carrier-sense, multiple access
with collision detection
CD
composite triple beat
CTB
decibel
dB
decibel referred to one
dBm
milliwatt; 0 dBm = 1 m W
decibel referred to one millivolt;
dBmV
OdBmV = 1 mV
directional coupler
DC
direct current
dc
data carrier level
DCL
noise level, noise floor
En
electromagnetic interference
EMI
equivalent noise input in dBm V
ENI
noise figure
F
Fl,F2
carrier frequencies on a system
Federal Aviation
FAA
Administration
Federal Communications
FCC
Commission
frequency-division
FDM
multiplexing
FM
frequency modulation
frequency shift keying
FSK
amplifier gain
G
gigahertz; 1,000,000,000 cycles
GHz
per second
harmonically-related carriers
HRC
A
ac
AGC
AM
amp
B
C
126
Hz
IMD
IRC
ITFS
Hertz; 1 cycle per second
inter modulation distortion
interval-related carriers
instructional television, fixed
service
kHz
kilohertz; 1,000 cycles per
second
local area network
LAN
MATV master antenna television
manual gain control
MGC
MHz
megahertz; 1,000,000 cycles per
second
modem modulator-demodulator
mux
multiplexer
millivolts
mV
number of amplifiers in cascade
N
number of data carriers in a 6NC
MHz channel
National Cable Television
NCTA
Association
public address
PA
packet communication unit
PCU
random access memory
RAM
radio frequency interference
RFI
remote job entry workstation
RJE
read-only memory
ROM
R/WM read/write memory
RF
radio frequency
amplifier output level in dBm V
S
s
seconds
signal-to-noise
ratio
SIN
TOM
time-division multiplexing
TVRO
television, receive-only
ultra-high frequency; 300-3000
UHF
MHz
microvolts
/.tV
V
Volts
video carrier level
VCL
very-high frequency; 30-300
VHF
MHz
XLTR
translator
cross-modulation distortion
XM
Appendix C: Broadband Symbols
~
TWO-WAY SPLITTER
DIRECTIONAL COUPLER 8 dB
~
PAD VALUE
EQUALIZER
lC[>---
{ LOW LOSS LEG
THREE-WAY SPLITTER
ONE-WAY AMPLIFIER
TWO-WAY AMPLIFIER
MANUAL GAIN
~VALUE
~24L
TWO PORT TAP
r:;:-vI~
VALUE
FOUR PORT TAP
TWO-WAY AMPLIFIER
AGC RETURN GAIN
TWO-WAY AMPLIFIER
AGC FORWARD GAIN
VALUE
EIGHT PORT TAP
TWO-WAY AMPLIFIER
WITH BRIDGER
FIXED EQUALIZER
1, 2, 3 & 4 OUTPUTS
VARIABLE EQUALIZER
~
TWO-WAY AMPLIFIER
AGC FORWARD GAIN
NO RETURN AMPLIFIER
DESIGNATES HEADEND
-----<.~ IAC BLOCKING TERMINATOR
F FITTING COMPONENTS
--¥M
L
H
DIPLEX FILTER
S = SUBSPLIT
M = MIDSPLIT
r-8
FOUR PORT TAP
60 VOLT POWER SUPPLY
ONE PORT TAP
--tIe}-POWER COUPLER
,---SIZE
.4121500' __ FEET
CABLE I D
FILTER
-8 dB -LOSS
0.750 INCH CABLE
0.500 INCH CABLE
0.412 INCH CABLE
Figure C-l. Broadband Symbols
127
TERMINATOR
SPLITTER
Appendices
Appendix D: Frequency Allocations
BROADCAST
CABLE
174 MHz
I
BROADCAST
CABLE
300 MHz
I
BROADCAST
CABLE
Figure D-1. Frequency Allocation Chart
128
Table D-1. Headend Channel Assignment Reference Table
FORMER ASSIGNMENT
INCREMENTAL (IRC) FREQUENCY ASSIGNMENT
FREQUENCY
CHANNEL
DESIGNATION
2
3
4
5
6
A
B
C
D
E
F
G
H
I
7
8
9
10
11
12
13
J
K
L
M
N
0
P
Q
R
S
T
U
V
W
RANGE
MHz
54-60
60-66
66-72
76-82
82-88
120-126
126-132
132-138
138-144
144-150
150-156
156-162
162-168
168-174
174-180
180-186
186-192
192-198
198-204
204-210
210-216
216-222
222-228
228-234
234-240
240-246
246-252
252-258
258-264
264-270
270-276
276-282
282-288
288-294
294-300
Incremental assignments
are used for non-phaselocked
headendsand forincrementalIy related carrier I RC phaselock headends
Harmonic assignments are
used for harmonically related
carner HAC
head ends.
phaselock
FREQUENCY PICTURE
CHANNEL
DESIGNATION
2
3
4
56141
151
161
171
181
191
201
211
221
7
8
9
10
11
12
13
231
241
251
261
271
281
291
301
311
321
331
341
351
361
371
381
391
401
411
421
431
441
451
461
471
481
491
501
511
521
531
541551561571
581
591
601
RANGE
MHz
54-69
60-66
66-72
76-82
82-88
120-126
126-132
132-138
138-144
144-150
150-156
156-162
162-168
168-174
174-180
180-186
186-192
192-198
198-204
204-210
210-216
216-222
222-228
228-234
234-240
240-246
246-252
252-258
258-264
264-270
270-276
276-282
282-286
228-234
294-300
300-306
306-312
312-318
318-324
324-330
330-336
336-342
342-348
348-354
354-360
360-366
366-372
372-378
378-384
384-390
390-396
396-402
72-78
78-84
84-90
90-96
96-102
102-108
108-114
114-120
CARRIER CARRIER
MHz
MHz
55.25
61.25
67.25
77.25
83.25
121.25
127.25
133.25
139.25
145.25
151.25
157.25
163.25
169.25
175.25
181.25
187.25
193.25
199.25
205.25
211.25
217.25
223.25
229.25
235.25
241.25
247.25
253.25
259.25
265.25
271.25
277.25
283.25
289.25
295.25
301.25
307.25
313.25
319.25
325.25
331.25
337.25
343.25
349.25
355.25
361.25
367.25
373.25
379.25
385.25
391.25
397.25
73.25
79.25
85.25
91.25
97.25
103.25
109.25
115.25
HARMONIC (HRC) FREQUENCY ASSIGNMENT
SOUND
59.75
65.75
71.75
81.75
87.75
125.75
131.75
137.75
143.75
149.75
155.75
161.75
167.75
173.75
179.75
185.75
191.75
197.75
203.75
209.75
215.75
221.75
227.75
233.75
239.75
245.75
251.75
257.75
263.75
269.75
275.75
281.75
287.75
293.75
299.75
305.75
311.75
317.75
323.75
329.75
335.75
341.75
347.75
353.75
359.75
365.75
371.75
377.75
383.75
389.75
395.75
401.75
77.75
83.75
89.75
95.75
101.75
107.75
113.75
119.75
·Use either 5 and 6 or 541,551. and 561.
FREQUENCY ASSIGNMENTS FOR SUBSPLIT
RETURN CHANNELS ON TWO-WAY SYSTEMS
FREQUENCY
CHANNEL
DESIGNATIO~
RANGE
MHz
2H
3H
4H
5"
6'14H
15H
16H
17H
18H
19H
20H
21H
22H
7H
8H
9H
10H
11H
12H
13H
23H
24H
25H
26H
27H
28H
29H
30H
31H
32H
33H
34H
35H
36H
37H
38H
39H
40H
41H
42H
43H
44H
45H
46H
47,H
48H
49H
50H
51H
52H
53H
54H
55H"
56W'
57H
58H
59H
60H
61H
52.75-58.75
58.75-64.75
64.75-70.75
76.75-82.75
82.75-88.75
118.75-124.75
124.75-130.75
130.75-136.75
136.75-142.75
142.75-148.75
148.75-154.75
154.75-160.75
160.75-166.75
166.75-172.75
172.75-178.75
178.75-184.75
184.75-190.75
190.75-196.75
196.75-202.75
202.75-209.75
208.75-214.75
214.75-220.75
220.75-226.75
226.75-232.75
232.75-238.75
238.75-244.75
244.75-250.75
250.75-256.75
256.75-262.75
262.75-268.75
268.75-274.75
274.75-280.75
280.75-286.75
286.75-292.75
292.75-298.75
298.75-304.75
304.75-310.75
310.75-316.75
316.75-322.75
322.75-328.75
328.75-334.75
334.75-340.75
340.75-346.75
346.75-352.75
352.75-358.75
358.75-364.75
364.75-370.75
370.75-376.75
376.75-382.75
382.75-388.75
388.75-394.75
394.75-400.75
70.75-76.75
76.75-82.75
82.75-88.75
88.75-94.75
94.75-100.75
100.75-106.75
106.75-112.75
112.75-118.75
SOUND
54
60
66
78
84
120
126
132
138
144
150
156
162
168
174
180
186
192
198
204
210
216
222
228
234
240
246
252
258
264
270
276
282
288
294
300
308
312
318
324
330
336
342
348
354
360
366
372
378
384
390
396
72
78
84
90
96
102
108
114
58.5
64.5
70.5
82.5
88.5
124.5
130.5
136.5
142.5
148.5
154.5
160.5
166.5
172.5
178.5
184.5
190.5
196.5
202.5
208.5
214.5
220.5
226.5
232.5
238.5
244.5
250.5
256.5
262.5
268.5
274.5
280.5
286.5
292.5
298.5
304.5
310.5
316.5
322.5
328.5
334.5
340.5
346.5
352.5
358.5
364.5
370.5
376.5
382.5
388.5
394.5
400.5
76.5
82.5
88.5
94.5
100.5
106.5
112.5
118.5
**5H and 6H are same as 55H and 56H.
CHANNEL
T7
T8
T9
no
129
ICTURE
CARRIER CARRIER
MHz
MHz
FREQUENCY PICTURE
5.75-11.75
11.75-17.75
17.75-23.75
23.75-29.75
7
13
19
25
SOUND
11.5
17.5
23.5
29.5
Appendices
Appendix E: RF Calculations
This appendix provides some details on calculations used in the design of RF systems.
These details supplement material provided in chapter five on broadband system
design. Also, the final section contains a summary list of broadband design equations.
The dBm and dBm V
As previously stated, 0 dBm V = 1 m V. Another commonly-used unit is the decibel
referred to one milliwatt, abbreviated dBm, and defined as
Number of dBm = 1010g(P/1 mW)
where P = power in milliwatts.
This equation provides a positive value when P is greater than 1 m W, and a negative
value when P is less than the reference value of 1 m W.
A useful equation relates dBm V and dBm. This equation is based on an assumed
impedance value. The common value for CATV systems is 75 Ohms. For a
1 m V drop across a 75-0hm resistor, the power consumed can be calculated and
converted to dBm.
Given 0 dBm V = 1 m V across 75 Ohms,
P = V 2 /R = 1.3(10)-8 Watts, and in dBm,
P = 1010g (1.3(10y8)/(lO)-3) dBm
P = -49 dBm
Therefore,
a dBm V
= -49 dBm in a 7S-0hm system.
Noise
The basic definition of thermal noise power is
P n = kTB Watts
where k = 1.38(lOy23 Joule/"K (Boltzmann's constant)
T = ambient temperature in OK
B = bandwidth in Hertz
When room temperature is assumed (3000K),
P = 4.IB(lOy21 Watts (at T=3000K)
Converting this to dBm gives
P = 1010g(4.lB(lOy21 /(10y3)
P = -174
130
+
1010g(B) dBm
A noise level in dBm V can also be calculated.
E = -174 + 10log(B) + 49 dBmV
E = -125 + lOlog(B) dBmV
The minimum noise level (noise floor) for a 4-MHz channel can be found.
En = -125 +IOlog(4(lO)6)
En = -59 dBmV
Carrier-to-Noise Ratio
The noise output of a single amplifier, En1 ' is
En! = Enf + G + Fo dBm V
where Enf = noise floor
G = amplifier gain
FO = amplifier noise figure
The carrier-to-noise ratio of one amplifier (C/NO) is
C/NO= S - Enl
= S - Enf - G - FO
= -Enf + A - FO
where S = amplifier output level
A = amplifier input level
For example, if noise floor is -59 dBm V, input level is lO dBm V, and FO is 7 dB,
C/NO = -(-59) + lO - 7 = 62 dB
In a cascaded system where N is the number of cascaded amplifiers,
System noise figure = F = FO + 10log(N)
System C/N = -Enf + A - F
= -Enf + A - FO - 100og(N)
= C/NO - 100og(N)
131
Appendices
Broadband Design Equations
This section lists some of the equations used in designing broadband networks. Most
of these have been covered in the text, and they are listed here together for
convenience. Consult appendix B for abbreviations.
A subscript of zero (0) indicates that the parameter is, for example, for an
individual amplifier. A subscript of 'c' indicates that the parameter is for a cascade of
amplifiers.
dB:
1Olog(P 11 P 2) = 20log(V 11 V2)
dBm:
IOlog(PI/l mW)
dBmV:
20log(V 111 m V)
Unity Gain:
Flat Loss
Rate of change of coaxial
cable attenuation
with temperature:
1% per 10 degree Fahrenheit change
Carrier Derating:
DCL = VCL - IOlog(NC)
Noise Floor:
En = -125
Noise Figure of a Cascaded
System:
Fc = Fa
Noise Floor of a Cascaded
System with B = 4MHz:
En = -59
CIN of a Single Amplifier:
CINo
CIN of a Cascaded System:
CINc = CINo - IOlog(N)
Usable Gain:
Minimum Full Gain - (Module Losses
Maximum Cascaded
Amplifier Output Level:
Sc = So - IOlog(N)
C/H of a Cascaded System:
C/Hc = C/Ho
+ 20log(N)
CTB of a Cascaded System:
CTBc = CTBO
+
132
=
+ Cable Loss
+
= Amplifier Gain
IOlog(B) dBmV
+ IOlog(N)
+
Fa
+
IOlog(N) dBm V
Input Level - En - Fa
20log(N)
+
Reserve Gain)
Appendix F: Test Equipment
This appendix lists test equipment useful for broadband communication networks,
including manufacturer's model numbers. Section 6 briefly describes some of the
instruments listed here. This list is not comprehensive, nor does it indicate preference
for one device over another. It does provide alternate choices in some of the categories.
Table F-l.
Test Equipment for RF Network Certification
Manufacturer
Description
Model
Spectrum Analyzer .4-450 MHz
VSM5B
Digital Storage
DS9
Spectrum Analyzers
Texscan
-----------------------
Hewlett Packard
.~-------------------------------------
Analyzer .01-450 MHz
8557 A, 8558B
Display
182T
Option for dBm V Calibration 75 Ohm
001
--------------------
Tektronix
--
Analyzer 100 Hz-1.5 GHz
8568A
Option 75 Ohm BNC 100 Hz-1500 MHz
001
Option Handle-Flange Kit
909
Option Spare Manual
910
Option Slide Kit
E-lO
IEEE 488 Cable
1063lB
Analyzer
462
-------------------------
Tektronix
----------------------------------------
492
Analyzer
-------------
1880
Wavetek
Sweep Generators
Wavetek
Sweep/Signal Generator 1-500 MHz
180lB
-------------------
Wavetek
Hewlett Packard
133
Pilot Carrier Notches Option
A-7
Sweeper 1-2500 MHz
2002A
Option Rack Mount
K108
Gen. 100 kHz -990 MHz
8656A
50 To 75 Ohm Adaptor
11687A
Option Handle-Flange Kit
909
Appendices
Table F-l.
Test Equipment for RF Network Certification (Continued)
Manufacturer
Description
Model
Texscan
Sweep Generator 1-1000 MHz
VS-60B
75-0hm Impedance
Z Option
Rack Mount Option
RMOption
Sweep Transmitter 1-450 MHz
1855B
Notch Filters, Two Adjustable
B-2
Sweep Analyzer 5-450 MHz
1865B
Camera with Bezel
C-l
SIMa Sweep 1-450 MHz
9551T
Receiver
9551RS
Cable Sweep Systems
Wavetek
Texscan
Digital Storage Option
Field Strength Meters
Texscan
Digital Field Strength Meter
Digitek
Field Strength Meter
7272
Computerized FSM
SAM III D
Field Strength Meter
SAM III
Computerized FSMj Analyzer
SAM IV
Bloodhound Transmitter
FDM-l
Bloodhound Receiver
FDM-2
Mounting Bracket
MB-2
Wavetek
Calibrated Dipole Antenna
RD-l
Comsonics
Sniffer
S200H-l
Multi-Channel Generator
5516
Wavetek
Radiation Testers
Texscan
Carrier Generators
Dix Hill
134
Table F-l.
Test Equipment for RF Network Certification (Continued)
Manufacturer
Description
Model
Fluke
AC/DC Meter
8024B
Hewlett Packard
Frequency Counter
5303B
Texscan
Variable BPF
3U 95/ 1905XX
Jerrold
Fixed BPF
PBF-*
Wavetek
Variable BPF
PP- 5-110
PP-110-220
PP-220-400
Wavetek
Variable Attenuator
7580.1
Texscan
Switchable Attenuator
SA-70F
Pad Kit
KFP-75 F Kit
Cable Bridges
RCB-3/75
Storage Scope
468
Digital Meters and
Frequency Counters
Variable and Fixed
Bandpass Filters
Variable and Fixed
Attenuators
Bridges
Texscan
Storage Scope
Tektronix
Cameras
Tektronix
C-3013
Wavetek
C-1
Power Meters
Hewlett Packard
435B
Wavetek
1034A
135
Appendices
Appendix G: Tools for Installation and Maintenance
This appendix lists tools that are helpful in installing and maintaining broadband
systems. All items included were available when this list was compiled. This list does
not indicate a preference for the items listed here over similar items that are not
included.
Trunk Cable Tools
Tool
Manufacturer
Model
Cable Trailer 2 Reel w / 54" dia.
Lemco Tool Corp.
6354
Cable Trailer 2 Reel w / 54" dia.
Lemco Tool Corp.
5254
Reel Buck
Lemco Tool Corp.
1220
Cable Block
Lemco Tool Corp.
PY-750
Cable Block
Lemco Tool Corp.
M1070-1
Cable Block
General Machine
8093
Cable Block
General Machine
G1315
Corner Block 90 degree
Lemco Tool Corp.
K468J
Corner Block 45 degree
Lemco Tool Corp.
M370J
Cable Chute
Lemco Tool Corp.
S1059
Traction Dynamometer (Tension gage)
Dillon
AX-1A
Pulling Grip (Sling)
Kellems
033-03-010
Loop Former, 0.500 Cable
Q-E Tools & EqUip.
R500J
Cable Cutters
Benner-Nawman
UP-B76
Jacket Stripper
Cablematic
JST-500
Stripper, Side Cut
Cablematic
DST-500
Corer
Cable Prep
DCT-500
Corer
Jerrold
CPT-500
Corer
Cablematic
CCT-500
Corer
Lemco Tool Compo
T-500
Corer (Power)
Lemco Tool Compo
0-500
Combination Stripper (Basic Power)
Cable Prep
SCT-500
136
Brake for above
SCT3006
T-Handle for above
TB-3003
Tool
Manufacturer
Model
Center Conductor Cleaner
Cablematic
CC457
Center Conductor Cleaner
Cable Prep
4010
Tool
Manufacturer
Model
Trimmer
Cablematic
UT6000
Crimper
Cable Prep
HCT-659
Crimper
Cablematic
CR-596
Crimper
Blonder-Tongue
4899
Shield Expander RG-6 Cable
Cablematic
FT-6
Tool
Manufacturer
Model
Female "F" to Male "G" Adapter
Jerrold
PGF
Pot Screwdriver
Bourns
Trimpot
Drop Cable Tools
Miscellaneous
137
Appendices
Appendix H: Broadband Equipment Suppliers
Full-Line Manufacturers
C-Cor Cable TV Industries
60 Decibel Road
State College, PA 16801
(814) 238-2461
Scientific Atlanta
Box 105027 Dept. A-R
Atlanta, Georgia 30348
(404) 441-4000
General Instrument Corp.
RF Systems Division
4229 S. Fremont Ave.
Tucson, Arizona 85714
(602) 294-1600
Jerrold Division
Hatboro, PA
(215) 679-4800
Sylvania
CATV Division
1790 Lee Trevino
Suite 600
El Paso, Texas 79936
(915) 591-3555
Magnavox
CATV Systems, Inc.
133 W. Seneca St.
Manlius, New York 13104
(315) 682-9105
RCA Cablevision System
West Coast
8500 Balboa BId.
Van Nuys, Calif. 91409
(213) 891-7911/ (800) 423-5617
East Coast (800) 345-8104
138
Theta-Com Division
122 Cutter Mill Rd.
Great Neck, NY 11021
(800) 528-4066
Support Manufacturers and Distributors
Anixter-Pruzan
18435 Olympic Street
Tukwila, Washington 98124
251-5287
(800) 323-8166
Berk-Tec Cable
Box 60 Rd 1
Reading, Penn 19607
(215) 376-8071
Blonder-Tongue
One Jake Brown Rd.
Old Bridge, NJ 08857
(201) 998-0695
Catel Division
United Scientific Corp.
4800 Patrick Henry Dr.
Santa Clara, CA 95054
(408) 988-7722
Cerro (Capscan Inc.)
Halls Mill Rd
Adedhia, NJ 07710
(201) 462-8700
Com-Scope
RT 1, Box 199A
Catawba, N.C. 28609
(704) 241-3142
Copperweld Corp.
Bi-Metallics Group
2 Robinson Plaza
Pittsburgh, PA 15205
(412) 777-3000
General Cable
P.O. Box 700
Woodbridge, NJ 07095
(201) 636-5500
Gilbert Connector Co.
3700 N. 36 Avenue
Phoenix, Arizona 85019
(800) 528-5567
Oak Communications
CATV Division
Crystal Lake, IL 60014
(815) 459-5000
139
North Supply
10951 Lakeview Ave.
Lenexa, Kansas 66219
(913) 888-9800
Phasecom Corp
6365 Arizona Circle
Los Angeles, Calif. 90045
(213) 641-3501
Pioneer
3518 Riverside Drive
Columbus, OH 43221
(614) 451-7694
RMS
50 Antin PI.
Bronx, New York 10462
(800) 223-8312
3M TeleComm Products
Dept. TL81-35
P.O. Box 33600
St. Paul, Minnesota 55133
Times Fiber Communications
Cable TV Division
P.O. Box 384
Wallingford, CT
(203) 265-2361
Tomco Communications Div.
United Scientific Corp.
4800 Patrick Henry Dr.
Santa Clara, CA 95054
(408) 988-7722
Toner Cable Equip.
969 Horsham Rd.
Horsham,PA.19044
(800) 523-5947
Wavetek
P.O. Box 190
Beech Grove, IN 46107
(317) 787-3332
Appendices
Appendix H: Broadband Equipment Suppliers (Continued)
Satellite Equipment Manufacturers
A Television Receive Only (TVRO) earth station receives low-power microwave
television signals from orbiting communication satellites. The signals are captured
with a parabolic antenna, amplified and converted to standard television signals, and
then fed to a conventional television distribution system.
Avantek
3175 Bowers Ave.
Santa Clara, CA 95051
(408) 496-6710
Hughes Aircraft Compo
P.O. Box 2999
Torrance, Calif. 90509
(213) 534-2170
Comtech Data Corp.
350 N. Hayden
Scottsdale, AZ 85257
(602) 949-1155
Microwave Assoc.
63 3rd Ave.
Burlington, MA 01803
(617) 272-3100
Gardiner Communications
3605 Security Street
Garland, TX 75042
(800) 527-1392
Scientific Atlanta
Box 105027 Dept. A-R
Atlanta, Georgia 30348
General Instrument Corp.
RF Systems Division
4229 S. Fremont Ave.
Tucson, AZ 85714
(602) 294-1600
Jerrold Division
Hatboro, PA
(215) 679-4800
140
SCN
RD. 1, Box 3114
Basking Ridge, NJ 07920
(201) 658-3838
Toner Cable Equipment Inc.
969 Horsham Rd.
Horsham, PA. 19044
(800) 523-5947
Appendix I: System Grounding
Introduction
The problem of ground potential differential between buildings or different areas of a
broadband system is often a concern for the installing engineer. This situation occurs
when either significantly different loading situations occur in different buildings, or
when the actual grounding apparatus for each building is at a different ground
potential.
In a single distribution environment such as a single building or factory, a ground
grid normally exists. This grid can comprise the structural steel, the plumbing piping,
or any other generally available conducting system which is tied to earth through a
ground rod or other metallic means. The wire from the grid to the ground is normally
#6 copper.
Broadband System Grounding
Broadband systems, just as all other electromagnetic systems, must be grounded.
Grounding is important for protection of individuals from shock hazard while using
or working on the system, and for proper operation of the active components of the
system.
A basic broadband system is grounded at the headend by attaching its power
supply to the building's ground. This establishes the initial ground potential of the
cable's shield throughout the system.
In all broadband systems, it is recommended that an attachment from each
amplifier be made to the ground grid. This minimizes the possible effects of shield
currents on power supplies. When this ground attachment is made with hanger
clamps to the structural steel, ensure that the paint has been adequately cleaned to
make a good metal-to-metal contact.
Grounding individual taps from the trunk cable is not necessary, but can improve
power line isolation for the device at the end of the drop. This technique is only
suggested for environments where the ground grid is readily accessible and can be
used as the mechanical support for the device.
Annual inspection of the trunk cable can identify potential ground failures caused
by corrosion, physical damage, or vandalism.
Grounding Between Sites
Broadband systems often connect more than one building or site, and Significant
ground potential problems might exist. A large difference can cause shield currents on
the cable that could degrade the performance of the amplifiers in that area of the
system. Reduce these intersite potential differences as much as possible.
~
Grounding individual amplifiers reduces this problem.
~
Installing suitable earth rods at the exit and entry points of each site also helps.
However, ensure that these earth rod grounds do not function as overall grounds
141
Appendices
for the entire site. Information on earth rod techniques is available from
manufacturers of this type of equipment, and in references listed in a publication
of Copperweld Steel Company. * A megger test of the actual ground is one way to
assure a reasonable potential to ground.
Above Ground Exposed Trunks
Intersite trunks which run above ground are exposed to natural energies that are not
normally encountered inside a facility. CATV equipment is designed specifically for
this environment. Surge protectors in the amplifiers protect the internal circuitry from
this type of overload.
~
Ground every eighth or twelfth support pole with a good earth rod ground to
protect the system from the energy transmitted by these forces.
~
Ground all amplifiers.
Establish a preventive maintenance program to replace the surge protectors
annually, especially in areas of high electrical activity.
~
~
Periodically inspect the grounds for such problems as corrosion, physical damage,
and vandalism.
Ground Loops
Even with a properly designed system using good ground techniques, ground loop
problems might occur. These problems are often a result of an inadequate ground grid
at the site. The first point of attack should be a thorough survey of the current ground
system and of the methods for reducing or eliminating any ground potential
differences at the site. If this does not solve the problem, try a sectored approach to
ensure that the system ground in each area of the building is adequate. Then ensure
that the power system ground potential for the individual devices that are attached in
that area are tied together.
It is unusual for a large enough ground potential difference to exist to hamper
system performance. A thorough investigation of the system should be conducted
before taking these steps to ensure that ground loops are the problem.
Conclusion
While elimination of ground potential differences is the best approach, other
techniques are available to eliminate or at least minimize the impact of these problems.
The following list identifies the key elements in controlling ground problems in
broadband systems.
~
Ground all amplifiers either directly to the ground grid or to the nearest ground
point with #6 copper ground .
• "Principles and Practices for Securing Safe, Dependable Grounds by Means of Driven Electrodes,"
Copperweld Steel Company.
142
~
For between-site potential differences, first attempt to resolve them using good
ground engineering techniques.
~
When between-site potential differences are significant and cannot be eliminated,
use properly selected earth grounds at exit and entry points of the sites.
~
Make periodic inspections of the trunk cable and amplifiers to identify potential
problems.
~
For overland trunks, use the CATV practice of grounding on every eighth or
twelfth pole.
~
For overland trunks, institute a preventive maintenance program for surge
protector replacement and cable inspection.
~
For suspected ground loops, first test to ensure that ground loops are really the
problem and then use a sectored ground approach.
143
Appendices
Appendix J: RF Connector Details
This appendix contains some specific details on RF connectors used in broadband cable
systems.
A
B
(a) Feed Thru
A
Body with integral mandrel.
B
Locking nut for seizing and retaining cable outer
conductor.
=~
c
B
A
(b) Pin Type
A
Body with cable center conductor seizing pin.
B
Main nut with integral mandrel.
C
Locking nut for seizing and retaining cable outer
conductor.
C
B
A
(c) Splice
A
Body with cable center conductor seizing device.
B
Main nut with integral mandrel (two on splice).
C
Locking nut for seizing and retaining cable outer
conductor (two on splice).
Figure J-l. Terminology for Connector Parts
144
B
C
A
B
C
(d) F Female
A
Body with F female port and cable center conductor
seizing device.
B
Main nut with integral mandrel.
C
Locking nut for seizing and retaining cable outer
conductor.
A
B
C
(e) F Male
A
Body with F male connection and cable center
conductor seizing device.
B
Main nut with integral mandrel.
C
Locking nut for seizing and retaining cable outer
conductor.
A
B
(f) Terminator
A
Body contains RF signal termination and AC power
blocking along with cable center conductor seizing
devices.
B
Main nut with integral mandrel.
C
Locking nut for seizing and retaining cable outer
conductor.
Figure 1-1. Terminology for Connector Parts (continued)
145
C
Appendices
The entry port configuration of most equipment (amplifiers, extenders, taps, etc.) used with coaxial cable
has the following dimensions.
1-1/8"
TYPICAL
%" DIA. COUNTER BORE FOR
CONNECTOR O-RING SEAL
t
5/8-24 THO
x 7/16 DEEP
CENTER CONDUCTOR
OR PIN TYPE CONNECTOR
SEIZING TERMINAL.
·-'vAI:JL.1:
Entry connectors, both feed thru and pin type, are manufactured to be compatible with this port design.
5/8 DIA. O-RING
5/8-24 THO - - -
1.50
~.31-1
---I
PIN-TYPE
Figure /-2. Connector/Equipment Interface
146
F-59/F-6
A connector that seizes only the outer braid and
jacket of the coaxial cable. The center conductor
extends through this connector becoming the
center contact. Male. 3/8-32 threads.
F-81
This connector is utilized to join together two cables.
Female/Female splice. 3/8-32 threads.
F-61
Equipment or panel mounted connector with solder lug. 3/8-32 threads.
F-71
Double ended F connector equivalent to short
jumper cable. Male/male splice. 3/8-32 threads.
This connector terminates the RF signal. The 60 Hz
power may optionally be blocked in specific devices.
TERMINATORS
CRIMP RING
OJ
Standard lock nut with 3/8-32 thread; 1/2 inch hex
and .093 thick.
PANEL NUT
WASHER
o
Figure J-3. F Connector Types and Parts
147
Metallic ring which is mechanically deformed to
provide retention of the coaxial cable braid and
jacket onto the F-59/F-6 connectors. Sometimes
incorporated into these connectors.
Copper disc with hole diameter .385 and 1/2 inch
0.0 ..025 thick for mounting F connectors to
sheetmetal panels.
Appendices
Appendix K: Bibliography
Some of these publications might be out of print, but can often be found on the dusty
shelves of the local electronic distributor or book store. Many texts can be located in the
technical section of good public or university libraries.
Ameco, Inc., Cable Installation Handbook, Milwaukee, McGrawEdison Power Systems
Division, #67100.
Baum, Robert E. and Theodore B., 101 Questions & Answers About CATV & MATV,
Indianapolis, Indiana, Howard W. Sams & Co. Inc., 1968. #20655.
Beever, Jack, Engineer's Guide To Specifications For Multiple Television Distribution
Systems, Philadelphia, Jerrold Electronics Corporation, 1966.
Buckwalter, Len, 99 Ways To Improve Your TV Reception, Indianapolis, Howard W. Sams
& Co. Inc., 1969. #20708.
Camps, Albert, and Markum, Joseph A., Microwave Primer, Indianapolis, Howard W.
Sams & Co. Inc., 1965. #MMC-1.
Cantor, Lon, How To Select And Install Antennas, New York, Hayden Book Company,
1969. #0786.
Cantor, Lon, Planning & Installing Master Antenna TV Systems, New York, John F. Rider
Publisher, Inc., 1965. #0388.
Cooper, Bemarr, ITFS Instructional Television Fixed Service (2500 Megahertz) What It IsHow To Plan, Washington, D.C., National Education Association, 1967.
Cooper, Robert B., Jr., CATV System Maintenance, Blue Ridge Summit, Pa., Tab Books,
1970. #T-82.
Cooper, Robert B., Jr., CATV System Management & Operation, Blue Ridge Summit, Pa.,
Tab Books, 1966, #T100.
Copperweld Bimetallics Group, Practical Grounding, Pittsburgh, Pa.
Cunningham, John E., Cable Television, Second Edition, Indianapolis, Howard W. Sams
& Co., Inc., 1980. #21755.
Darr, Jack, Eliminating Man-Made Interference, Indianapolis, Howard W. Sams & Co.,
Inc., 1960. #MMD-1.
Diamond, Robert M., A Guide To Instructional Television, New York, McGraw-Hill Book
Company, 1964.
Editors of "Electronic Technicial/Dealer", A Practical Guide To MATV/CATV System
Design & Service, Blue Ridge Summit, Pa., Tab Books, 1974. #731.
Elroy, Hansen & Lawrence, Paul, Home TV-FM Antennas, Indianapolis, Howard W.
Sams & Co., Inc., 1974. #21076.
FCC Rules and Regulations, Technical Standards for CATV, Section 15, Subpart K,
Sections 76-601 to 76-617.
Hewlett Packard, Cable Television System Measurements Handbook, Santa Rosa, Ca.,
Hewlett-Packard Co., 1977. #5955-8509.
148
Kamen, Ira, Questions And Answers About Pay TV, Indianapolis, Howard W. Sams & Co.,
Inc., 1973. #20971.
Kamen, Ira and Doundoulakis, George, Scatter Propagation, Indianapolis, Howard W.
Sams & Co., Inc., 1956. #SPK-1.
Kuecken, John A., Antennas And Transmission Lines, Indianapolis, Howard W. Sams Co.,
Inc., 1969. #20716.
Lytel, Allen, ABC's of Antennas, Indianapolis, Howard W. Sams Co., Inc., 1966. #AAL-1.
Lytel, Allen, Microwave Test & Measurement Techniques, Indianapolis, Howard W. Sams
Co., Inc., 1964. #MIL-1.
Lyte!, Allen, UHF Television Antennas And Converters, New York, John F. Rider
Publisher, Inc., 1953. #153.
Mivec, F. Johnathan, Microwave Systems Fundamentals, Indianapolis, Howard W. Sams
& Co. Inc., 1963. #MSM-1.
National Cable Television Association, Standards of Good Engineering Practices for
Measurements on Cable Television Systems, Distrib;,:tion System, Washington, D.C.,
NCTA, 1977. #008-0477.
Pawlowski, Allen, MATV Systems Handbook-Design, Installation & Maintenance, Blue
Ridge Summit, Pa., Tab Books, 1973. #657.
Ray, Verne M., CATV Operator's Handbook, Blue Ridge Summit, Pa., Tab Books, 1967.
Rheinfelder, William A., CATV System Engineering, Blue Ridge Summit, Pa., Tab Books,
1970. #298.
Salvati, M.J., RF Interference Handbook, New York, Sony Corporation of America,
Technical Publications Department, 1977.
Salvati, M.}., TV Antennas And Signal Distribution Systems, Indianapolis, Howard W.
Sams & Co. Inc., 1979. #21584.
Sands, Leo G., Installing TV & FM Antennas, Blue Ridge Summit, Pa., Tab Books, 1974.
#636.
Shrock, Clifford B., No Loose Ends, The Tektronix Proof-of-Performance Program for CATV,
Tektronix Application Note, 1973. #26W-4889.
Simons, Ken, Technical Handbook For CATV Systems, Third edition, Hatboro, Pa.,
General Instrument Corporation, Jerrold Division, 1968. #436-001-01.
Television Publications, Inc., National Standards for CATV Systems, Graphic Symbols,
Oklahoma City, Oklahoma 73107.
Wortman, Leon A., Closed Circuit Television Handbook, Indianapolis, Howard W. Sams &
Co. Inc., 1964. #CLC-1.
149
Appendices
Appendix L: dBmV-to-Voltage Conversion Chart
Table L-l.
dBmV /Voltage Conversion Chart
dBmV
-40
/-tV
10
dBmV
0
/-tV
1,000
40
/-tV
100,000
110,000
-39
11
1
1,100
41
-38
13
2
1,300
42
130,000
-37
14
3
1,400
43
140,000
-36
16
4
1,600
44
160,000
-35
18
5
1,800
45
180,000
-34
20
6
2,000
46
200,000
-33
22
7
2,200
47
220,000
-32
25
8
2,500
48
250,000
-31
28
9
2,800
49
280,000
-30
32
10
3,200
50
320,000
-29
36
11
3,600
51
360,000
-28
40
12
4,000
52
400,000
-27
45
13
4,500
53
450,000
-26
50
14
5,000
54
500,000
-25
56
15
5,600
55
560,000
-24
63
16
6,300
56
630,000
-23
70
17
7,000
57
700,000
-22
80
18
8,000
58
800,000
-21
90
19
9,000
59
-20
100
20
10,000
60
1.0 volt
-19
110
21
11,000
61
1.1 volts
-18
130
22
13,000
62
1.2 volts
-17
140
23
14,000
63
1.4 volts
-16
160
24
16,000
64
1.6 volts
-15
180
25
18,000
65
1.8 volts
-14
200
26
20,000
66
2.0 volts
2.2 volts
900,000
-13
220
27
22,000
67
-12
250
28
25,000
68
2.5 volts
-11
280
29
28,000
69
2.8 volts
-10
320
30
32,000
70
3.2 volts
-9
360
31
36,000
71
3.6 volts
-8
400
32
40,000
72
4.0 volts
-7
450
33
45,000
73
4.5 volts
-6
500
34
50,000
74
5.0 volts
-5
560
35
56,000
75
5.6 volts
-4
630
36
63,000
76
6.3 volts
-3
700
37
70,000
77
7.0 volts
-2
800
38
80,000
78
8.0 volts
-1
900
39
90,000
79
9.0 volts
-0
1,000
40
100,000
80
10.0 volts
Definition of dBm V: 0 dBm V = 1,000 ,tV across 75 ohms.
150
dBmV
•
Index
Index
A
AlB switch, 33
abbreviations and acronyms, 126
absolute voltage levels, 21
ac power, 51, 112
ac-blocking terminator, 60
access, 66
acronyms, 126
active components, 18
adjacent channels, 67
advantages, broadband network, 8
advantages, narrow bandwidth, 74
advantages, unity gain, 22
ACe, 48, 79
agile, frequency, 66
airport, 68
alignment documentation, 103
alignment, forward path, 107
alignment method, round robin, 110
alignment methods, lO4ff
alignment of dual cable systems, 110
alignment of single cable systems, 106
alignment process, 100
alignment, return path, 108
alignment, system, 100ff
allocation, bandwidth, 68
allocation chart, frequency, 128
allocation, frequency, 64
aluminum shield, 19,43,52
amplifier, 20, 46ff, 64, 75, 83, 87, 110
amplifier alignment, MCe, 104
amplifier, backup, 88
amplifier, bridging, 49, 72
amplifier capacity, 36
amplifier cascade, 80ff
amplifier characteristics, 47-48
amplifier configurations, trunk, 101
152
amplifier cost, 47
amplifier, dual cable, 32
amplifier, forward, 59
amplifier gain, 22, 72, 80, 86, 100
amplifier gain control, 47
amplifier gain setting, 22
amplifier input level, 84
amplifier input, minimum, 85
amplifier, internal distribution, 50
amplifier, line extender, 50
amplifier, midsplit, 30
amplifier module additions, 50
amplifier operating current and voltage,
52
amplifier output level, 80ff
amplifier output level, cascaded, 48
amplifier output, maximum, 85
amplifier, overloading, 37
amplifier power, 51
amplifier selection, 79
amplifier, subsplit, 30
amplifier, trunk, 30
amplifier unit, 32
amplitude, 20
amplitude, outlet signal, 71
antenna site, 3
antenna siting and cabling, 63
antennas, 67
applications, broadband network, 8
applications, educational, 9
applications, local area network, 4
applications, video, 39
architecture, 12, 14
architecture and topology, 63
armored cable, 43
assignment table, channel, 129
assignments, channel, 67
attenuation, 23
attenuation, cable, 19
attenuators, 50, 86
automatic gain control (AGC), 48, 79
B
backbone, 3
backbone design, 36
backbone network, 13
backbone trunk, 36
backup, 38
backup amplifiers, 88
backup trunk, 87
bandpass, 110
bandpass filters (BPF), 59
bandstop filters (BSF), 59
bandwidth, 28-31, 36, 64-67, 75, 94
bandwidth allocation, 68
bandwidth compression, 39
bandwidth costs, 38-39
bandwidth, dual cable system, 31
bandwidth expansion factors, 64-65
bandwidth requirements, 65
bandwidth, return path, 16
bandwidth, single and dual systems, 36
bandwidth use, 39
barrel connector, 60
beat frequencies, 83
bidirectional, 3, 14, 46, 50, 148
bidirectional communication, 17
bidirectional signal distribution, 28
bidirectional single-cable systems, 16
bit error rates, 40
bloodhound, 102
BPF,59
branch cable, 18, 54
branch line, 13
branch signal levels, 51
breakdown, network, 38
breaks, cable, 86
bridger, 49
bridger amplifiers, 110
bridging amplifiers, 49, 51, 72
broadband,2
broadband communications, 2
broadband component quality, 87
broadband components, 42ff
153
broadband design equations, 132
broadband device receive level ( + 6
dBmV), 25, 71
broadband device transmit level ( + 56
dBmV), 25, 71
broadband equipment suppliers, 138
broadband local area network, 2, 5, 16
broadband network advantages, 8
broadband network applications, 8
broadband network characteristics, 3
broadband network design, 62ff
broadband network elements, 12
broadband symbols, 95, 127
Brown University, 9
BSF,59
building, 68
building codes, 44
building construction, 44
build in;; survey, 63
buildings, connecting, 36
buried cable, 43
bursty traffic, 66
bus network, 14
business uses, 6
C
cable attenuation, 19,45
cable bandwidth, 42
cable, branch, 54
cable breaks, 86
cable certification, 103
cable composition, 42
cable, defective, 103
Cable Design and Consulting Group
(CDCG), ii
cable diameter, 18
cable, discontinuities in, 104
cable distribution network, 3
cable distribution scheme, 91
cable, drop, 69
cable faults, 102
cable handling, 44
cable installation, 44, 54
cable layout, 49
cable loop resistance, 52
cable loss, 18, 22-23
Cable Marketing Magazine, 31
Index
cable power, 19,51
cable reflectometer, 102
cable resistance, 52
cable slope, 45
cable specification, 19
cable spectrum, 15
cable sweeping test, 104
cable systems, multiple, 36
cable television customers, 3
Cable Television, Manhattan, 33
cable television (see CATV)
cable tilt, 23, 45-46, 57
cable, trunk, 68
cable, underground, 48
calculating signal levels, 20
calculations, C/N, 131
calculations, noise, 130
calculations, RF, 130
camera, television, 65
capacity, amplifier, 36
capacity, channel, 66-67
carrier, data, 38
carrier derating formula, 72
carrier level, derated, 37
carrier levels, narrow bandwidth, 72ff
carrier signal, 25, 66-67
carrier signal, pilot, 79
carrier-to-hum (C/H) ratio, 82
carrier-to-noise (C/N) ratio, 75, 82, 84, 131
cascade,40,48,50,72,80, 84,86
cascade, effect on noise, 75-76
cascaded C/H, 82
cascaded intermodulation, 83
cascading amplifiers, 48
CATV, 3, 9-10, 14, 16,28,94
CATV amplifiers, 48
CATV converters, 28
CATV dual trunk systems, 33
CATV industrial trunks, 7
CATV multichannel headends, 94
CATV network, 34
CATV operators, 68
CATV standards, 49
CATV taps, 59
CATV test equipment, 103
CATV trunk, 34
CCTV, 65, 67
CDCG, ii
154
certification, coaxial cable, 103
C/H,82
changing tap value, 57
channel,4, 16, 28, 37,65-66, 68, 72-73
channel access, 3
channel access protocols, 14, 66
channel, adjacent, 67
channel assignment, 67
channel assignment table, 129
channel capacity, 66-67
channel processors, 67
characteristic impedance, 18
characteristics, amplifier, 47-48
characteristics, drop cable, 43
characteristics, trunk cable, 43
checking an outlet, 113
closed circuit television (CCTV), 65, 67
C/N, 75, 82, 84, 131
C/N calculations, 131
coaxial cable, 3-4, 42ff
coaxial cable certification, 103
coaxial switches, 87
co-channel interference, 67
codes, fire, 44
color subcarrier, 72
comb generator, 94
combined single and dual cable systems,
34-35
combiner, 77
combiner, power, 93
combiners, eight-way, 92
commercial distribution, 33
commercial trunk, 33
common port, diplexer, 59
communication networks, data, 29
communications utility, 3
Community Antenna Television (see
CATV)
comparing single and dual cable systems,
36
compensate, 23, 46
compensation, frequency, 47
compensator circuit, thermal, 47
component failures, passive, 112
component performance test, 104
component quality, 87
components, 12, 14, 18,28, 42ff
components, passive, 64
composite triple beat (CTB), 83, 85, 94
conduit, 43-44
connecting buildings, 36
connecting networks, 33
connections, 9, 13
connections, faulty, 86
connections, trunk, 54
connectivity, 18
connector, barrel, 60
connector details, 144
connector types, 55
connector /equipment interface, 146
connectors and hardware, 54
construction, 63
continuous access, 66
contractor, 63
control systems, technical, 88
conventional wiring, 6
conversion, frequency, 67
converters, CATV, 28
converting one-way to two-way system,
31
Copperweld Company, 53,142
cost, amplifier, 47
cost estimate, 63
cost of cable, 44
cost of converting, 31
costs, bandwidth, 38
couplers, 20
couplers, directional, 54, 68
Cox Cable Communications, 40
CTB, 83, 85, 94
current, 20
D
data carrier, 38
data carrier level (DCL), 72
data channel, 73
data communication networks, 29
data communications, 65
data communications devices, 72
data signals, narrow bandwidth, 37
data sub channels, 72
dB,20
dBm, 130
dBm V, 20, 130
dBm V-to-voltage conversion chart, 21, 150
DCL,72
155
decibel (dB), 20
decibel referred to one millivolt (dBmV),
20, 130
decibel referred to one milliwatt (dBm),
130
decibels per 100 feet, 19
defective cable, 103
defects, installation, 111
definition, carrier derating, 72
definition of dBm V, 20
definition of decibel, 20
definition of noise floor, 75
definition of terms, 118££
definition, unity gain, 22
delay, envelope, 59
derated carrier level, 37
derating formula, carrier, 72
Design and Consulting Group, Cable, ii
design and performance calculations, 79££
deSign, backbone, 36
design cascade, 80
design characteristics, system, 81
design, distribution system, 86
design equations, 132
design errors, 111
design, headend, 89ff
design of broadband networks, 19
design of the physical layout, 68-69
design, trunk, 63
detailed headend configuration, 92
device receive level, 25, 71
device transmit level, 25, 71
digital control signals, 67
diplex filter, 59
diplex filter loss, 80
diplexers,59
directional coupler, 54, 60, 68
directivity, 56
disconnecting feeders, 51
discontinuities in cable, 104
distortion, 37, 50, 86
distortion, amplifier, 47
distortion, dual cabte network, 32
distortion, inter modulation, 75, 83
distortion level, 84
distribution cable, 43, 49
distribution, commercial, 33
distribution, institutional, 33
Index
distribution legs, 58
distribution network, 3, 14, 16-17,70
distribution network design, 19
distribution, RF signal, 20
distribution scheme, 91
distribution, signal, 68
distribution system, 54
distribution system design, 86
distribution, television, 67
documentation, alignment, 103
drawing standards, 95
drop cable, 13, 18, 43, 69
drop cable characteristics, 43
drop cable problems, 112
dual cable amplifier, 32
dual cable systems, 28, 31ff
dual cable systems, alignment of, 110
dual trunk systems, CATV, 33
dual trunk/dual feeder, 35
dual trunk/dual feeder with crossover, 35
dual trunk/single feeder, 34
E
educational uses, 8
eight-way combiners, 92
electromagnetic interference (EMI), 4, 42,
103
emergency power, 87
EM!, 4, 42, 103
enclosures, 44
engineering contractors, 63
EN!, 84-85
envelope delay, 59
equalization, 47
equalizer, 23, 46
equalizer loss, 80
equalizers, 50
equipment suppliers, broadband, 138
equipment, test, 102, 133
equivalent noise input (ENI), 84-85
errors, design, 111
example cable distribution scheme, 91
example, DCL, 73-74
examples, network, 9
expansion, 49
expansion, bandwidth, 64-65
expansion, future, 63
156
F
FAA, 68, 94
failure, 9, 86-88
failure, network, 38
failures, system, 54
faults, cable, 102
faulty connections, 86
FCC, 68, 94
F-connector, 60
F-connector types and parts, 147
FDM, 3, 67
Federal Aviation Administration (FAA),
68,94
Federal Communications Commission
(FCC), 68, 94
feeder cable, 43
feeder cable selection criteria, 43
feeder disconnect circuit, 51
field strength meter (FSM), 102
figure, noise, 47, 75, 84
filter, RF, 93
filter, system, 93
filters, 32, 59
fire codes, 44
fixed frequency, 66
flat amplifier input method, 105
flat amplifier output method, 104
flat loss, 22-23
flat midspan method, 106
flat response, 104
flexibility, 13
flooding compound, 43
flooding gel, 43
floor, noise, 21, 75, 82, 84
format, highsplit, 16
format, midsplit, 16
format, subsplit, 16
forward amplifier, 59
forward band, 16, 28-31, 68
forward path, 49, 67, 92
forward path alignment, 107
forward path amplifier gain, 80
forward path bandwidth, 36
forward path loss, 71
forward path, noise, 77
frequency agile, 40, 66
frequency agile modems, 40
frequency allocation chart, 64, 128
frequency allocations, 63
frequency band, 33
frequency compensation, 47
frequency considerations, system, 64
frequency conversion, 67
frequency division multiplexing (FDM),
3,67
frequency multiplexing, 4
frequency shift keying, 67
frequency spectrum, 16, 28, 68, 102
frequency translation, 15
frequency variation of cable attenuation,
45
FSM,102
full duplex, 16, 66
full duplex operation, 16
fuses, 112
future developments, 39
future expansion, 49, 63
G
gain, 100
gain, 22, 47, 80, 86, 100
gain control, 47, 79
gain control, return path, 79
gain, reserve, 80
gateways, 33
general amplifier characteristics, 47
generator, comb, 94
geographic independence, 6
geographical coverage, 63
geographically independent, 3
graph, system level, 84
ground loops, 143
grounding, 53
grounding between sites, 142
Grounding Principles, 53
grounding, system, 141
guard band, 15-16,68
Guidelines for Handling Trunk and
Feeder Cables, 44
H
handling cable, 44
hardware, 54
harmonically related carrier (HRC), 94
157
headend, 12-13,28,33,51-52,63
headend channel assignment table, 129
head end configuration, 90
headend design, 89££
headend translator, 15-17
high port, diplexer, 59
high-resolution video, 67
highsplit, 16, 31
highway, 16
home terminal, 7
HRC, 94
I
IEEE-488 interface bus, 88
IEEE-802, 29
IMD, 75, 83
impedance, 18,42
improper connector tightening, 111
inbound, 15, 31
inbound and outbound cables, 110
inbound cable, 110
incorrect orientation of passive taps, 112
independence,vendo~8
independent, geographically, 3
industrial outlet levels, 98
industrial trunking, 33-34
industrial trunks, 10
industrial uses, 7
information utility, 63
ingress, 42, 60, 103, 112
input level, amplifier, 84
input side, amplifier, 49
input signal level, 21
insertion loss, 53, 57
insertion loss, amplifier, 47
inspection, physical, 87
installation, 56, 103
installation and maintenance, tools for,
136
installation, cable, 54
installation considerations, cable, 44
installation damage, 112
installation defects, 111
installation, overhead cable, 44
institutional distribution, 33
intelligent amplifiers, 51
Index
interconnecting LANs, 33
interface bus, IEEE-488, 88
interface device, 66, 70, 73
interference, 4, 37
interference, co-channel, 67
intermittent loss of signal, 112
inter modulation distortion (lMO), 75, 83
internal distribution amplifiers, 50
interval related carrier (IRC), 95
inverted tree, 13
IRC, 95
isolation, 32, 51, 56
isolation pad, 108
J
junction, return path signal levels at a,
109
L
LAN, 2, 7,14-16,29,34,38,48,68
LANs, interconnecting, 33
large CATV multichannel headends, 94
layout, 13, 38, 63
layout, cable, 49
layout, physical, 68-69
line amplifiers, 50
line extender amplifiers, 50, 79
little or no RF signal, 111
local area network applications, 4
local area network (LAN), 2, 7, 14-16,29,
34, 38, 48, 68
local origination, 67
logarithmic units, 20
logical topology, 12, 14
loop resistance, 52
loops, ground, 143
loss, 53, 56
loss, cable, 18,22
loss, diplex filter, 80
loss, equalizer, 80
loss, flat, 22-23
loss, insertion, 57
loss of signal, 112
loss, signal, 19
low port, diplexer, 59
158
M
maintenance, periodic, 87
manager, network, 62
Manhattan Cable Television, 33
manual gain control (MGC), 47, 79
margin, output level, 86
MATV systems, 68
maximum amplifier output, 85
mean-time-between failure (MTBF), 87
messengers, 43
meter, field strength, 102
method, flat amplifier input, 105
method, flat amplifier output, 104
method, flat midspan, 106
method, modified flat midspan, 106
MGC, 47, 79
MGC amplifier alignment, 104
microphone, 65
midsplit, 16, 29, 31
midsplit amplifier, 30
millivolt, 20, 130
minimum amplifier input, 85
modem, 29
modems, frequency agile, 40
modems, point-to-point RF, 66
modified flat midspan method, 106
modulator, 65
modulators, television, 71
module failure, 9
monitoring system performance, 113
monitoring systems, 63
monitoring systems, status, 88
mounting hardware, 54
MSOs (multiple station operators), 40
MTBF,87
multichannel headends, CATV, 94
multilane highway, 16
multimeter,102
multi-path distortion, 13
multiple cable systems, 36
multiple cable systems, problems, 36
multiple services, 6
multiple station operators (MSOs), 40
multiplexed modems, 66
multiplexing, 3
multi-taps, 51, 56
Murphy's law, 39
N
narrow bandwidth advantages, 74
narrow bandwidth carrier levels, 72ff
narrow bandwidth data signals, 37
NC (number of carriers), 72
network,15
network architecture and topology, 63
network breakdown, 38
network deSign, 19
network design, broadband 62ff
network elements, 12
network examples, 9
network failure, 38
network manager, 62
network RF test point, 93
network size, 9
network, troubleshooting, 51
networks, connecting, 33
node, 13
noise at a splitter/combiner, 77-78
noise calculations, 130
noise figure, 47, 50, 75, 84
noise figure, typical, 82
noise floor, 21, 75, 82, 84
noise floor, defined 75
noise level, 40, 75ff
noise spike, 79
noise, system, 81
noise, thermal, 75
non-amplified system, 107
non translated devices, 93
number of carriers (NC), 72
o
office of the future, 7
office outlet levels, 98
one-way service, 40
one-way system, 16,31
one-way trunk, 33
operating current and voltage, amplifier,
52
operating experience, 9
operating window, 84
outbound, 15, 31
outbound cable, 110
outlet, 9, 13, 38-39, 43, 51, 60, 69-70
outlet, checking an, 113
159
outlet drop cable problems, 112
outlet, dual cable system, 32, 39
outlet level, 24
outlet levels, industrial, 98
outlet levels, office, 98
outlet signal amplitude, 71
outlets, single and dual cable systems, 39
output level, 47, 84
output level, amplifier, 84
output level margin, 86
output ports, 112
output side, amplifier, 49
outside noise sources, 82
overcompensation, 48, 79
overdrive, 72
overhead cable installation, 44
overloading amplifiers, 37
overtightening, 111
p
packet communication unit, 29, 66
pad selection, 108
passive component failures, 112
passive components, 18, 53ff, 64
passive loss, 23, 53
passive taps, 20
patch panel, 93
path layout, 38
path loss, forward, 71
path, return, 40
performance test, 104
periodic maintenance, 87
physical inspection, 87
physical layout, 68-69
physical topology, 12
picture quality for C/N values, 77
pilot carrier signal, 79
point-to-point RF modems, 66
pole mounting, 43
ports, 57
power, 20
power, cable, 19
power capacity of amplifiers, 37
power combiner, 51, 93
power distribution, 51
powe~emergency,87
Index
power supply, 51ft, 93
power supply, rule-of-thumb, 52
problem: design errors, 111
problem: improper connector tightening,
111
problem: incorrect orientation of passive
taps, 112
problem: installation defects, 111
problem: intermittent loss of signal, 112
problem: little or no RF signal, 111
problem: loss of signal, 112
problem: outlet drop cable, 112
problem: overtightening, 111
problem: passive component failures, 112
problem: radiation and signal ingress, 112
problem: short circuit, 112
problems and solutions, 111
problems with multiple cable systems, 36
procedure, frequency allocation, 64
process, alignment, 100
processors, channel, 67
programmable taps, 58
protector, surge, 93
protocols, 14, 66
PVC-coated cables, 44
Q
quality, component, 87
quality for C/N values, picture, 77
R
radiation, 102
radiation and signal ingress, 112
radio frequency interference, 8
radio frequency interference (RFI), 4, 42,
103
receive, 28
receive level (+6 dBmV), 25, 71
receive reference level, 70
receiver, television, 65
receiving unit, 16-17
redundancy, 63, 86££
redundancy, single and dual cable
systems, 38-39
redundant trunks and components, 87
reference level, 70
160
reference, television signal, 104
reflectometer, 102
regulations, 68
reliability, 10, 63, 86££
repair, 63, 87
replacement and repair, 87
reserve gain, 80
reserved channels, 16
resistance, cable, 52
resistance, ground, 53
return amplifier, 59
return band, 16, 28ft, 49, 67-68, 92
return loss test, 103
return path alignment, 108
return path amplifiers, 50
return path bandwidth, 16,36
return path gain control, 79
return path ingress, 113
return path noise, 77
return path signal levels, 49
return path signal levels at a junction, 109
reverse band, 29
RF calculations, 130
RF carrier signals, 3
RF connections, 9
RF connector, 39
RF connector details, 144
RF filter, 93
RF modems, point-to-point, 66
RF radiation, 102
RF signal, 12-13,65
RF signal distribution, 20
RF signal level changes, 46
RF signal, little or no, 111
RF spectrum analyzer, 102
RF sweep generator, 102
RFI, 4, 42, 103
RG-11,43
RG-59,43
RG-6,43
ring network, 15
robust, 3
round robin alignment method, 110
rule-of-thumb, AGC amplifiers, 48
rule-of-thumb, cable attenuation vs.
temperature, 46
rule-of-thumb, cascading amplifiers, 48
rule-of-thumb, conduit installation cost,
44
rule-of-thumb, pad selection, 108
rule-of-thumb, power supply, 52
S
safety, 63
sample system design characteristics, 81
second-order beat frequencies, 83
self-terminating outlet, 39, 60
series amplifiers, 48
services, 16, 29, 63-65, 69
services, special, 67
shared access, 66
shield, 18,32,42
shield, aluminum, 43
short circuit, 112
shrink tubing, 54
signal, 65
signal amplitude, 20
signal distribution, 9, 68
signal distribution, bidirectional, 28
signal frequency, 15
signal ingress, 42
signal level, 19-20, 70ff
signal level, calculating, 20
signal level changes, 46
signal level gain, 47
signal level, return path, 49
signal level, standard, 24
signal level, typical, 56
signal loss, 19
signal, loss of, 112
signal strength, 69
signals, carrier, 66-67
signal-to-noise ratio (SIN), 75, 84, 86
single cable system, 16, 28f£, 36, 64
single cable systems, alignment of, 106
size, network, 9
slope, cable, 45
slope control, 106
small businesses, 7
small discontinuities in cable, 104
SIN, 75, 84, 86
sniffer, 102
solutions, 111
sound carrier, television, 72
161
span, 111
special services, 67
specification, cable, 19
specifications, system, 96
spectrum, 15-16,28,37,68,72-74,77
spectrum analyzer, 102
spectrum analyzer connection, 108
splitter, 56, 77
standard connectors, 39
standard headend, 89
standard signal level, 24
standards, CATV, 49
standards, drawing, 95
standby power units, 52
status monitoring systems, 51, 63, 88
structural return loss test, 103
studies, broadband communications, 40
studio, television, 67
subcarrier, color, 72
subchannels, 74
subsplit, 16, 28
subsplit amplifier, 30
surge protection, 52
surge protector, 93
sweep generator, 102
sweep testing, 103-104
switch, AlB, 33
switches, coaxial, 87
symbols, broadband, 95, 127
symptoms, 111
system alignment, 100ff
system carrier-to-noise ratio, 76
system design characteristics, 81
system design, transparent, 24
system diagrams, 89ff
system failure, 51, 54
system filter, 93
system frequency considerations, 64
system grounding, 141
system level graph, 84
system losses, 22
system noise, 81
system performance, monitoring, 113
system response, 46
system specifications, 96
system structure, 64
Sytek, ii
Index
T
T1 channel, 29
tap loss, 56
tap point, 54
tap value, changing, 57
taps, passive, 20
TASO,67
TOM,3
technical control systems, 88
Teflon cable, 44
TeleNet,34
telephone system, FOM, 67
Television Allocations Study
Organization (TASO), 67
television and data carriers, 74
television camera, 65
television channel, 28, 72
television distribution, 67
television modulators, 71
television receiver, 65
television set, 71
television signal reference, 104
television signals, 3, 70
television studio, 67
television visual carrier signal, 25
temperature variation of cable
attenuation, 46
terminators, 60
test, cable sweeping, 104
test, component performance, 104
test equipment, 102, 133
test equipment, CATV, 103
test patch panel, 93
test point, network RF, 93
test points, amplifiers, 49
test, return loss, 103
testing, sweep, 103
thermal compensator circuit, 47
thermal noise, 75
third-order beat frequencies, 83
tilt, 104
tilt, cable, 45-46, 57
time-division multiplexing (TOM), 3
Times Fiber Communications, 44
token-passing ring network, 15
tools for installation and maintenance,
136
topology, 12, 63
total system failure, 86
162
traffic, bursty, 66
translation, 68
translator, 15-17
transmit, 28
transmit level (+ 56 dBm V), 25, 37, 70-71
transmitter site, 68
transmitting unit, 16-17
transparency, advantages, 24
transparent, 24
transparent distribution system, 70
transparent system, 72
transparent system design, 24
transponders, 88
tree architecture, 64
troubleshooting a network, 51
trunk amplifier, 30, 79, 110
trunk amplifier characteristics, 48
trunk amplifier configurations, 101
trunk, backbone, 36
trunk, backup, 87
trunk cable, 18, 20, 43, 68
trunk cable characteristics, 43
trunk, CATV, 34
trunk connections, 54
trunk design, 63
trunk design, unity gain, 22
trunk levels, 71
trunk line, 13
trunk, one-way, 33
trunk/bridger amplifiers, 79
trunking, industrial, 33-34
trunks, above ground, 142
trunks and components, redundant, 87
trunks, industrial, 10
tuned RF voltmeter, 102
tuning, 66
turnkey service, 63
two-way communications, 28
two-way operation, 16
two-way service, 40
two-way system, 31, 66
two-way transmission, 34
Tymnet,34
types and parts, F-connector, 147
types, connector, 55
types of amplifiers, 48ff
types of cables, 43
typical signal levels, 56
U
underground cable, 48
underground installations, 54
unidirectional, 4, 14
unity gain criterion, 22
upgrading, 16
upgrading a network, 39
usable gain, 80
usable gain, amplifier, 47
user outlets, 13
uses, business, 6
uses, educational, 8
uses, industrial, 7
utility, 3
utility, information, 63
variation of cable attenuation,
temperature, 46
VCL,72
vendor independence, 8
ventilation, 44
VHF television channels, 28
video applications, 39
video bandpass, 72
video carrier level (VCL), 72
video reference level, 70
visual carrier signal, 25
voltage, 20
voltage dependent, amplifier, 52
voltage levels, 21
W
V
valves, 20
variable equalizers, 50
variation of cable attenuation, frequency,
45
163
Wang Laboratories, 31
water delivery system, 20
work-at-home, 7
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

advertising